diff --git a/.dvc/.gitignore b/.dvc/.gitignore deleted file mode 100644 index 528f30c71c687de473bbb506c071e902beba6cd9..0000000000000000000000000000000000000000 --- a/.dvc/.gitignore +++ /dev/null @@ -1,3 +0,0 @@ -/config.local -/tmp -/cache diff --git a/.dvc/config b/.dvc/config deleted file mode 100644 index 2f569628e11a710c99dc12b1eaede351c4748113..0000000000000000000000000000000000000000 --- a/.dvc/config +++ /dev/null @@ -1,6 +0,0 @@ -[cache] - dir = /data/dvc-cache/FintoAI-data-YSO - shared = group - type = symlink -[core] - autostage = true diff --git a/.dvcignore b/.dvcignore deleted file mode 100644 index 2801e4d488fcc0bddbf7f6ca95690ba6072bf441..0000000000000000000000000000000000000000 --- a/.dvcignore +++ /dev/null @@ -1,4 +0,0 @@ -# Add patterns of files dvc should ignore, which could improve -# the performance. Learn more at -# https://dvc.org/doc/user-guide/dvcignore -*.pdf diff --git a/.gitattributes b/.gitattributes deleted file mode 100644 index 6ddd5e901133d06650bc1b64e40507cbe5532954..0000000000000000000000000000000000000000 --- a/.gitattributes +++ /dev/null @@ -1,54 +0,0 @@ -*.7z filter=lfs diff=lfs merge=lfs -text -*.arrow filter=lfs diff=lfs merge=lfs -text -*.bin filter=lfs diff=lfs merge=lfs -text -*.bz2 filter=lfs diff=lfs merge=lfs -text -*.ckpt filter=lfs diff=lfs merge=lfs -text -*.ftz filter=lfs diff=lfs merge=lfs -text -*.gz filter=lfs diff=lfs merge=lfs -text -*.h5 filter=lfs diff=lfs merge=lfs -text -*.joblib filter=lfs diff=lfs merge=lfs -text -*.lfs.* filter=lfs diff=lfs merge=lfs -text -*.mlmodel filter=lfs diff=lfs merge=lfs -text -*.model filter=lfs diff=lfs merge=lfs -text -*.msgpack filter=lfs diff=lfs merge=lfs -text -*.npy filter=lfs diff=lfs merge=lfs -text -*.npz filter=lfs diff=lfs merge=lfs -text -*.onnx filter=lfs diff=lfs merge=lfs -text -*.ot filter=lfs diff=lfs merge=lfs -text -*.parquet filter=lfs diff=lfs merge=lfs -text -*.pb filter=lfs diff=lfs merge=lfs -text -*.pickle filter=lfs diff=lfs merge=lfs -text -*.pkl filter=lfs diff=lfs merge=lfs -text -*.pt filter=lfs diff=lfs merge=lfs -text -*.pth filter=lfs diff=lfs merge=lfs -text -*.rar filter=lfs diff=lfs merge=lfs -text -*.safetensors filter=lfs diff=lfs merge=lfs -text -saved_model/**/* filter=lfs diff=lfs merge=lfs -text -*.tar.* filter=lfs diff=lfs merge=lfs -text -*.tar filter=lfs diff=lfs merge=lfs -text -*.tflite filter=lfs diff=lfs merge=lfs -text -*.tgz filter=lfs diff=lfs merge=lfs -text -*.wasm filter=lfs diff=lfs merge=lfs -text -*.xz filter=lfs diff=lfs merge=lfs -text -*.zip filter=lfs diff=lfs merge=lfs -text -*.zst filter=lfs diff=lfs merge=lfs -text -*tfevents* filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-en/omikuji-model/tree0.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-en/omikuji-model/tree1.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-en/omikuji-model/tree2.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-en/vectorizer filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-fi/omikuji-model/tree0.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-fi/omikuji-model/tree1.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-fi/omikuji-model/tree2.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-fi/vectorizer filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-sv/omikuji-model/tree0.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-sv/omikuji-model/tree1.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-sv/omikuji-model/tree2.cbor filter=lfs diff=lfs merge=lfs -text -data/projects/yso-bonsai-sv/vectorizer filter=lfs diff=lfs merge=lfs -text -data/projects/yso-en/nn-model.keras filter=lfs diff=lfs merge=lfs -text -data/projects/yso-fasttext-en/fasttext-model filter=lfs diff=lfs merge=lfs -text -data/projects/yso-fasttext-fi/fasttext-model filter=lfs diff=lfs merge=lfs -text -data/projects/yso-fasttext-sv/fasttext-model filter=lfs diff=lfs merge=lfs -text -data/projects/yso-fi/nn-model.keras filter=lfs diff=lfs merge=lfs -text -data/projects/yso-sv/nn-model.keras filter=lfs diff=lfs merge=lfs -text -data/vocabs/yso/subjects.ttl filter=lfs diff=lfs merge=lfs -text diff --git a/.gitignore b/.gitignore deleted file mode 100644 index ea041bea6977d9b527a59629e1318b624b0c6c77..0000000000000000000000000000000000000000 --- a/.gitignore +++ /dev/null @@ -1,2 +0,0 @@ -venv -venv-installed diff --git a/LICENSE b/LICENSE deleted file mode 100644 index 0e259d42c996742e9e3cba14c677129b2c1b6311..0000000000000000000000000000000000000000 --- a/LICENSE +++ /dev/null @@ -1,121 +0,0 @@ -Creative Commons Legal Code - -CC0 1.0 Universal - - CREATIVE COMMONS CORPORATION IS NOT A LAW FIRM AND DOES NOT PROVIDE - LEGAL SERVICES. DISTRIBUTION OF THIS DOCUMENT DOES NOT CREATE AN - ATTORNEY-CLIENT RELATIONSHIP. CREATIVE COMMONS PROVIDES THIS - INFORMATION ON AN "AS-IS" BASIS. CREATIVE COMMONS MAKES NO WARRANTIES - REGARDING THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS - PROVIDED HEREUNDER, AND DISCLAIMS LIABILITY FOR DAMAGES RESULTING FROM - THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS PROVIDED - HEREUNDER. - -Statement of Purpose - -The laws of most jurisdictions throughout the world automatically confer -exclusive Copyright and Related Rights (defined below) upon the creator -and subsequent owner(s) (each and all, an "owner") of an original work of -authorship and/or a database (each, a "Work"). - -Certain owners wish to permanently relinquish those rights to a Work for -the purpose of contributing to a commons of creative, cultural and -scientific works ("Commons") that the public can reliably and without fear -of later claims of infringement build upon, modify, incorporate in other -works, reuse and redistribute as freely as possible in any form whatsoever -and for any purposes, including without limitation commercial purposes. -These owners may contribute to the Commons to promote the ideal of a free -culture and the further production of creative, cultural and scientific -works, or to gain reputation or greater distribution for their Work in -part through the use and efforts of others. - -For these and/or other purposes and motivations, and without any -expectation of additional consideration or compensation, the person -associating CC0 with a Work (the "Affirmer"), to the extent that he or she -is an owner of Copyright and Related Rights in the Work, voluntarily -elects to apply CC0 to the Work and publicly distribute the Work under its -terms, with knowledge of his or her Copyright and Related Rights in the -Work and the meaning and intended legal effect of CC0 on those rights. - -1. Copyright and Related Rights. A Work made available under CC0 may be -protected by copyright and related or neighboring rights ("Copyright and -Related Rights"). Copyright and Related Rights include, but are not -limited to, the following: - - i. the right to reproduce, adapt, distribute, perform, display, - communicate, and translate a Work; - ii. moral rights retained by the original author(s) and/or performer(s); -iii. publicity and privacy rights pertaining to a person's image or - likeness depicted in a Work; - iv. rights protecting against unfair competition in regards to a Work, - subject to the limitations in paragraph 4(a), below; - v. rights protecting the extraction, dissemination, use and reuse of data - in a Work; - vi. database rights (such as those arising under Directive 96/9/EC of the - European Parliament and of the Council of 11 March 1996 on the legal - protection of databases, and under any national implementation - thereof, including any amended or successor version of such - directive); and -vii. other similar, equivalent or corresponding rights throughout the - world based on applicable law or treaty, and any national - implementations thereof. - -2. Waiver. To the greatest extent permitted by, but not in contravention -of, applicable law, Affirmer hereby overtly, fully, permanently, -irrevocably and unconditionally waives, abandons, and surrenders all of -Affirmer's Copyright and Related Rights and associated claims and causes -of action, whether now known or unknown (including existing as well as -future claims and causes of action), in the Work (i) in all territories -worldwide, (ii) for the maximum duration provided by applicable law or -treaty (including future time extensions), (iii) in any current or future -medium and for any number of copies, and (iv) for any purpose whatsoever, -including without limitation commercial, advertising or promotional -purposes (the "Waiver"). Affirmer makes the Waiver for the benefit of each -member of the public at large and to the detriment of Affirmer's heirs and -successors, fully intending that such Waiver shall not be subject to -revocation, rescission, cancellation, termination, or any other legal or -equitable action to disrupt the quiet enjoyment of the Work by the public -as contemplated by Affirmer's express Statement of Purpose. - -3. Public License Fallback. Should any part of the Waiver for any reason -be judged legally invalid or ineffective under applicable law, then the -Waiver shall be preserved to the maximum extent permitted taking into -account Affirmer's express Statement of Purpose. In addition, to the -extent the Waiver is so judged Affirmer hereby grants to each affected -person a royalty-free, non transferable, non sublicensable, non exclusive, -irrevocable and unconditional license to exercise Affirmer's Copyright and -Related Rights in the Work (i) in all territories worldwide, (ii) for the -maximum duration provided by applicable law or treaty (including future -time extensions), (iii) in any current or future medium and for any number -of copies, and (iv) for any purpose whatsoever, including without -limitation commercial, advertising or promotional purposes (the -"License"). The License shall be deemed effective as of the date CC0 was -applied by Affirmer to the Work. Should any part of the License for any -reason be judged legally invalid or ineffective under applicable law, such -partial invalidity or ineffectiveness shall not invalidate the remainder -of the License, and in such case Affirmer hereby affirms that he or she -will not (i) exercise any of his or her remaining Copyright and Related -Rights in the Work or (ii) assert any associated claims and causes of -action with respect to the Work, in either case contrary to Affirmer's -express Statement of Purpose. - -4. Limitations and Disclaimers. - - a. No trademark or patent rights held by Affirmer are waived, abandoned, - surrendered, licensed or otherwise affected by this document. - b. Affirmer offers the Work as-is and makes no representations or - warranties of any kind concerning the Work, express, implied, - statutory or otherwise, including without limitation warranties of - title, merchantability, fitness for a particular purpose, non - infringement, or the absence of latent or other defects, accuracy, or - the present or absence of errors, whether or not discoverable, all to - the greatest extent permissible under applicable law. - c. Affirmer disclaims responsibility for clearing rights of other persons - that may apply to the Work or any use thereof, including without - limitation any person's Copyright and Related Rights in the Work. - Further, Affirmer disclaims responsibility for obtaining any necessary - consents, permissions or other rights required for any use of the - Work. - d. Affirmer understands and acknowledges that Creative Commons is not a - party to this document and has no duty or obligation with respect to - this CC0 or use of the Work. diff --git a/README.md b/README.md deleted file mode 100644 index 3b884ea5da12097a271c183844f65597df0aa500..0000000000000000000000000000000000000000 --- a/README.md +++ /dev/null @@ -1,49 +0,0 @@ ---- -license: cc0-1.0 -language: -- fi -- sv -- en -pipeline_tag: text-classification -thumbnail: https://raw.githubusercontent.com/NatLibFi/FintoAI/main/ai.finto.fi/static/img/finto-ai-social.png -tags: -- glam -- lam -- subject indexing -- annif ---- -# FintoAI-data-YSO -This repository is for the Annif projects with the -[YSO vocabulary](https://finto.fi/yso) -used at the [Finto AI service](https://ai.finto.fi/). -The current models were published there 2023-09-04. -The models have been trained on Python 3.8.10 with [Annif](https://annif.org) version 1.0.0. -See [projects.toml](projects.toml) for the configurations of the models. - -This repository is mirrored from GitHub to the 🤗 Hugging Face Hub; -the GitHub repository does not contain the model files, but only the configurations for the projects and the DVC pipeline, see below. - -The training corpora that are public can be found from the [Annif-corpora repository](https://github.com/NatLibFi/Annif-corpora/). - -The [notebook](/repository-metrics-analysis/analyse-theseus-tietolinja.ipynb) contains analysis of Annif suggestions in [Theseus repository](https://www.theseus.fi/). - -## Models -The downloadable directories for projects and vocabularies are stored in the -[`/data`](https://huggingface.co/juhoinkinen/FintoAI-data-YSO/tree/main/data) -directory of this repository in the 🤗 Hugging Face Hub. - -## DVC pipeline -The projects are trained and evaluated using a [DVC (Data Version Control) pipeline](https://dvc.org/doc/start/data-management/data-pipelines) defined in [dvc.yaml](./dvc.yaml). - -The pipeline takes care of - -1. installing Annif in a venv, -2. loading the vocabulary, -3. training the projects, -4. evaluating the projects. - -When the necessary vocabulary and training corpora are in place the pipeline can be run using the command - - dvc repro - -For more information about using DVC with Annif projects see the [DVC exercise of Annif tutorial](https://github.com/NatLibFi/Annif-tutorial/blob/master/exercises/OPT_dvc.md). diff --git a/corpora/.gitignore b/corpora/.gitignore deleted file mode 100644 index f72aa5c1124b6fe1d9737452dde1dab0a2024003..0000000000000000000000000000000000000000 --- a/corpora/.gitignore +++ /dev/null @@ -1 +0,0 @@ -/yso-skos.ttl diff --git a/corpora/fulltext-test/en/.gitignore b/corpora/fulltext-test/en/.gitignore deleted file mode 100644 index c62954f71699023c78ea8bdadd3e40f0ad0bc1b6..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/en/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -/abo-theses -/jyu-theses -/kirjaesittelyt2021 -/vapaakappaleet-orig diff --git a/corpora/fulltext-test/en/abo-theses.dvc b/corpora/fulltext-test/en/abo-theses.dvc deleted file mode 100644 index 77f4895ddb762df20ac9969c8a23541c93ce37e4..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/en/abo-theses.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 3eb7c18baf9f361da8535b971115dc94 -deps: -- md5: 7743e923695da632586b6426318d6ac6.dir - size: 181887549 - nfiles: 306 - path: /data/Annif-corpora/fulltext/abo-theses/eng/test/ - hash: md5 -outs: -- md5: 8543bf23d8ae5c415973beeac864929c.dir - size: 10823806 - nfiles: 227 - path: abo-theses diff --git a/corpora/fulltext-test/en/jyu-theses.dvc b/corpora/fulltext-test/en/jyu-theses.dvc deleted file mode 100644 index 4211312e4640396a40310cfdfb804a0ea5fde259..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/en/jyu-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 8d33b71700855688cf72611c81332782 -deps: -- md5: 35f96b1279030a69ca1c67c792768faf.dir - size: 70776290 - nfiles: 603 - path: /data/Annif-corpora/fulltext/jyu-theses/eng-test/ - hash: md5 -outs: -- md5: 35f96b1279030a69ca1c67c792768faf.dir - size: 70776290 - nfiles: 603 - path: jyu-theses - hash: md5 diff --git a/corpora/fulltext-test/en/kirjaesittelyt2021.dvc b/corpora/fulltext-test/en/kirjaesittelyt2021.dvc deleted file mode 100644 index 8f31fa538e8853675a98cbd5bdb237da7a699ea8..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/en/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: c689b1b0c1c84f3fcf11afcb41675684 -deps: -- md5: 7d714ce2b7222e511507946d7eed43ca.dir - size: 603188 - nfiles: 894 - path: /data/Annif-corpora-restricted/kirjaesittelyt2021/yso/eng/test/ - hash: md5 -outs: -- md5: 7d714ce2b7222e511507946d7eed43ca.dir - size: 603188 - nfiles: 894 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-test/en/vapaakappaleet-orig.dvc b/corpora/fulltext-test/en/vapaakappaleet-orig.dvc deleted file mode 100644 index 77fc1cdd8f4f2e9c6f33c19d68f542786d31a5c9..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/en/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 2259d4d7676ffa10002217ce68653b73 -deps: -- md5: d2fe684741eb7b4ef338da37a5c24fbf.dir - size: 250449809 - nfiles: 2102 - path: /data/Annif-corpora-local/vapaakappaleet-orig/en/test/ - hash: md5 -outs: -- md5: 015ae9b1152b530ea0b90e430f97ac97.dir - size: 250449809 - nfiles: 2102 - path: vapaakappaleet-orig diff --git a/corpora/fulltext-test/fi/.gitignore b/corpora/fulltext-test/fi/.gitignore deleted file mode 100644 index be81e25e3b86ef9cb2a956251ded9b907a1f7e15..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/.gitignore +++ /dev/null @@ -1,8 +0,0 @@ -/kirjastonhoitaja -/jyu-theses -/satakunnan-kansa-1 -/satakunnan-kansa-2 -/satakunnan-kansa-3 -/satakunnan-kansa-4 -/kirjaesittelyt2021 -/vapaakappaleet-orig diff --git a/corpora/fulltext-test/fi/jyu-theses.dvc b/corpora/fulltext-test/fi/jyu-theses.dvc deleted file mode 100644 index c777d7cd53fb641994c01f74919b294c39ad5dfd..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/jyu-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 788eb07f26d934aaf602601da4c5ce7d -deps: -- md5: 84ad598c69b51ccf24fba7cf5770a8e9.dir - size: 156492348 - nfiles: 1532 - path: /data/Annif-corpora/fulltext/jyu-theses/fin-test/ - hash: md5 -outs: -- md5: 84ad598c69b51ccf24fba7cf5770a8e9.dir - size: 156492348 - nfiles: 1532 - path: jyu-theses - hash: md5 diff --git a/corpora/fulltext-test/fi/kirjaesittelyt2021.dvc b/corpora/fulltext-test/fi/kirjaesittelyt2021.dvc deleted file mode 100644 index e4a79866fbab0ce27bf37d12fc52b4549b560084..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 0686f9e231846e10e78c35062a1439f7 -deps: -- md5: 669296255c4cc5e064bee77d5d0e9f90.dir - size: 1915579 - nfiles: 2622 - path: /data/Annif-corpora-restricted/kirjaesittelyt2021/yso/fin/test/ - hash: md5 -outs: -- md5: 669296255c4cc5e064bee77d5d0e9f90.dir - size: 1915579 - nfiles: 2622 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-test/fi/kirjastonhoitaja.dvc b/corpora/fulltext-test/fi/kirjastonhoitaja.dvc deleted file mode 100644 index 8bad1e44cef11aa23cc64ec12d698052d5ea2022..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/kirjastonhoitaja.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 63d69ce511c1262977bf8db01fe2f752 -deps: -- md5: 90d818aa356f6b5d8e3992817702a0df.dir - size: 466317 - nfiles: 624 - path: /data/Annif-corpora/fulltext/kirjastonhoitaja/test/ - hash: md5 -outs: -- md5: 90d818aa356f6b5d8e3992817702a0df.dir - size: 466317 - nfiles: 624 - path: kirjastonhoitaja - hash: md5 diff --git a/corpora/fulltext-test/fi/satakunnan-kansa-1.dvc b/corpora/fulltext-test/fi/satakunnan-kansa-1.dvc deleted file mode 100644 index b6e44a8ec3cc76617f85ad3fe413a323c92413fb..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/satakunnan-kansa-1.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: ff82a7714027f4a3fd2c5b98b82ac0d8 -deps: -- md5: 70c57356422bf59c87ad4545c6975393.dir - size: 122589 - nfiles: 100 - path: /data/Annif-corpora-restricted/satakunnan-kansa/test1/ - hash: md5 -outs: -- md5: 70c57356422bf59c87ad4545c6975393.dir - size: 122589 - nfiles: 100 - path: satakunnan-kansa-1 diff --git a/corpora/fulltext-test/fi/satakunnan-kansa-2.dvc b/corpora/fulltext-test/fi/satakunnan-kansa-2.dvc deleted file mode 100644 index 9d140e1c190657df4e7f593cd37f4a56abe4eeea..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/satakunnan-kansa-2.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 94f4c590e519ceb83745134706d4941d -deps: -- md5: 438f14417893527d7d4ab74cbc2b63a0.dir - size: 119038 - nfiles: 100 - path: /data/Annif-corpora-restricted/satakunnan-kansa/test2/ - hash: md5 -outs: -- md5: 438f14417893527d7d4ab74cbc2b63a0.dir - size: 119038 - nfiles: 100 - path: satakunnan-kansa-2 diff --git a/corpora/fulltext-test/fi/satakunnan-kansa-3.dvc b/corpora/fulltext-test/fi/satakunnan-kansa-3.dvc deleted file mode 100644 index a59d758d3e4fe10223a62297e71e35dbfa6481f2..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/satakunnan-kansa-3.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 4d840ec899949b09e42ec29f7c0ec592 -deps: -- md5: e8011b405e848a6c7ecb1d54b5f63456.dir - size: 122239 - nfiles: 100 - path: /data/Annif-corpora-restricted/satakunnan-kansa/test3/ - hash: md5 -outs: -- md5: e8011b405e848a6c7ecb1d54b5f63456.dir - size: 122239 - nfiles: 100 - path: satakunnan-kansa-3 diff --git a/corpora/fulltext-test/fi/satakunnan-kansa-4.dvc b/corpora/fulltext-test/fi/satakunnan-kansa-4.dvc deleted file mode 100644 index c29318cf778953a084a2bc05bda749533ec024e4..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/satakunnan-kansa-4.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: d88ed0e407d028bfffd012a84873fe44 -deps: -- md5: 43d672dfadc16ffffd6537f2ffbbd82b.dir - size: 122546 - nfiles: 100 - path: /data/Annif-corpora-restricted/satakunnan-kansa/test4/ - hash: md5 -outs: -- md5: 43d672dfadc16ffffd6537f2ffbbd82b.dir - size: 122546 - nfiles: 100 - path: satakunnan-kansa-4 diff --git a/corpora/fulltext-test/fi/vapaakappaleet-orig.dvc b/corpora/fulltext-test/fi/vapaakappaleet-orig.dvc deleted file mode 100644 index 577ae27c44f1dc4b37ac599e7069d7b755bab6e8..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/fi/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: e3eacb5db12f34d67376e6c6efc022e0 -deps: -- md5: 4d6ea13ca02024a016142c25567ded7d.dir - size: 226413101 - nfiles: 2080 - path: /data/Annif-corpora-local/vapaakappaleet-orig/fi/test/ - hash: md5 -outs: -- md5: ea34a6a16b98b497edb2ea21cf41bb96.dir - size: 226413101 - nfiles: 2080 - path: vapaakappaleet-orig diff --git a/corpora/fulltext-test/sv/.gitignore b/corpora/fulltext-test/sv/.gitignore deleted file mode 100644 index c62954f71699023c78ea8bdadd3e40f0ad0bc1b6..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/sv/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -/abo-theses -/jyu-theses -/kirjaesittelyt2021 -/vapaakappaleet-orig diff --git a/corpora/fulltext-test/sv/abo-theses.dvc b/corpora/fulltext-test/sv/abo-theses.dvc deleted file mode 100644 index eeeb3285f25d4df231cc77fe8085e0dbaf4b95a4..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/sv/abo-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: f3c85e3ead529de7d073ece9515cd097 -deps: -- md5: 49e25736b4df7cdddd4eddc4018ba461.dir - size: 135949330 - nfiles: 452 - path: /data/Annif-corpora/fulltext/abo-theses/swe/test/ - hash: md5 -outs: -- md5: 92766d2134bd95255b00015baa4e5537.dir - size: 17219952 - nfiles: 337 - path: abo-theses - hash: md5 diff --git a/corpora/fulltext-test/sv/jyu-theses.dvc b/corpora/fulltext-test/sv/jyu-theses.dvc deleted file mode 100644 index 663a65bbd97cc3f0bd3c1982f33de0f731d2d6a6..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/sv/jyu-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: cc9aa22b429e0c9f7e45ab53e0b75379 -deps: -- md5: dc55be91f5b3157e4bc2e4b010bc6847.dir - size: 5734687 - nfiles: 64 - path: /data/Annif-corpora/fulltext/jyu-theses/swe-test/ - hash: md5 -outs: -- md5: dc55be91f5b3157e4bc2e4b010bc6847.dir - size: 5734687 - nfiles: 64 - path: jyu-theses - hash: md5 diff --git a/corpora/fulltext-test/sv/kirjaesittelyt2021.dvc b/corpora/fulltext-test/sv/kirjaesittelyt2021.dvc deleted file mode 100644 index 77e390252ae2513931acd8c7168e6e49b9d8df07..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/sv/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 5c70d93cc7f357ab78ea013751785a08 -deps: -- md5: 03528a8434a6e4ba9e35468ce1d8bc7c.dir - size: 677165 - nfiles: 1120 - path: /data/Annif-corpora-restricted/kirjaesittelyt2021/yso/swe/test - hash: md5 -outs: -- md5: 03528a8434a6e4ba9e35468ce1d8bc7c.dir - size: 677165 - nfiles: 1120 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-test/sv/vapaakappaleet-orig.dvc b/corpora/fulltext-test/sv/vapaakappaleet-orig.dvc deleted file mode 100644 index fe5027675dd3c62de82ed28348520c9cdde00e3c..0000000000000000000000000000000000000000 --- a/corpora/fulltext-test/sv/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 1983227e4acbc6ea45546111ddcd20a8 -deps: -- md5: 19cc037d5da0a35ca7df8d587789aed9.dir - size: 43104636 - nfiles: 352 - path: /data/Annif-corpora-local/vapaakappaleet-orig/sv/test/ - hash: md5 -outs: -- md5: 9a0bbb9a19c71df69ec7623ccbff005e.dir - size: 43104636 - nfiles: 352 - path: vapaakappaleet-orig diff --git a/corpora/fulltext-train/en/.gitignore b/corpora/fulltext-train/en/.gitignore deleted file mode 100644 index 9557846ffb85ceec79c8140c62984c9e1e7cd053..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/en/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -/vapaakappaleet-orig -/kirjaesittelyt2021 -/abo-theses -/jyu-theses diff --git a/corpora/fulltext-train/en/abo-theses.dvc b/corpora/fulltext-train/en/abo-theses.dvc deleted file mode 100644 index a255c0c3e91779a8cbec4b81c9cdb8d6b49ef742..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/en/abo-theses.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: ef3e52d478dfc68126670c72a123f172 -deps: -- md5: 3fad4fdc3b9b767e91e6d5f8c78f66f9.dir - size: 844863786 - nfiles: 1343 - path: /data/Annif-corpora-local/fulltext-train/en/abo-theses - hash: md5 -outs: -- md5: bba9a6f747e56765005c9a33c32f078a.dir - size: 47738259 - nfiles: 997 - path: abo-theses diff --git a/corpora/fulltext-train/en/jyu-theses.dvc b/corpora/fulltext-train/en/jyu-theses.dvc deleted file mode 100644 index 95a1fd59c8d5caa61822becdd38da7abe25f1af8..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/en/jyu-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: ad203010df2d637157aac2f7596d576f -deps: -- md5: 6520f92450253e62d8165bd62e4ad0a2.dir - size: 119290546 - nfiles: 998 - path: /data/Annif-corpora-local/fulltext-train/en/jyu-theses - hash: md5 -outs: -- md5: 6520f92450253e62d8165bd62e4ad0a2.dir - size: 119290546 - nfiles: 998 - path: jyu-theses - hash: md5 diff --git a/corpora/fulltext-train/en/kirjaesittelyt2021.dvc b/corpora/fulltext-train/en/kirjaesittelyt2021.dvc deleted file mode 100644 index b4b9f4e3ce5b16d56284576083e897b02dfb6458..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/en/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: c34a43c0b010c387892ef193b06865e0 -deps: -- md5: d5cb69f6a8969fcdb45661c4cd07fd38.dir - size: 1232470 - nfiles: 1652 - path: /data/Annif-corpora-local/fulltext-train/en/kirjaesittelyt2021 - hash: md5 -outs: -- md5: d5cb69f6a8969fcdb45661c4cd07fd38.dir - size: 1232470 - nfiles: 1652 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-train/en/vapaakappaleet-orig.dvc b/corpora/fulltext-train/en/vapaakappaleet-orig.dvc deleted file mode 100644 index b9d2eead72e6917623a99b5ca6d59294b252c4e2..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/en/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: d349357d0fe88fcd67853ddf180ac966 -deps: -- md5: fa2f6d99b3b80057c3277a459b71878b.dir - size: 100089232 - nfiles: 932 - path: /data/Annif-corpora-local/fulltext-train/en/vapaakappaleet-orig - hash: md5 -outs: -- md5: 4e970e8206aa0b8db67cf8e29bac8c4b.dir - size: 100089232 - nfiles: 932 - path: vapaakappaleet-orig diff --git a/corpora/fulltext-train/fi/.gitignore b/corpora/fulltext-train/fi/.gitignore deleted file mode 100644 index d1c582ab6f4c1580a496d64408a0aea1c3bbc396..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/.gitignore +++ /dev/null @@ -1,5 +0,0 @@ -/vapaakappaleet-orig -/kirjaesittelyt2021 -/kirjastonhoitaja -/jyu-theses -/satakunnan-kansa diff --git a/corpora/fulltext-train/fi/jyu-theses.dvc b/corpora/fulltext-train/fi/jyu-theses.dvc deleted file mode 100644 index 678f6fc2f0114db98ca50c47f4390a049a3b94e2..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/jyu-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 51402df088f59db105b4100fcc6d0a22 -deps: -- md5: 5915d784b3dee464cf79dfc7ed8972fd.dir - size: 111873334 - nfiles: 1000 - path: /data/Annif-corpora-local/fulltext-train/fi/jyu-theses - hash: md5 -outs: -- md5: 5915d784b3dee464cf79dfc7ed8972fd.dir - size: 111873334 - nfiles: 1000 - path: jyu-theses - hash: md5 diff --git a/corpora/fulltext-train/fi/kirjaesittelyt2021.dvc b/corpora/fulltext-train/fi/kirjaesittelyt2021.dvc deleted file mode 100644 index a3fa417adb509664e11f04fdce68543298cbe800..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 2f01d0b71ea32c24598860cafe5510b6 -deps: -- md5: 3900e0f22556c3c69f84115ee81bf6fd.dir - size: 693029 - nfiles: 1000 - path: /data/Annif-corpora-local/fulltext-train/fi/kirjaesittelyt2021 - hash: md5 -outs: -- md5: 3900e0f22556c3c69f84115ee81bf6fd.dir - size: 693029 - nfiles: 1000 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-train/fi/kirjastonhoitaja.dvc b/corpora/fulltext-train/fi/kirjastonhoitaja.dvc deleted file mode 100644 index 03684b5c517c9619b89c5a25fdd322d4692a3f2b..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/kirjastonhoitaja.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: c53fbdaa4c77397e06e6ec6ced29fda9 -deps: -- md5: 8979846e372fb084dec5663088f7cdc2.dir - size: 758017 - nfiles: 1000 - path: /data/Annif-corpora-local/fulltext-train/fi/kirjastonhoitaja - hash: md5 -outs: -- md5: 8979846e372fb084dec5663088f7cdc2.dir - size: 758017 - nfiles: 1000 - path: kirjastonhoitaja - hash: md5 diff --git a/corpora/fulltext-train/fi/satakunnan-kansa.dvc b/corpora/fulltext-train/fi/satakunnan-kansa.dvc deleted file mode 100644 index a4eceee5c12e875ea677ce1c44d1e348b865b07b..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/satakunnan-kansa.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 54e7cd77b751dccfb90c850b9fadf144 -deps: -- md5: 477591160a10d6289a6eb92204e64b48.dir - size: 128726 - nfiles: 100 - path: /data/Annif-corpora-local/fulltext-train/fi/satakunnan-kansa - hash: md5 -outs: -- md5: 477591160a10d6289a6eb92204e64b48.dir - size: 128726 - nfiles: 100 - path: satakunnan-kansa diff --git a/corpora/fulltext-train/fi/vapaakappaleet-orig.dvc b/corpora/fulltext-train/fi/vapaakappaleet-orig.dvc deleted file mode 100644 index aa9990ae79c23d26e047ec0d85bab23b5781bdd5..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/fi/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: efcd970176b36e6cf1dc11f135bb48e5 -deps: -- md5: a3fba8d50ca8bab8d53c49dc32c32cf3.dir - size: 300405374 - nfiles: 2477 - path: /data/Annif-corpora-local/fulltext-train/fi/vapaakappaleet-orig - hash: md5 -outs: -- md5: 8f479ecda100df8c244485b1419724dc.dir - size: 300405374 - nfiles: 2477 - path: vapaakappaleet-orig diff --git a/corpora/fulltext-train/sv/.gitignore b/corpora/fulltext-train/sv/.gitignore deleted file mode 100644 index 9557846ffb85ceec79c8140c62984c9e1e7cd053..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/sv/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -/vapaakappaleet-orig -/kirjaesittelyt2021 -/abo-theses -/jyu-theses diff --git a/corpora/fulltext-train/sv/abo-theses.dvc b/corpora/fulltext-train/sv/abo-theses.dvc deleted file mode 100644 index 248d74f560362b6e7f1aa5a4f151450b7a714a0c..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/sv/abo-theses.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 3ef7f112a8a284fb154ca2e7c6c0994f -deps: -- md5: 97cd40ae5e815b3ca56955a87628f048.dir - size: 815900768 - nfiles: 2104 - path: /data/Annif-corpora-local/fulltext-train/sv/abo-theses - hash: md5 -outs: -- md5: 2566088025265f800f55e1f04db01625.dir - size: 81123768 - nfiles: 1565 - path: abo-theses - hash: md5 diff --git a/corpora/fulltext-train/sv/jyu-theses.dvc b/corpora/fulltext-train/sv/jyu-theses.dvc deleted file mode 100644 index 9a9016d6da707d028f9e95ff546a0cc7708460a9..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/sv/jyu-theses.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 5ddae68412821047810dfcda806dd38a -deps: -- md5: c1af325f14aabba73fb65baf77d85543.dir - size: 19596525 - nfiles: 206 - path: /data/Annif-corpora-local/fulltext-train/sv/jyu-theses - hash: md5 -outs: -- md5: c1af325f14aabba73fb65baf77d85543.dir - size: 19596525 - nfiles: 206 - path: jyu-theses diff --git a/corpora/fulltext-train/sv/kirjaesittelyt2021.dvc b/corpora/fulltext-train/sv/kirjaesittelyt2021.dvc deleted file mode 100644 index a332433c87b0ca98e5c7ee0156515fea97b3b13d..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/sv/kirjaesittelyt2021.dvc +++ /dev/null @@ -1,13 +0,0 @@ -md5: 847ede919aacc94af9fef348e8021721 -deps: -- md5: 7bde899cf4bbbf7c8baa0014a3d8634e.dir - size: 918716 - nfiles: 1488 - path: /data/Annif-corpora-local/fulltext-train/sv/kirjaesittelyt2021 - hash: md5 -outs: -- md5: 7bde899cf4bbbf7c8baa0014a3d8634e.dir - size: 918716 - nfiles: 1488 - path: kirjaesittelyt2021 - hash: md5 diff --git a/corpora/fulltext-train/sv/vapaakappaleet-orig.dvc b/corpora/fulltext-train/sv/vapaakappaleet-orig.dvc deleted file mode 100644 index e43a94e8a179fa23af9fa1dbebf40786cf91150b..0000000000000000000000000000000000000000 --- a/corpora/fulltext-train/sv/vapaakappaleet-orig.dvc +++ /dev/null @@ -1,12 +0,0 @@ -md5: 87bd7c808ae8e6c24981c9505c1bf175 -deps: -- md5: 7da4409207e17533342df239c1fa7986.dir - size: 53458333 - nfiles: 490 - path: /data/Annif-corpora-local/fulltext-train/sv/vapaakappaleet-orig - hash: md5 -outs: -- md5: cc91ab54440572bffee62cfa75ecaee2.dir - size: 53458333 - nfiles: 490 - path: vapaakappaleet-orig diff --git a/corpora/shorttext-train/en/.gitignore b/corpora/shorttext-train/en/.gitignore deleted file mode 100644 index 18bddf279679c74f1147fa9ff289efc0f22c708f..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/en/.gitignore +++ /dev/null @@ -1 +0,0 @@ -/yso-finna-en.tsv.gz diff --git a/corpora/shorttext-train/en/yso-finna-en.tsv.gz.dvc b/corpora/shorttext-train/en/yso-finna-en.tsv.gz.dvc deleted file mode 100644 index 516fc7abdd2337a19c6aa378771e61526059abec..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/en/yso-finna-en.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: 71c24280ff9d20993ef67d4ad4d0de61 -deps: -- md5: c45518688682af2848246b8d6db617a5 - size: 98828900 - path: /data/Annif-corpora/training/yso-finna-en.tsv.gz - hash: md5 -outs: -- md5: c45518688682af2848246b8d6db617a5 - size: 98828900 - path: yso-finna-en.tsv.gz - hash: md5 diff --git a/corpora/shorttext-train/fi/.gitignore b/corpora/shorttext-train/fi/.gitignore deleted file mode 100644 index d695cc8dc46ed1d663768a4bf6797f7e4d9a7e42..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/fi/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -/yso-finna-fi-01.tsv.gz -/yso-finna-fi-02.tsv.gz -/yso-finna-fi-03.tsv.gz -/yso-finna-fi-04.tsv.gz diff --git a/corpora/shorttext-train/fi/yso-finna-fi-01.tsv.gz.dvc b/corpora/shorttext-train/fi/yso-finna-fi-01.tsv.gz.dvc deleted file mode 100644 index 3ac5a27bcf73c436f55a55d9f7aa019abf7b7c1e..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/fi/yso-finna-fi-01.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: 564736494bd3e9f6dcbad43db05e4fcd -deps: -- md5: 3d5ecada7cbe4ad320c0e5f8a54846f7 - size: 99401083 - path: /data/Annif-corpora/training/yso-finna-fi-01.tsv.gz - hash: md5 -outs: -- md5: 3d5ecada7cbe4ad320c0e5f8a54846f7 - size: 99401083 - path: yso-finna-fi-01.tsv.gz - hash: md5 diff --git a/corpora/shorttext-train/fi/yso-finna-fi-02.tsv.gz.dvc b/corpora/shorttext-train/fi/yso-finna-fi-02.tsv.gz.dvc deleted file mode 100644 index 96dc2e7c107b0c7dc55925a447697f476a3952ff..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/fi/yso-finna-fi-02.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: f6efdf2e9a81c0369226f14d007a7af8 -deps: -- md5: 3dac9f314f705870ed66e9243a07ab70 - size: 99477086 - path: /data/Annif-corpora/training/yso-finna-fi-02.tsv.gz - hash: md5 -outs: -- md5: 3dac9f314f705870ed66e9243a07ab70 - size: 99477086 - path: yso-finna-fi-02.tsv.gz - hash: md5 diff --git a/corpora/shorttext-train/fi/yso-finna-fi-03.tsv.gz.dvc b/corpora/shorttext-train/fi/yso-finna-fi-03.tsv.gz.dvc deleted file mode 100644 index 7b624723ff611f43d36ca37eaa71a98d8d6daa27..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/fi/yso-finna-fi-03.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: e5b52fed49c0c9f64110aa4254e17885 -deps: -- md5: be00ff7dc5fd91d7e13b366e0089089c - size: 99411108 - path: /data/Annif-corpora/training/yso-finna-fi-03.tsv.gz - hash: md5 -outs: -- md5: be00ff7dc5fd91d7e13b366e0089089c - size: 99411108 - path: yso-finna-fi-03.tsv.gz - hash: md5 diff --git a/corpora/shorttext-train/fi/yso-finna-fi-04.tsv.gz.dvc b/corpora/shorttext-train/fi/yso-finna-fi-04.tsv.gz.dvc deleted file mode 100644 index 7ef6865f0d981c48e46fee0e7e535c7ae5484e0d..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/fi/yso-finna-fi-04.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: a7af220a29741c0b213676bd77c221ee -deps: -- md5: a8f16f5d33de8c534010e7d2191a5e71 - size: 60270897 - path: /data/Annif-corpora/training/yso-finna-fi-04.tsv.gz - hash: md5 -outs: -- md5: a8f16f5d33de8c534010e7d2191a5e71 - size: 60270897 - path: yso-finna-fi-04.tsv.gz - hash: md5 diff --git a/corpora/shorttext-train/sv/.gitignore b/corpora/shorttext-train/sv/.gitignore deleted file mode 100644 index c133f04cc1cecf3b684e7d59b319163ef860b6eb..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/sv/.gitignore +++ /dev/null @@ -1 +0,0 @@ -/yso-finna-sv.tsv.gz diff --git a/corpora/shorttext-train/sv/yso-finna-sv.tsv.gz.dvc b/corpora/shorttext-train/sv/yso-finna-sv.tsv.gz.dvc deleted file mode 100644 index 34b5811673a9d3aea38cfee0c4ae043cd15713f5..0000000000000000000000000000000000000000 --- a/corpora/shorttext-train/sv/yso-finna-sv.tsv.gz.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: 629c63b1650efa48cdd983002bcaabb5 -deps: -- md5: 35b3ae2a8eba426645c0bf22fd609b87 - size: 40515070 - path: /data/Annif-corpora/training/yso-finna-sv.tsv.gz - hash: md5 -outs: -- md5: 35b3ae2a8eba426645c0bf22fd609b87 - size: 40515070 - path: yso-finna-sv.tsv.gz - hash: md5 diff --git a/corpora/yso-skos.ttl.dvc b/corpora/yso-skos.ttl.dvc deleted file mode 100644 index 768324670e5b76c52b96459163e7a723ac440c22..0000000000000000000000000000000000000000 --- a/corpora/yso-skos.ttl.dvc +++ /dev/null @@ -1,11 +0,0 @@ -md5: 723d53e9f88b27482a340c3f4e02f5a7 -deps: -- md5: c3d9a5148c46efa4fbf11ee866154ebf - size: 32953533 - path: /data/Annif-corpora/vocab/yso-skos.ttl - hash: md5 -outs: -- md5: c3d9a5148c46efa4fbf11ee866154ebf - size: 32953533 - path: yso-skos.ttl - hash: md5 diff --git a/dvc.lock b/dvc.lock deleted file mode 100644 index 90887ee8396928437e4c54af3b19a992c7a21f7f..0000000000000000000000000000000000000000 --- a/dvc.lock +++ /dev/null @@ -1,1568 +0,0 @@ -schema: '2.0' -stages: - install: - cmd: - - python3 -m venv venv - - source venv/bin/activate && pip install -U pip wheel setuptools && pip install - -r requirements.txt - - cp requirements.txt venv-installed - deps: - - path: requirements.txt - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - outs: - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - load-vocab: - cmd: venv/bin/annif load-vocab --force yso corpora/yso-skos.ttl - deps: - - path: corpora/yso-skos.ttl - hash: md5 - md5: c3d9a5148c46efa4fbf11ee866154ebf - size: 32953533 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - outs: - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - train-mllm@fi: - cmd: venv/bin/annif train yso-mllm-fi corpora/fulltext-train/fi/*/ -j 16 - deps: - - path: corpora/fulltext-train/fi - hash: md5 - md5: f5c1820afb398fa8145181cf22905336.dir - size: 413860133 - nfiles: 5583 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-fi: - name: YSO MLLM Finnish - language: fi - backend: mllm - analyzer: voikko(fi) - vocab: yso - limit: '1000' - transform: limit(3000000) - access: hidden - outs: - - path: data/projects/yso-mllm-fi - hash: md5 - md5: fb05a7c1e6b2ed72fee85fbbc5b7374b.dir - size: 36157873 - nfiles: 2 - train-mllm@en: - cmd: venv/bin/annif train yso-mllm-en corpora/fulltext-train/en/*/ -j 16 - deps: - - path: corpora/fulltext-train/en - hash: md5 - md5: 426f141f77c5bac77b32784e4b827d31.dir - size: 268351812 - nfiles: 4584 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-en: - name: YSO MLLM English - language: en - backend: mllm - analyzer: snowball(english) - vocab: yso - limit: '1000' - transform: limit(2500000) - access: hidden - outs: - - path: data/projects/yso-mllm-en - hash: md5 - md5: 129793bd06d231413a66ed5611180dbe.dir - size: 39175771 - nfiles: 2 - train-mllm@sv: - cmd: venv/bin/annif train yso-mllm-sv corpora/fulltext-train/sv/*/ -j 16 - deps: - - path: corpora/fulltext-train/sv - hash: md5 - md5: c64480b5f34b1895db972d774abe12cc.dir - size: 155098642 - nfiles: 3754 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-sv: - name: YSO MLLM Swedish - language: sv - backend: mllm - analyzer: snowball(swedish) - vocab: yso - limit: '1000' - transform: limit(3000000) - access: hidden - outs: - - path: data/projects/yso-mllm-sv - hash: md5 - md5: 88dd945c235bdfc4549fa44f5ea582a4.dir - size: 19736546 - nfiles: 2 - train-omikuji@sv: - cmd: venv/bin/annif train yso-bonsai-sv corpora/shorttext-train/sv/yso-finna-sv*.tsv.gz - deps: - - path: corpora/shorttext-train/sv/ - hash: md5 - md5: 33a49e42ec12daf0c973c792928a2cb0.dir - size: 40515364 - nfiles: 3 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-sv: - name: YSO Omikuji Bonsai Swedish - language: sv - backend: omikuji - analyzer: snowball(swedish) - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '2' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: data/projects/yso-bonsai-sv - hash: md5 - md5: 77a54001b5d18a7a474bcf7d4e9577e1.dir - size: 1029311381 - nfiles: 6 - train-omikuji@en: - cmd: venv/bin/annif train yso-bonsai-en corpora/shorttext-train/en/yso-finna-en*.tsv.gz - deps: - - path: corpora/shorttext-train/en/ - hash: md5 - md5: 19e76af78210f39cf02a5c8bbee38f60.dir - size: 98829194 - nfiles: 3 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-en: - name: YSO Omikuji Bonsai English - language: en - backend: omikuji - analyzer: simple - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '5' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: data/projects/yso-bonsai-en - hash: md5 - md5: b1a5a925fbe1a11c153ce3133176c717.dir - size: 2321603849 - nfiles: 6 - train-omikuji@fi: - cmd: venv/bin/annif train yso-bonsai-fi corpora/shorttext-train/fi/yso-finna-fi*.tsv.gz - deps: - - path: corpora/shorttext-train/fi/ - hash: md5 - md5: 5a2f47124433a7215e6c391a48c0aeca.dir - size: 358561386 - nfiles: 9 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-fi: - name: YSO Omikuji Bonsai Finnish - language: fi - backend: omikuji - analyzer: snowball(finnish) - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '5' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: data/projects/yso-bonsai-fi - hash: md5 - md5: 84a1ae3ca24702d0cbb901f3215e1675.dir - size: 5354349482 - nfiles: 6 - train-fasttext@sv: - cmd: venv/bin/annif train yso-fasttext-sv corpora/shorttext-train/sv/yso-finna-sv*.tsv.gz - deps: - - path: corpora/shorttext-train/sv/ - hash: md5 - md5: 33a49e42ec12daf0c973c792928a2cb0.dir - size: 40515364 - nfiles: 3 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-sv: - name: YSO fastText Swedish - language: sv - backend: fasttext - analyzer: snowball(swedish) - dim: '560' - lr: '0.974349' - epoch: '110' - minn: '2' - maxn: '6' - minCount: '2' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: data/projects/yso-fasttext-sv - hash: md5 - md5: eccbf5ba8ce07f2777b4bee583ed783b.dir - size: 4913883852 - nfiles: 2 - train-fasttext@fi: - cmd: venv/bin/annif train yso-fasttext-fi corpora/shorttext-train/fi/yso-finna-fi*.tsv.gz - deps: - - path: corpora/shorttext-train/fi/ - hash: md5 - md5: 5a2f47124433a7215e6c391a48c0aeca.dir - size: 358561386 - nfiles: 9 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-fi: - name: YSO fastText Finnish - language: fi - backend: fasttext - analyzer: voikko(fi) - dim: '660' - lr: '0.506539' - epoch: '75' - minn: '2' - maxn: '7' - minCount: '2' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: data/projects/yso-fasttext-fi - hash: md5 - md5: 06ea14ea97351f72fbaeb13517011c88.dir - size: 7547313091 - nfiles: 2 - train-fasttext@en: - cmd: venv/bin/annif train yso-fasttext-en corpora/shorttext-train/en/yso-finna-en*.tsv.gz - deps: - - path: corpora/shorttext-train/en/ - hash: md5 - md5: 19e76af78210f39cf02a5c8bbee38f60.dir - size: 98829194 - nfiles: 3 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-en: - name: YSO fastText English - language: en - backend: fasttext - analyzer: snowball(english) - dim: '430' - lr: '0.506539' - epoch: '115' - minn: '4' - maxn: '5' - minCount: '1' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: data/projects/yso-fasttext-en - hash: md5 - md5: ca3fa22f9a2b6d46246e351c7a1e3256.dir - size: 4091197300 - nfiles: 2 - train-nn-ensemble@sv: - cmd: venv/bin/annif train yso-sv corpora/fulltext-train/sv/*/ -j 16 - deps: - - path: corpora/fulltext-train/sv - hash: md5 - md5: c64480b5f34b1895db972d774abe12cc.dir - size: 155098642 - nfiles: 3754 - - path: data/projects/yso-bonsai-sv - hash: md5 - md5: 77a54001b5d18a7a474bcf7d4e9577e1.dir - size: 1029311381 - nfiles: 6 - - path: data/projects/yso-fasttext-sv - hash: md5 - md5: eccbf5ba8ce07f2777b4bee583ed783b.dir - size: 4913883852 - nfiles: 2 - - path: data/projects/yso-mllm-sv - hash: md5 - md5: 88dd945c235bdfc4549fa44f5ea582a4.dir - size: 19736546 - nfiles: 2 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-sv: - name: ALLFO svenska (2023.1.Ghosha) - language: sv - backend: nn_ensemble - sources: yso-mllm-sv:0.1439,yso-fasttext-sv:0.3302,yso-bonsai-sv:0.5259 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: data/projects/yso-sv - hash: md5 - md5: 9cab94546614acf5ac28514687cf26ad.dir - size: 1259460957 - nfiles: 3 - train-nn-ensemble@en: - cmd: venv/bin/annif train yso-en corpora/fulltext-train/en/*/ -j 16 - deps: - - path: corpora/fulltext-train/en - hash: md5 - md5: 426f141f77c5bac77b32784e4b827d31.dir - size: 268351812 - nfiles: 4584 - - path: data/projects/yso-bonsai-en - hash: md5 - md5: b1a5a925fbe1a11c153ce3133176c717.dir - size: 2321603849 - nfiles: 6 - - path: data/projects/yso-fasttext-en - hash: md5 - md5: ca3fa22f9a2b6d46246e351c7a1e3256.dir - size: 4091197300 - nfiles: 2 - - path: data/projects/yso-mllm-en - hash: md5 - md5: 129793bd06d231413a66ed5611180dbe.dir - size: 39175771 - nfiles: 2 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-en: - name: YSO English (2023.1.Ghosha) - language: en - backend: nn_ensemble - sources: yso-mllm-en:0.3426,yso-fasttext-en:0.1419,yso-bonsai-en:0.5155 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: data/projects/yso-en - hash: md5 - md5: cf09853815a7e8c621b352c2bc70f9e1.dir - size: 1259460957 - nfiles: 3 - train-nn-ensemble@fi: - cmd: venv/bin/annif train yso-fi corpora/fulltext-train/fi/*/ -j 16 - deps: - - path: corpora/fulltext-train/fi - hash: md5 - md5: f5c1820afb398fa8145181cf22905336.dir - size: 413860133 - nfiles: 5583 - - path: data/projects/yso-bonsai-fi - hash: md5 - md5: 84a1ae3ca24702d0cbb901f3215e1675.dir - size: 5354349482 - nfiles: 6 - - path: data/projects/yso-fasttext-fi - hash: md5 - md5: 06ea14ea97351f72fbaeb13517011c88.dir - size: 7547313091 - nfiles: 2 - - path: data/projects/yso-mllm-fi - hash: md5 - md5: fb05a7c1e6b2ed72fee85fbbc5b7374b.dir - size: 36157873 - nfiles: 2 - - path: data/vocabs/yso - hash: md5 - md5: ad8a2ea707ceac4f455d2b3b2f7a7c90.dir - size: 61626265 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fi: - name: YSO suomi (2023.1.Ghosha) - language: fi - backend: nn_ensemble - sources: yso-mllm-fi:0.1492,yso-fasttext-fi:0.6090,yso-bonsai-fi:0.2418 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: data/projects/yso-fi - hash: md5 - md5: a0961bd4dfcfee8367b22ad17d780dbb.dir - size: 1259460957 - nfiles: 3 - eval-fi@mllm: - cmd: - - venv/bin/annif eval yso-mllm-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-mllm-fi -j 6 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-mllm-fi -j 4 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-mllm-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - md5: 169cae26a0b93733aed807f8e9d9ca40.dir - size: 385626157 - nfiles: 7267 - - path: data/projects/yso-mllm-fi - md5: c42dd794bb146ecf7cf60e5f49167ab2.dir - size: 36078206 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/mllm-fi-jyu-theses.json - md5: 017241308807e76f5c4ecc675c280bfa - size: 92 - - path: reports/mllm-fi-kirjaesittelyt2021.json - md5: b289df9d0158a0e8c6cd74dd2df7d9c2 - size: 94 - - path: reports/mllm-fi-kirjastonhoitaja.json - md5: 1ed10e3905a72ffe5181bc74ec27599d - size: 93 - - path: reports/mllm-fi-satakunnan-kansa.json - md5: 7100b02bb18dad3f8e8d073b7bff9e50 - size: 93 - - path: reports/mllm-fi-vapaakappaleet-orig.json - md5: bb3ed842b8468708c1c064452b6b24f1 - size: 93 - eval-fi@bonsai: - cmd: - - venv/bin/annif eval yso-bonsai-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-bonsai-fi -j 6 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-bonsai-fi -j 4 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-bonsai-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - md5: 169cae26a0b93733aed807f8e9d9ca40.dir - size: 385626157 - nfiles: 7267 - - path: data/projects/yso-bonsai-fi - md5: e065094c56ff3a61e2a4648e47156c8f.dir - size: 5277448954 - nfiles: 6 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/bonsai-fi-jyu-theses.json - md5: 660095be4380591aa8c7f839ffbf1aa4 - size: 92 - - path: reports/bonsai-fi-kirjaesittelyt2021.json - md5: 61540b3360540911b4ec3a934abb74ca - size: 94 - - path: reports/bonsai-fi-kirjastonhoitaja.json - md5: 95696330bd44571520752bdb4a460287 - size: 93 - - path: reports/bonsai-fi-satakunnan-kansa.json - md5: 2bb5923811b0760695cb5ad2465469e5 - size: 93 - - path: reports/bonsai-fi-vapaakappaleet-orig.json - md5: d369966bf45650f6b9643469343ab65d - size: 93 - eval-fi@fasttext: - cmd: - - venv/bin/annif eval yso-fasttext-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-fasttext-fi -j 6 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-fasttext-fi -j 4 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-fasttext-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - md5: 169cae26a0b93733aed807f8e9d9ca40.dir - size: 385626157 - nfiles: 7267 - - path: data/projects/yso-fasttext-fi - md5: 0f5c9eb966671d610e8d8556b270652c.dir - size: 7519964475 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/fasttext-fi-jyu-theses.json - md5: 339a05b2faa098074a7717ef6677048b - size: 93 - - path: reports/fasttext-fi-kirjaesittelyt2021.json - md5: b0425400c2513d8580bf726bc0b487d3 - size: 94 - - path: reports/fasttext-fi-kirjastonhoitaja.json - md5: 7fe1e06726e05effef46a1cb97078d72 - size: 92 - - path: reports/fasttext-fi-satakunnan-kansa.json - md5: 63b1b03ad124c17bc3570188f1855ec9 - size: 93 - - path: reports/fasttext-fi-vapaakappaleet-orig.json - md5: 0acd6e385633742bd18513109a5a87dc - size: 94 - eval-sv@mllm: - cmd: - - venv/bin/annif eval yso-mllm-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-mllm-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-mllm-sv -j 6 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - md5: 2d29064639aa0f8900e4bf58a781826e.dir - size: 66510456 - nfiles: 1876 - - path: data/projects/yso-mllm-sv - md5: fc2e344bac4abd0048c7c286ffdba0eb.dir - size: 19904278 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/mllm-sv-abo-theses.json - md5: 0d6f871b92dfb263c56e17f644e99513 - size: 92 - - path: reports/mllm-sv-jyu-theses.json - md5: 81f8eb0a6908169aeb2ba236a0baf3f7 - size: 91 - - path: reports/mllm-sv-kirjaesittelyt2021.json - md5: 0886460a79805dd7698d6c7f60f2ab2d - size: 93 - - path: reports/mllm-sv-vapaakappaleet-orig.json - md5: d55b237633198997361b7aab0ba7d10b - size: 92 - eval-sv@bonsai: - cmd: - - venv/bin/annif eval yso-bonsai-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-bonsai-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-bonsai-sv -j 6 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - md5: 2d29064639aa0f8900e4bf58a781826e.dir - size: 66510456 - nfiles: 1876 - - path: data/projects/yso-bonsai-sv - md5: 325628ed10b5330b42fdf93489ab870d.dir - size: 1035735788 - nfiles: 6 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/bonsai-sv-abo-theses.json - md5: 65fd5a684e967d01b8527489b6ca38c3 - size: 93 - - path: reports/bonsai-sv-jyu-theses.json - md5: 69084213079115f7f97cf16115d58010 - size: 91 - - path: reports/bonsai-sv-kirjaesittelyt2021.json - md5: 1c2fe10783ce03e22dede97beb120c65 - size: 93 - - path: reports/bonsai-sv-vapaakappaleet-orig.json - md5: f6e85d73d66e9f6cfc0cfa359ba6de73 - size: 92 - eval-sv@fasttext: - cmd: - - venv/bin/annif eval yso-fasttext-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-fasttext-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-fasttext-sv -j 6 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - md5: 2d29064639aa0f8900e4bf58a781826e.dir - size: 66510456 - nfiles: 1876 - - path: data/projects/yso-fasttext-sv - md5: 8d272558b0ec13606c0027af74be47e8.dir - size: 4915577395 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/fasttext-sv-abo-theses.json - md5: ea4e5c8d3c72e853ce0bda1757cd4e41 - size: 93 - - path: reports/fasttext-sv-jyu-theses.json - md5: 8e73efa922698dd1fa968f6dba4c779c - size: 92 - - path: reports/fasttext-sv-kirjaesittelyt2021.json - md5: 42fa4ff3fa951ecf07ce404a1c68d990 - size: 93 - - path: reports/fasttext-sv-vapaakappaleet-orig.json - md5: c8c0290553be948bbd127572efc0fa95 - size: 93 - eval-en@fasttext: - cmd: - - venv/bin/annif eval yso-fasttext-en -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-fasttext-en -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-fasttext-en -j 6 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-en -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - md5: e6fd23c87a07631f24e52f568fad23ea.dir - size: 331772939 - nfiles: 3825 - - path: data/projects/yso-fasttext-en - md5: 625be5253a0eee61c44741e63acb1021.dir - size: 4077282108 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/fasttext-en-abo-theses.json - md5: 66940ff0ea42d44ab1b429905c51858a - size: 93 - - path: reports/fasttext-en-jyu-theses.json - md5: 8bdfe2503c1a002d31121f48daf493db - size: 92 - - path: reports/fasttext-en-kirjaesittelyt2021.json - md5: 1d8d4ce09c768c644d34382497b5bb15 - size: 92 - - path: reports/fasttext-en-vapaakappaleet-orig.json - md5: a8060ee8a76b5b3e7a5134d0edfdc5d9 - size: 92 - eval-en@mllm: - cmd: - - venv/bin/annif eval yso-mllm-en -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-mllm-en -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-mllm-en -j 6 -m F1@5 -m NDCG --metrics-file reports/mllm-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-en -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - md5: e6fd23c87a07631f24e52f568fad23ea.dir - size: 331772939 - nfiles: 3825 - - path: data/projects/yso-mllm-en - md5: 03fe928341b019387a004ce33b85b220.dir - size: 38643312 - nfiles: 2 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/mllm-en-abo-theses.json - md5: 0cc714a7ee3c0a5d42f941da5719dd8c - size: 92 - - path: reports/mllm-en-jyu-theses.json - md5: c8bdc5f4f57799201b97ed62191c608c - size: 92 - - path: reports/mllm-en-kirjaesittelyt2021.json - md5: 121d0ac40869b4dd7b0a1fc6094e2c42 - size: 94 - - path: reports/mllm-en-vapaakappaleet-orig.json - md5: e763f75300be26b3699d40a4cc119526 - size: 94 - eval-en@bonsai: - cmd: - - venv/bin/annif eval yso-bonsai-en -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-bonsai-en -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-bonsai-en -j 6 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-en -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - md5: e6fd23c87a07631f24e52f568fad23ea.dir - size: 331772939 - nfiles: 3825 - - path: data/projects/yso-bonsai-en - md5: 7b3c8486bbce25710e969ce706e9ac71.dir - size: 2246619174 - nfiles: 6 - - path: venv-installed - md5: abf841d22e1cdf25eb9e9ef2368c240d - size: 49 - outs: - - path: reports/bonsai-en-abo-theses.json - md5: 109beeb488b023241e61fb4fddba1d35 - size: 93 - - path: reports/bonsai-en-jyu-theses.json - md5: f917141ae8af29e6834359c35998e98d - size: 92 - - path: reports/bonsai-en-kirjaesittelyt2021.json - md5: 711601791fe7245a961e0bd5fdb6dccd - size: 92 - - path: reports/bonsai-en-vapaakappaleet-orig.json - md5: 2324260b996366b775fbf3c38a14a309 - size: 93 - eval-sv@fasttext-sv: - cmd: - - venv/bin/annif eval yso-fasttext-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-fasttext-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-fasttext-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - hash: md5 - md5: 8424019ec4c261915627865324ac2f73.dir - size: 66737727 - nfiles: 1878 - - path: data/projects/yso-fasttext-sv - hash: md5 - md5: eccbf5ba8ce07f2777b4bee583ed783b.dir - size: 4913883852 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-sv: - name: YSO fastText Swedish - language: sv - backend: fasttext - analyzer: snowball(swedish) - dim: '560' - lr: '0.974349' - epoch: '110' - minn: '2' - maxn: '6' - minCount: '2' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: reports/fasttext-sv-abo-theses.json - hash: md5 - md5: 50e4860a0829e104eb2051c13b7786c2 - size: 94 - - path: reports/fasttext-sv-jyu-theses.json - hash: md5 - md5: 194c5356f9792f2d6938c26ffc935f37 - size: 91 - - path: reports/fasttext-sv-kirjaesittelyt2021.json - hash: md5 - md5: c49bb0271033d10d951ccce116a0c0c4 - size: 93 - - path: reports/fasttext-sv-vapaakappaleet-orig.json - hash: md5 - md5: a85d38c085eac4297435d392077cb7f6 - size: 94 - eval-sv@sv: - cmd: - - venv/bin/annif eval yso-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - hash: md5 - md5: 8424019ec4c261915627865324ac2f73.dir - size: 66737727 - nfiles: 1878 - - path: data/projects/yso-sv - hash: md5 - md5: 9cab94546614acf5ac28514687cf26ad.dir - size: 1259460957 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-sv: - name: ALLFO svenska (2023.1.Ghosha) - language: sv - backend: nn_ensemble - sources: yso-mllm-sv:0.1439,yso-fasttext-sv:0.3302,yso-bonsai-sv:0.5259 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: reports/sv-abo-theses.json - hash: md5 - md5: 87917904500a6eefe80e1924a0cebe99 - size: 92 - - path: reports/sv-jyu-theses.json - hash: md5 - md5: bb440b30d4c995bc9d9ffef140b53817 - size: 91 - - path: reports/sv-kirjaesittelyt2021.json - hash: md5 - md5: e9b35e7f38fa85945eb1278bdf01ed30 - size: 92 - - path: reports/sv-vapaakappaleet-orig.json - hash: md5 - md5: dbd8e93725f32ba3fcba848e9cb61876 - size: 92 - eval-en@bonsai-en: - cmd: - - venv/bin/annif eval yso-bonsai-en -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-bonsai-en -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-bonsai-en -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-en -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - hash: md5 - md5: f1883db613d00c7bce08a1211787f984.dir - size: 332654375 - nfiles: 3831 - - path: data/projects/yso-bonsai-en - hash: md5 - md5: b1a5a925fbe1a11c153ce3133176c717.dir - size: 2321603849 - nfiles: 6 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-en: - name: YSO Omikuji Bonsai English - language: en - backend: omikuji - analyzer: simple - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '5' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: reports/bonsai-en-abo-theses.json - hash: md5 - md5: 23fb86f0954118ead50392c7801ce834 - size: 91 - - path: reports/bonsai-en-jyu-theses.json - hash: md5 - md5: 0f8b8c4db657ed8aac3b494a98fa6c15 - size: 92 - - path: reports/bonsai-en-kirjaesittelyt2021.json - hash: md5 - md5: f047174bbd0de4588072412f2431d288 - size: 92 - - path: reports/bonsai-en-vapaakappaleet-orig.json - hash: md5 - md5: 3091f929ce7e0b070d7fc2dfd0fff4c0 - size: 92 - eval-en@en: - cmd: - - venv/bin/annif eval yso-en -j 1 -m F1@5 -m NDCG --metrics-file reports/en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-en -j 1 -m F1@5 -m NDCG --metrics-file reports/en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-en -j 10 -m F1@5 -m NDCG --metrics-file reports/en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-en -j 10 -m F1@5 -m NDCG --metrics-file reports/en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - hash: md5 - md5: f1883db613d00c7bce08a1211787f984.dir - size: 332654375 - nfiles: 3831 - - path: data/projects/yso-en - hash: md5 - md5: cf09853815a7e8c621b352c2bc70f9e1.dir - size: 1259460957 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-en: - name: YSO English (2023.1.Ghosha) - language: en - backend: nn_ensemble - sources: yso-mllm-en:0.3426,yso-fasttext-en:0.1419,yso-bonsai-en:0.5155 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: reports/en-abo-theses.json - hash: md5 - md5: 8d526675b9af9b477a94213eb1254841 - size: 91 - - path: reports/en-jyu-theses.json - hash: md5 - md5: 623001128bd0cdf0caedcc97457f0b1c - size: 92 - - path: reports/en-kirjaesittelyt2021.json - hash: md5 - md5: 0dbf7262d50072866df3943231f3517e - size: 92 - - path: reports/en-vapaakappaleet-orig.json - hash: md5 - md5: 6968cecb575018eb88e530ad2997888b - size: 94 - eval-en@fasttext-en: - cmd: - - venv/bin/annif eval yso-fasttext-en -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-fasttext-en -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-fasttext-en -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-en -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - hash: md5 - md5: f1883db613d00c7bce08a1211787f984.dir - size: 332654375 - nfiles: 3831 - - path: data/projects/yso-fasttext-en - hash: md5 - md5: ca3fa22f9a2b6d46246e351c7a1e3256.dir - size: 4091197300 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-en: - name: YSO fastText English - language: en - backend: fasttext - analyzer: snowball(english) - dim: '430' - lr: '0.506539' - epoch: '115' - minn: '4' - maxn: '5' - minCount: '1' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: reports/fasttext-en-abo-theses.json - hash: md5 - md5: 931f2f652825441088cac77c30dbf3a9 - size: 92 - - path: reports/fasttext-en-jyu-theses.json - hash: md5 - md5: bcad7d3bae638d6008d4898ae93cd52a - size: 93 - - path: reports/fasttext-en-kirjaesittelyt2021.json - hash: md5 - md5: 85e8cdaa79beb6cfa897146bb45460b2 - size: 93 - - path: reports/fasttext-en-vapaakappaleet-orig.json - hash: md5 - md5: 55ce7777768ebcb9421c06551fa9235d - size: 93 - eval-fi@fasttext-fi: - cmd: - - venv/bin/annif eval yso-fasttext-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-fasttext-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fasttext-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-fasttext-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-fasttext-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fasttext-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - hash: md5 - md5: 0fe8c48afc24e8c843a78fd6d76609a4.dir - size: 385776340 - nfiles: 7267 - - path: data/projects/yso-fasttext-fi - hash: md5 - md5: 06ea14ea97351f72fbaeb13517011c88.dir - size: 7547313091 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fasttext-fi: - name: YSO fastText Finnish - language: fi - backend: fasttext - analyzer: voikko(fi) - dim: '660' - lr: '0.506539' - epoch: '75' - minn: '2' - maxn: '7' - minCount: '2' - wordNgrams: '2' - loss: hs - limit: '1000' - chunksize: '24' - vocab: yso - transform: limit(15000),filter_lang,limit(5000) - access: hidden - outs: - - path: reports/fasttext-fi-jyu-theses.json - hash: md5 - md5: f44b29e37e940212365d1faba24d807f - size: 94 - - path: reports/fasttext-fi-kirjaesittelyt2021.json - hash: md5 - md5: 8557276f66c15738048271a5890256ea - size: 93 - - path: reports/fasttext-fi-kirjastonhoitaja.json - hash: md5 - md5: 1f88e3fc726015d206d9323c3b4efab3 - size: 92 - - path: reports/fasttext-fi-satakunnan-kansa.json - hash: md5 - md5: c9e037fb4399ee097b07318090a503fa - size: 94 - - path: reports/fasttext-fi-vapaakappaleet-orig.json - hash: md5 - md5: 52c838f448cf38789a5a73e72eb121c5 - size: 93 - eval-fi@mllm-fi: - cmd: - - venv/bin/annif eval yso-mllm-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-mllm-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-mllm-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-mllm-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - hash: md5 - md5: 0fe8c48afc24e8c843a78fd6d76609a4.dir - size: 385776340 - nfiles: 7267 - - path: data/projects/yso-mllm-fi - hash: md5 - md5: fb05a7c1e6b2ed72fee85fbbc5b7374b.dir - size: 36157873 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-fi: - name: YSO MLLM Finnish - language: fi - backend: mllm - analyzer: voikko(fi) - vocab: yso - limit: '1000' - transform: limit(3000000) - access: hidden - outs: - - path: reports/mllm-fi-jyu-theses.json - hash: md5 - md5: e6a84d414ed22b9d9d3fee4a7d955d85 - size: 92 - - path: reports/mllm-fi-kirjaesittelyt2021.json - hash: md5 - md5: 6f2cb724ae31fb838b22731c460974d2 - size: 94 - - path: reports/mllm-fi-kirjastonhoitaja.json - hash: md5 - md5: 3247bcd04d35ffb318d7e48666a90976 - size: 94 - - path: reports/mllm-fi-satakunnan-kansa.json - hash: md5 - md5: 016830e79dfffc8d5ea99fa523e31ba2 - size: 93 - - path: reports/mllm-fi-vapaakappaleet-orig.json - hash: md5 - md5: aa637226b7c53ecea64bca339d20687c - size: 93 - eval-fi@bonsai-fi: - cmd: - - venv/bin/annif eval yso-bonsai-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-bonsai-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-bonsai-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-bonsai-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - hash: md5 - md5: 0fe8c48afc24e8c843a78fd6d76609a4.dir - size: 385776340 - nfiles: 7267 - - path: data/projects/yso-bonsai-fi - hash: md5 - md5: 84a1ae3ca24702d0cbb901f3215e1675.dir - size: 5354349482 - nfiles: 6 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-fi: - name: YSO Omikuji Bonsai Finnish - language: fi - backend: omikuji - analyzer: snowball(finnish) - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '5' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: reports/bonsai-fi-jyu-theses.json - hash: md5 - md5: 76e3cc2484c71e18d965fdd4acd1ba8a - size: 93 - - path: reports/bonsai-fi-kirjaesittelyt2021.json - hash: md5 - md5: f2987884156a29aa053ad510993e74a2 - size: 94 - - path: reports/bonsai-fi-kirjastonhoitaja.json - hash: md5 - md5: f12f961721bb62fb1a9e94e5d2636aec - size: 92 - - path: reports/bonsai-fi-satakunnan-kansa.json - hash: md5 - md5: ad6ad6e95073021b40aaa643171e5773 - size: 93 - - path: reports/bonsai-fi-vapaakappaleet-orig.json - hash: md5 - md5: 5904dff99322d3b6019d9d7a1aa7cd8b - size: 93 - eval-en@mllm-en: - cmd: - - venv/bin/annif eval yso-mllm-en -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-en-abo-theses.json - corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-mllm-en -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-en-jyu-theses.json - corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-mllm-en -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-en-kirjaesittelyt2021.json - corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-en -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-en-vapaakappaleet-orig.json - corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/en - hash: md5 - md5: f1883db613d00c7bce08a1211787f984.dir - size: 332654375 - nfiles: 3831 - - path: data/projects/yso-mllm-en - hash: md5 - md5: 129793bd06d231413a66ed5611180dbe.dir - size: 39175771 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-en: - name: YSO MLLM English - language: en - backend: mllm - analyzer: snowball(english) - vocab: yso - limit: '1000' - transform: limit(2500000) - access: hidden - outs: - - path: reports/mllm-en-abo-theses.json - hash: md5 - md5: 45509a75f80910070d39b864ba64bdc4 - size: 90 - - path: reports/mllm-en-jyu-theses.json - hash: md5 - md5: 0cc98fc0e98ec579ae7acd1bcc1c9cde - size: 93 - - path: reports/mllm-en-kirjaesittelyt2021.json - hash: md5 - md5: b2a3cec2dc5e54321c6365324f2f616e - size: 92 - - path: reports/mllm-en-vapaakappaleet-orig.json - hash: md5 - md5: c36acbab1e4a06d010ae7ecd8e961406 - size: 93 - eval-sv@bonsai-sv: - cmd: - - venv/bin/annif eval yso-bonsai-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-bonsai-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-bonsai-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-bonsai-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/bonsai-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - hash: md5 - md5: 8424019ec4c261915627865324ac2f73.dir - size: 66737727 - nfiles: 1878 - - path: data/projects/yso-bonsai-sv - hash: md5 - md5: 77a54001b5d18a7a474bcf7d4e9577e1.dir - size: 1029311381 - nfiles: 6 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-bonsai-sv: - name: YSO Omikuji Bonsai Swedish - language: sv - backend: omikuji - analyzer: snowball(swedish) - vocab: yso - cluster_balanced: 'False' - cluster_k: '100' - max_depth: '3' - min_df: '2' - ngram: '2' - limit: '1000' - transform: limit(5000) - access: hidden - outs: - - path: reports/bonsai-sv-abo-theses.json - hash: md5 - md5: c96c3200961b7e9e12ef94ce0a09623d - size: 92 - - path: reports/bonsai-sv-jyu-theses.json - hash: md5 - md5: 6db98665f360001f8647af363fb38ff4 - size: 91 - - path: reports/bonsai-sv-kirjaesittelyt2021.json - hash: md5 - md5: 8937047ce1e09f29306cfd2f5d1b0a68 - size: 92 - - path: reports/bonsai-sv-vapaakappaleet-orig.json - hash: md5 - md5: 1bf7c202c1be124aa26fa510900020b5 - size: 93 - eval-sv@mllm-sv: - cmd: - - venv/bin/annif eval yso-mllm-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-abo-theses.json - corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-mllm-sv -j 1 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-jyu-theses.json - corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-mllm-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-kirjaesittelyt2021.json - corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-mllm-sv -j 10 -m F1@5 -m NDCG --metrics-file reports/mllm-sv-vapaakappaleet-orig.json - corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/sv - hash: md5 - md5: 8424019ec4c261915627865324ac2f73.dir - size: 66737727 - nfiles: 1878 - - path: data/projects/yso-mllm-sv - hash: md5 - md5: 88dd945c235bdfc4549fa44f5ea582a4.dir - size: 19736546 - nfiles: 2 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-mllm-sv: - name: YSO MLLM Swedish - language: sv - backend: mllm - analyzer: snowball(swedish) - vocab: yso - limit: '1000' - transform: limit(3000000) - access: hidden - outs: - - path: reports/mllm-sv-abo-theses.json - hash: md5 - md5: 7b95bec0ae5f0a4346b5ded0c7adfab8 - size: 92 - - path: reports/mllm-sv-jyu-theses.json - hash: md5 - md5: c6f2150aebf21e4ddb8d1ebade0aefa5 - size: 91 - - path: reports/mllm-sv-kirjaesittelyt2021.json - hash: md5 - md5: ddddbc0b9097fea43c34b5133e2b24ae - size: 93 - - path: reports/mllm-sv-vapaakappaleet-orig.json - hash: md5 - md5: 7511ffde5122627168bb00855fe0e4e6 - size: 92 - eval-fi@fi: - cmd: - - venv/bin/annif eval yso-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fi-jyu-theses.json - corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fi-kirjaesittelyt2021.json - corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-fi -j 1 -m F1@5 -m NDCG --metrics-file reports/fi-kirjastonhoitaja.json - corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fi-satakunnan-kansa.json - corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-fi -j 10 -m F1@5 -m NDCG --metrics-file reports/fi-vapaakappaleet-orig.json - corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - path: corpora/fulltext-test/fi - hash: md5 - md5: 0fe8c48afc24e8c843a78fd6d76609a4.dir - size: 385776340 - nfiles: 7267 - - path: data/projects/yso-fi - hash: md5 - md5: a0961bd4dfcfee8367b22ad17d780dbb.dir - size: 1259460957 - nfiles: 3 - - path: venv-installed - hash: md5 - md5: 829dcd06bff073c054d28bb526b8b16f - size: 48 - params: - projects.toml: - yso-fi: - name: YSO suomi (2023.1.Ghosha) - language: fi - backend: nn_ensemble - sources: yso-mllm-fi:0.1492,yso-fasttext-fi:0.6090,yso-bonsai-fi:0.2418 - limit: '100' - vocab: yso - nodes: '100' - dropout_rate: '0.2' - epochs: '10' - outs: - - path: reports/fi-jyu-theses.json - hash: md5 - md5: f60303e36e729af9a6ba5fea084b440e - size: 91 - - path: reports/fi-kirjaesittelyt2021.json - hash: md5 - md5: e9ee9149f010d08d85bf4869644e25b8 - size: 94 - - path: reports/fi-kirjastonhoitaja.json - hash: md5 - md5: dec9471c35e4fef43ab484c4fd6ab0b3 - size: 91 - - path: reports/fi-satakunnan-kansa.json - hash: md5 - md5: 73c32b1eae7f182e1204f65b13022929 - size: 93 - - path: reports/fi-vapaakappaleet-orig.json - hash: md5 - md5: 50921e6a32b8d391ccfd01d55cbd23be - size: 94 diff --git a/dvc.yaml b/dvc.yaml deleted file mode 100644 index 2e0e9f2bac223577199ed52f9f77851e5c448813..0000000000000000000000000000000000000000 --- a/dvc.yaml +++ /dev/null @@ -1,181 +0,0 @@ -stages: - # Ensure Annif is installed - install: - cmd: - - python3 -m venv venv - - source venv/bin/activate && pip install -U pip wheel setuptools && pip install -r requirements.txt - - cp requirements.txt venv-installed - deps: - - requirements.txt - outs: - - venv-installed: - cache: false - # Load YSO vocabulary - load-vocab: - cmd: venv/bin/annif load-vocab --force yso corpora/yso-skos.ttl - deps: - - venv-installed - - corpora/yso-skos.ttl - outs: - - data/vocabs/yso - # Train MLLM projects - train-mllm: - foreach: - - fi - - sv - - en - do: - cmd: venv/bin/annif train yso-mllm-${item} corpora/fulltext-train/${item}/*/ -j 16 - deps: - - venv-installed - - corpora/fulltext-train/${item} - - data/vocabs/yso - params: - - projects.toml: - - yso-mllm-${item} - outs: - - data/projects/yso-mllm-${item} - # Train Omikuji Bonsai projects - train-omikuji: - foreach: - - fi - - sv - - en - do: - cmd: venv/bin/annif train yso-bonsai-${item} corpora/shorttext-train/${item}/yso-finna-${item}*.tsv.gz - deps: - - venv-installed - - corpora/shorttext-train/${item}/ - - data/vocabs/yso - params: - - projects.toml: - - yso-bonsai-${item} - outs: - - data/projects/yso-bonsai-${item} - # Train fasttext projects - train-fasttext: - foreach: - - fi - - sv - - en - do: - cmd: venv/bin/annif train yso-fasttext-${item} corpora/shorttext-train/${item}/yso-finna-${item}*.tsv.gz - deps: - - venv-installed - - corpora/shorttext-train/${item}/ - - data/vocabs/yso - params: - - projects.toml: - - yso-fasttext-${item} - outs: - - data/projects/yso-fasttext-${item} - # Train nn-ensemble projects - train-nn-ensemble: - foreach: - - fi - - sv - - en - do: - cmd: venv/bin/annif train yso-${item} corpora/fulltext-train/${item}/*/ -j 16 - deps: - - venv-installed - - corpora/fulltext-train/${item} - - data/vocabs/yso - - data/projects/yso-mllm-${item} - - data/projects/yso-bonsai-${item} - - data/projects/yso-fasttext-${item} - params: - - projects.toml: - - yso-${item} - outs: - - data/projects/yso-${item} - # Evaluate Finnish projects - eval-fi: - foreach: - - mllm-fi - - bonsai-fi - - fasttext-fi - - fi - do: - cmd: - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-jyu-theses.json corpora/fulltext-test/fi/jyu-theses/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-kirjaesittelyt2021.json corpora/fulltext-test/fi/kirjaesittelyt2021/ - - venv/bin/annif eval yso-${item} -j 1 -m F1@5 -m NDCG --metrics-file reports/${item}-kirjastonhoitaja.json corpora/fulltext-test/fi/kirjastonhoitaja/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-satakunnan-kansa.json corpora/fulltext-test/fi/satakunnan-kansa-?/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-vapaakappaleet-orig.json corpora/fulltext-test/fi/vapaakappaleet-orig/ - deps: - - venv-installed - - corpora/fulltext-test/fi - - data/projects/yso-${item} - params: - - projects.toml: - - yso-${item} - metrics: - - reports/${item}-jyu-theses.json: - cache: false - - reports/${item}-kirjaesittelyt2021.json: - cache: false - - reports/${item}-kirjastonhoitaja.json: - cache: false - - reports/${item}-satakunnan-kansa.json: - cache: false - - reports/${item}-vapaakappaleet-orig.json: - cache: false - # Evaluate Swedish projects - eval-sv: - foreach: - - mllm-sv - - bonsai-sv - - fasttext-sv - - sv - do: - cmd: - - venv/bin/annif eval yso-${item} -j 1 -m F1@5 -m NDCG --metrics-file reports/${item}-abo-theses.json corpora/fulltext-test/sv/abo-theses/ - - venv/bin/annif eval yso-${item} -j 1 -m F1@5 -m NDCG --metrics-file reports/${item}-jyu-theses.json corpora/fulltext-test/sv/jyu-theses/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-kirjaesittelyt2021.json corpora/fulltext-test/sv/kirjaesittelyt2021/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-vapaakappaleet-orig.json corpora/fulltext-test/sv/vapaakappaleet-orig/ - deps: - - venv-installed - - corpora/fulltext-test/sv - - data/projects/yso-${item} - params: - - projects.toml: - - yso-${item} - metrics: - - reports/${item}-abo-theses.json: - cache: false - - reports/${item}-jyu-theses.json: - cache: false - - reports/${item}-kirjaesittelyt2021.json: - cache: false - - reports/${item}-vapaakappaleet-orig.json: - cache: false - # Evaluate English projects - eval-en: - foreach: - - mllm-en - - bonsai-en - - fasttext-en - - en - do: - cmd: - - venv/bin/annif eval yso-${item} -j 1 -m F1@5 -m NDCG --metrics-file reports/${item}-abo-theses.json corpora/fulltext-test/en/abo-theses/ - - venv/bin/annif eval yso-${item} -j 1 -m F1@5 -m NDCG --metrics-file reports/${item}-jyu-theses.json corpora/fulltext-test/en/jyu-theses/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-kirjaesittelyt2021.json corpora/fulltext-test/en/kirjaesittelyt2021/ - - venv/bin/annif eval yso-${item} -j 10 -m F1@5 -m NDCG --metrics-file reports/${item}-vapaakappaleet-orig.json corpora/fulltext-test/en/vapaakappaleet-orig/ - deps: - - venv-installed - - corpora/fulltext-test/en - - data/projects/yso-${item} - params: - - projects.toml: - - yso-${item} - metrics: - - reports/${item}-abo-theses.json: - cache: false - - reports/${item}-jyu-theses.json: - cache: false - - reports/${item}-kirjaesittelyt2021.json: - cache: false - - reports/${item}-vapaakappaleet-orig.json: - cache: false diff --git a/projects.d/0-projects-yso.toml b/projects.d/0-projects-yso.toml deleted file mode 100644 index d22d94046c8978f5b1bc4bd5a7f84d64b88b8a8b..0000000000000000000000000000000000000000 --- a/projects.d/0-projects-yso.toml +++ /dev/null @@ -1,164 +0,0 @@ -[yso-fi] -name = "YSO suomi (2023.6.Hypatia)" -language = "fi" -backend = "nn_ensemble" -sources = "yso-mllm-fi:0.1492,yso-fasttext-fi:0.6090,yso-bonsai-fi:0.2418" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-sv] -name = "ALLFO svenska (2023.6.Hypatia)" -language = "sv" -backend = "nn_ensemble" -sources = "yso-mllm-sv:0.1439,yso-fasttext-sv:0.3302,yso-bonsai-sv:0.5259" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-en] -name = "YSO English (2023.6.Hypatia)" -language = "en" -backend = "nn_ensemble" -sources = "yso-mllm-en:0.3426,yso-fasttext-en:0.1419,yso-bonsai-en:0.5155" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-mllm-fi] -name = "YSO MLLM Finnish" -language = "fi" -backend = "mllm" -analyzer = "voikko(fi)" -vocab = "yso" -limit = "1000" -transform = "limit(3000000)" -access = "hidden" - -[yso-mllm-en] -name = "YSO MLLM English" -language = "en" -backend = "mllm" -analyzer = "snowball(english)" -vocab = "yso" -limit = "1000" -transform = "limit(2500000)" -access = "hidden" - -[yso-mllm-sv] -name = "YSO MLLM Swedish" -language = "sv" -backend = "mllm" -analyzer = "snowball(swedish)" -vocab = "yso" -limit = "1000" -transform = "limit(3000000)" -access = "hidden" - -[yso-bonsai-fi] -name = "YSO Omikuji Bonsai Finnish" -language = "fi" -backend = "omikuji" -analyzer = "snowball(finnish)" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "5" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-bonsai-sv] -name = "YSO Omikuji Bonsai Swedish" -language = "sv" -backend = "omikuji" -analyzer = "snowball(swedish)" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "2" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-bonsai-en] -name = "YSO Omikuji Bonsai English" -language = "en" -backend = "omikuji" -analyzer = "simple" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "5" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-fasttext-fi] -name = "YSO fastText Finnish" -language = "fi" -backend = "fasttext" -analyzer = "voikko(fi)" -dim = "660" -lr = "0.506539" -epoch = "75" -minn = "2" -maxn = "7" -minCount = "2" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" - -[yso-fasttext-sv] -name = "YSO fastText Swedish" -language = "sv" -backend = "fasttext" -analyzer = "snowball(swedish)" -dim = "560" -lr = "0.974349" -epoch = "110" -minn = "2" -maxn = "6" -minCount = "2" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" - -[yso-fasttext-en] -name = "YSO fastText English" -language = "en" -backend = "fasttext" -analyzer = "snowball(english)" -dim = "430" -lr = "0.506539" -epoch = "115" -minn = "4" -maxn = "5" -minCount = "1" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" diff --git a/projects.toml b/projects.toml deleted file mode 100644 index d22d94046c8978f5b1bc4bd5a7f84d64b88b8a8b..0000000000000000000000000000000000000000 --- a/projects.toml +++ /dev/null @@ -1,164 +0,0 @@ -[yso-fi] -name = "YSO suomi (2023.6.Hypatia)" -language = "fi" -backend = "nn_ensemble" -sources = "yso-mllm-fi:0.1492,yso-fasttext-fi:0.6090,yso-bonsai-fi:0.2418" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-sv] -name = "ALLFO svenska (2023.6.Hypatia)" -language = "sv" -backend = "nn_ensemble" -sources = "yso-mllm-sv:0.1439,yso-fasttext-sv:0.3302,yso-bonsai-sv:0.5259" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-en] -name = "YSO English (2023.6.Hypatia)" -language = "en" -backend = "nn_ensemble" -sources = "yso-mllm-en:0.3426,yso-fasttext-en:0.1419,yso-bonsai-en:0.5155" -limit = "100" -vocab = "yso" -nodes = "100" -dropout_rate = "0.2" -epochs = "10" - -[yso-mllm-fi] -name = "YSO MLLM Finnish" -language = "fi" -backend = "mllm" -analyzer = "voikko(fi)" -vocab = "yso" -limit = "1000" -transform = "limit(3000000)" -access = "hidden" - -[yso-mllm-en] -name = "YSO MLLM English" -language = "en" -backend = "mllm" -analyzer = "snowball(english)" -vocab = "yso" -limit = "1000" -transform = "limit(2500000)" -access = "hidden" - -[yso-mllm-sv] -name = "YSO MLLM Swedish" -language = "sv" -backend = "mllm" -analyzer = "snowball(swedish)" -vocab = "yso" -limit = "1000" -transform = "limit(3000000)" -access = "hidden" - -[yso-bonsai-fi] -name = "YSO Omikuji Bonsai Finnish" -language = "fi" -backend = "omikuji" -analyzer = "snowball(finnish)" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "5" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-bonsai-sv] -name = "YSO Omikuji Bonsai Swedish" -language = "sv" -backend = "omikuji" -analyzer = "snowball(swedish)" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "2" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-bonsai-en] -name = "YSO Omikuji Bonsai English" -language = "en" -backend = "omikuji" -analyzer = "simple" -vocab = "yso" -cluster_balanced = "False" -cluster_k = "100" -max_depth = "3" -min_df = "5" -ngram = "2" -limit = "1000" -transform = "limit(5000)" -access = "hidden" - -[yso-fasttext-fi] -name = "YSO fastText Finnish" -language = "fi" -backend = "fasttext" -analyzer = "voikko(fi)" -dim = "660" -lr = "0.506539" -epoch = "75" -minn = "2" -maxn = "7" -minCount = "2" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" - -[yso-fasttext-sv] -name = "YSO fastText Swedish" -language = "sv" -backend = "fasttext" -analyzer = "snowball(swedish)" -dim = "560" -lr = "0.974349" -epoch = "110" -minn = "2" -maxn = "6" -minCount = "2" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" - -[yso-fasttext-en] -name = "YSO fastText English" -language = "en" -backend = "fasttext" -analyzer = "snowball(english)" -dim = "430" -lr = "0.506539" -epoch = "115" -minn = "4" -maxn = "5" -minCount = "1" -wordNgrams = "2" -loss = "hs" -limit = "1000" -chunksize = "24" -vocab = "yso" -transform = "limit(15000),filter_lang,limit(5000)" -access = "hidden" diff --git a/projects/yso-mllm-en.zip b/projects/yso-mllm-en.zip deleted file mode 100644 index 4585cebe24bc10c459286bcacdec91684029eca0..0000000000000000000000000000000000000000 --- a/projects/yso-mllm-en.zip +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:de4a59f7620d455eb4c95ab09f5ba9a20d2f0887aad9cbe95544c9d48a34ca07 -size 2114994 diff --git a/reports/bonsai-en-abo-theses.json b/reports/bonsai-en-abo-theses.json deleted file mode 100644 index 68d2159b16f1e837ab507ebcac67932dbba32bff..0000000000000000000000000000000000000000 --- a/reports/bonsai-en-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3054043183401105, - "NDCG": 0.4286120533943176, - "Documents_evaluated": 69 -} \ No newline at end of file diff --git a/reports/bonsai-en-jyu-theses.json b/reports/bonsai-en-jyu-theses.json deleted file mode 100644 index a09d7894915fd5f0c5ed46b47bb12802fb463141..0000000000000000000000000000000000000000 --- a/reports/bonsai-en-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.5280977655960271, - "NDCG": 0.7093563675880432, - "Documents_evaluated": 301 -} \ No newline at end of file diff --git a/reports/bonsai-en-kirjaesittelyt2021.json b/reports/bonsai-en-kirjaesittelyt2021.json deleted file mode 100644 index fe89722b901bcd3b98430f3fe5d311acb7212f12..0000000000000000000000000000000000000000 --- a/reports/bonsai-en-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4702196887073211, - "NDCG": 0.6450860500335693, - "Documents_evaluated": 447 -} \ No newline at end of file diff --git a/reports/bonsai-en-vapaakappaleet-orig.json b/reports/bonsai-en-vapaakappaleet-orig.json deleted file mode 100644 index 0db1a623914022dcb3c4781df9fb31d54de16950..0000000000000000000000000000000000000000 --- a/reports/bonsai-en-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.452540523877909, - "NDCG": 0.6115207076072693, - "Documents_evaluated": 1048 -} \ No newline at end of file diff --git a/reports/bonsai-fi-jyu-theses.json b/reports/bonsai-fi-jyu-theses.json deleted file mode 100644 index 31df1332e0c009973d2edf62b55d6a6eeece3917..0000000000000000000000000000000000000000 --- a/reports/bonsai-fi-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.47928490438998533, - "NDCG": 0.6288755536079407, - "Documents_evaluated": 766 -} \ No newline at end of file diff --git a/reports/bonsai-fi-kirjaesittelyt2021.json b/reports/bonsai-fi-kirjaesittelyt2021.json deleted file mode 100644 index 2e398273c2a841c0836471c33693381b939ec39f..0000000000000000000000000000000000000000 --- a/reports/bonsai-fi-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.46373377236325175, - "NDCG": 0.6373720169067383, - "Documents_evaluated": 1311 -} \ No newline at end of file diff --git a/reports/bonsai-fi-kirjastonhoitaja.json b/reports/bonsai-fi-kirjastonhoitaja.json deleted file mode 100644 index aa648ccf2be2b4d385d7bcf7c4c05ee5f2623483..0000000000000000000000000000000000000000 --- a/reports/bonsai-fi-kirjastonhoitaja.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2806648479725403, - "NDCG": 0.3858826458454132, - "Documents_evaluated": 312 -} \ No newline at end of file diff --git a/reports/bonsai-fi-satakunnan-kansa.json b/reports/bonsai-fi-satakunnan-kansa.json deleted file mode 100644 index 2bb93ff38dc24681b2f4f962a691ea67992be1a0..0000000000000000000000000000000000000000 --- a/reports/bonsai-fi-satakunnan-kansa.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2727621316591905, - "NDCG": 0.35492458939552307, - "Documents_evaluated": 200 -} \ No newline at end of file diff --git a/reports/bonsai-fi-vapaakappaleet-orig.json b/reports/bonsai-fi-vapaakappaleet-orig.json deleted file mode 100644 index 088fde84b9f5244c87cb00a2d763cdd399af68b5..0000000000000000000000000000000000000000 --- a/reports/bonsai-fi-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4591248628471061, - "NDCG": 0.6167663335800171, - "Documents_evaluated": 1040 -} \ No newline at end of file diff --git a/reports/bonsai-sv-abo-theses.json b/reports/bonsai-sv-abo-theses.json deleted file mode 100644 index 17baf6a9287ed9e1f7c89830933d38bd0a23b1dc..0000000000000000000000000000000000000000 --- a/reports/bonsai-sv-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3039283938401189, - "NDCG": 0.4083988070487976, - "Documents_evaluated": 107 -} \ No newline at end of file diff --git a/reports/bonsai-sv-jyu-theses.json b/reports/bonsai-sv-jyu-theses.json deleted file mode 100644 index b17c4c2c56e3a6273f08624ee6ce7ee36ed85a92..0000000000000000000000000000000000000000 --- a/reports/bonsai-sv-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.5114636231823732, - "NDCG": 0.6833634972572327, - "Documents_evaluated": 32 -} \ No newline at end of file diff --git a/reports/bonsai-sv-kirjaesittelyt2021.json b/reports/bonsai-sv-kirjaesittelyt2021.json deleted file mode 100644 index 9ee79190e4060843e0bb59a4afa78b55b47492d6..0000000000000000000000000000000000000000 --- a/reports/bonsai-sv-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4658328182599977, - "NDCG": 0.6352542042732239, - "Documents_evaluated": 560 -} \ No newline at end of file diff --git a/reports/bonsai-sv-vapaakappaleet-orig.json b/reports/bonsai-sv-vapaakappaleet-orig.json deleted file mode 100644 index bf25c7f7650a493f9ef1a520f961bea8bbaea623..0000000000000000000000000000000000000000 --- a/reports/bonsai-sv-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.49705832364125313, - "NDCG": 0.6688342094421387, - "Documents_evaluated": 176 -} \ No newline at end of file diff --git a/reports/en-abo-theses.json b/reports/en-abo-theses.json deleted file mode 100644 index 08df9dcecbb4af72bf5af295e7832875f804e468..0000000000000000000000000000000000000000 --- a/reports/en-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3586282711669708, - "NDCG": 0.4873850643634796, - "Documents_evaluated": 69 -} \ No newline at end of file diff --git a/reports/en-jyu-theses.json b/reports/en-jyu-theses.json deleted file mode 100644 index dc3e76ece0ef1e6db639d2e2f645a5052bd9d8f1..0000000000000000000000000000000000000000 --- a/reports/en-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.5319231423955479, - "NDCG": 0.7081617116928101, - "Documents_evaluated": 301 -} \ No newline at end of file diff --git a/reports/en-kirjaesittelyt2021.json b/reports/en-kirjaesittelyt2021.json deleted file mode 100644 index 77d5ae254868e9a9ad4984c91bbe3748c1211ea2..0000000000000000000000000000000000000000 --- a/reports/en-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.46394802162537224, - "NDCG": 0.635046660900116, - "Documents_evaluated": 447 -} \ No newline at end of file diff --git a/reports/en-vapaakappaleet-orig.json b/reports/en-vapaakappaleet-orig.json deleted file mode 100644 index a51df2e8bb970229614b303893f7e34ed7406e0f..0000000000000000000000000000000000000000 --- a/reports/en-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.47304038133646015, - "NDCG": 0.6267992258071899, - "Documents_evaluated": 1048 -} \ No newline at end of file diff --git a/reports/fasttext-en-abo-theses.json b/reports/fasttext-en-abo-theses.json deleted file mode 100644 index 9227c70082e1118879654ac9300cb4c3e451e483..0000000000000000000000000000000000000000 --- a/reports/fasttext-en-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.20503141048638424, - "NDCG": 0.3023087680339813, - "Documents_evaluated": 69 -} \ No newline at end of file diff --git a/reports/fasttext-en-jyu-theses.json b/reports/fasttext-en-jyu-theses.json deleted file mode 100644 index fa44254c4fb60e47f3f03392259fc2ec86c0976f..0000000000000000000000000000000000000000 --- a/reports/fasttext-en-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.25546128859670025, - "NDCG": 0.3536200523376465, - "Documents_evaluated": 301 -} \ No newline at end of file diff --git a/reports/fasttext-en-kirjaesittelyt2021.json b/reports/fasttext-en-kirjaesittelyt2021.json deleted file mode 100644 index 68ed6f55f483b3dedb8d28df3e3f4f099d2e3dfe..0000000000000000000000000000000000000000 --- a/reports/fasttext-en-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.32929289961175834, - "NDCG": 0.4719170033931732, - "Documents_evaluated": 447 -} \ No newline at end of file diff --git a/reports/fasttext-en-vapaakappaleet-orig.json b/reports/fasttext-en-vapaakappaleet-orig.json deleted file mode 100644 index 12ee4d44abe55296c5ecdfe9f5f078535d9d66da..0000000000000000000000000000000000000000 --- a/reports/fasttext-en-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2840493228074519, - "NDCG": 0.3941110074520111, - "Documents_evaluated": 1048 -} \ No newline at end of file diff --git a/reports/fasttext-fi-jyu-theses.json b/reports/fasttext-fi-jyu-theses.json deleted file mode 100644 index 13b75da2d88d3fd808f0c42da92bb183ba8a3ab1..0000000000000000000000000000000000000000 --- a/reports/fasttext-fi-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.30858265219863545, - "NDCG": 0.41998833417892456, - "Documents_evaluated": 766 -} \ No newline at end of file diff --git a/reports/fasttext-fi-kirjaesittelyt2021.json b/reports/fasttext-fi-kirjaesittelyt2021.json deleted file mode 100644 index 379d0ddbbeb4a80dedc9192a83e3ccaa26976415..0000000000000000000000000000000000000000 --- a/reports/fasttext-fi-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3718261359386204, - "NDCG": 0.5144786834716797, - "Documents_evaluated": 1311 -} \ No newline at end of file diff --git a/reports/fasttext-fi-kirjastonhoitaja.json b/reports/fasttext-fi-kirjastonhoitaja.json deleted file mode 100644 index c4e295c5a2f9194a9b6b1cab382bd89be24d946b..0000000000000000000000000000000000000000 --- a/reports/fasttext-fi-kirjastonhoitaja.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2545131506669968, - "NDCG": 0.3733712136745453, - "Documents_evaluated": 312 -} \ No newline at end of file diff --git a/reports/fasttext-fi-satakunnan-kansa.json b/reports/fasttext-fi-satakunnan-kansa.json deleted file mode 100644 index 01f9c1ffdb54d2ebd65b9bbb50168b7ef334cde5..0000000000000000000000000000000000000000 --- a/reports/fasttext-fi-satakunnan-kansa.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.22725808994926644, - "NDCG": 0.32201603055000305, - "Documents_evaluated": 200 -} \ No newline at end of file diff --git a/reports/fasttext-fi-vapaakappaleet-orig.json b/reports/fasttext-fi-vapaakappaleet-orig.json deleted file mode 100644 index f79b951868f5dcf72541d647ade7daa1d720013b..0000000000000000000000000000000000000000 --- a/reports/fasttext-fi-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3193791850958433, - "NDCG": 0.4337014853954315, - "Documents_evaluated": 1040 -} \ No newline at end of file diff --git a/reports/fasttext-sv-abo-theses.json b/reports/fasttext-sv-abo-theses.json deleted file mode 100644 index 09a2cb30999e427e1242ca9df79855b6da5c9a63..0000000000000000000000000000000000000000 --- a/reports/fasttext-sv-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.22664895686361827, - "NDCG": 0.31691738963127136, - "Documents_evaluated": 107 -} \ No newline at end of file diff --git a/reports/fasttext-sv-jyu-theses.json b/reports/fasttext-sv-jyu-theses.json deleted file mode 100644 index 3a994ccc03147781ec93cbc21b33d05cc25849e5..0000000000000000000000000000000000000000 --- a/reports/fasttext-sv-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3478615134865135, - "NDCG": 0.4654184579849243, - "Documents_evaluated": 32 -} \ No newline at end of file diff --git a/reports/fasttext-sv-kirjaesittelyt2021.json b/reports/fasttext-sv-kirjaesittelyt2021.json deleted file mode 100644 index 8acee5db78c0e21a7ae1dcc1ea616a93af6d05ea..0000000000000000000000000000000000000000 --- a/reports/fasttext-sv-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3472656438292162, - "NDCG": 0.48715171217918396, - "Documents_evaluated": 560 -} \ No newline at end of file diff --git a/reports/fasttext-sv-vapaakappaleet-orig.json b/reports/fasttext-sv-vapaakappaleet-orig.json deleted file mode 100644 index adc8830a70a06b4fe798c45d53fa37e2c855d240..0000000000000000000000000000000000000000 --- a/reports/fasttext-sv-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.32061171839968006, - "NDCG": 0.42769214510917664, - "Documents_evaluated": 176 -} \ No newline at end of file diff --git a/reports/fi-jyu-theses.json b/reports/fi-jyu-theses.json deleted file mode 100644 index 7651588627ffc153e6abfaad70efc4cfbe041e61..0000000000000000000000000000000000000000 --- a/reports/fi-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.521254215931539, - "NDCG": 0.6806246042251587, - "Documents_evaluated": 766 -} \ No newline at end of file diff --git a/reports/fi-kirjaesittelyt2021.json b/reports/fi-kirjaesittelyt2021.json deleted file mode 100644 index 08d67407f5baea8d68f059512be337bc491b32d5..0000000000000000000000000000000000000000 --- a/reports/fi-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.47211753194689804, - "NDCG": 0.6451713442802429, - "Documents_evaluated": 1311 -} \ No newline at end of file diff --git a/reports/fi-kirjastonhoitaja.json b/reports/fi-kirjastonhoitaja.json deleted file mode 100644 index e8292f1d265680c05c6e2ae630d3142711c2c007..0000000000000000000000000000000000000000 --- a/reports/fi-kirjastonhoitaja.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4092650938804785, - "NDCG": 0.548991858959198, - "Documents_evaluated": 312 -} \ No newline at end of file diff --git a/reports/fi-satakunnan-kansa.json b/reports/fi-satakunnan-kansa.json deleted file mode 100644 index d8e948a7d42db68457b43032d9226d1f08bb2f96..0000000000000000000000000000000000000000 --- a/reports/fi-satakunnan-kansa.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.30209318785789374, - "NDCG": 0.4054855406284332, - "Documents_evaluated": 200 -} \ No newline at end of file diff --git a/reports/fi-vapaakappaleet-orig.json b/reports/fi-vapaakappaleet-orig.json deleted file mode 100644 index 5fea03a822ac6891cc01f0b6aa7df92c8def5a2c..0000000000000000000000000000000000000000 --- a/reports/fi-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.49103690820462104, - "NDCG": 0.6527368426322937, - "Documents_evaluated": 1040 -} \ No newline at end of file diff --git a/reports/metrics.tsv b/reports/metrics.tsv deleted file mode 100644 index 438b53581ed9a04be381d22260ffeb5e8eb2d778..0000000000000000000000000000000000000000 --- a/reports/metrics.tsv +++ /dev/null @@ -1,15 +0,0 @@ -lang test_set documents bonsai fasttext mllm nn -en abo-theses 63 0.3071 0.19563 0.33812 0.35514 -en jyu-theses 301 0.53481 0.25072 0.30314 0.52716 -en kirjaesittelyt2021 447 0.47426 0.33568 0.20642 0.45892 -en vapaakappaleet-orig 1048 0.45809 0.27554 0.25672 0.47289 -fi jyu-theses 766 0.47584 0.30561 0.43745 0.5175 -fi kirjaesittelyt2021 1311 0.463 0.37159 0.22769 0.47216 -fi kirjastonhoitaja 312 0.27809 0.26318 0.32339 0.40305 -fi satakunnan-kansa 200 0.28965 0.22578 0.2238 0.30408 -fi vapaakappaleet-orig 1040 0.4637 0.31516 0.37471 0.49595 -sv abo-theses 105 0.29877 0.22494 0.37944 0.38866 -sv jyu-theses 32 0.51605 0.34188 0.39801 0.56257 -sv kirjaesittelyt2021 560 0.46757 0.35114 0.19829 0.49215 -sv vapaakappaleet-orig 176 0.48741 0.32585 0.27866 0.51249 - diff --git a/reports/mllm-en-abo-theses.json b/reports/mllm-en-abo-theses.json deleted file mode 100644 index ca154aa0ab91c610c6b397286bd4cbe919c9bc9b..0000000000000000000000000000000000000000 --- a/reports/mllm-en-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.343417593185395, - "NDCG": 0.4538632929325104, - "Documents_evaluated": 69 -} \ No newline at end of file diff --git a/reports/mllm-en-jyu-theses.json b/reports/mllm-en-jyu-theses.json deleted file mode 100644 index 8c516775ca7407e1c12efed8d5bf43cb636da117..0000000000000000000000000000000000000000 --- a/reports/mllm-en-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3049068815501052, - "NDCG": 0.42262372374534607, - "Documents_evaluated": 301 -} \ No newline at end of file diff --git a/reports/mllm-en-kirjaesittelyt2021.json b/reports/mllm-en-kirjaesittelyt2021.json deleted file mode 100644 index af9a4343384d9b1961ae208918bd7b25709f74b1..0000000000000000000000000000000000000000 --- a/reports/mllm-en-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2019899981196284, - "NDCG": 0.2837369441986084, - "Documents_evaluated": 447 -} \ No newline at end of file diff --git a/reports/mllm-en-vapaakappaleet-orig.json b/reports/mllm-en-vapaakappaleet-orig.json deleted file mode 100644 index 86a3bb5dfb3b67e7e56c3e389a2aee72de3600a4..0000000000000000000000000000000000000000 --- a/reports/mllm-en-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2567412744702019, - "NDCG": 0.3595125675201416, - "Documents_evaluated": 1048 -} \ No newline at end of file diff --git a/reports/mllm-fi-jyu-theses.json b/reports/mllm-fi-jyu-theses.json deleted file mode 100644 index cfb9b8232e4da8bef53f71fec1826c07676f57c3..0000000000000000000000000000000000000000 --- a/reports/mllm-fi-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4430391050234972, - "NDCG": 0.5812211632728577, - "Documents_evaluated": 766 -} \ No newline at end of file diff --git a/reports/mllm-fi-kirjaesittelyt2021.json b/reports/mllm-fi-kirjaesittelyt2021.json deleted file mode 100644 index 9ab91811dc96af31ef26ce97eb88dfe595278e7e..0000000000000000000000000000000000000000 --- a/reports/mllm-fi-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.22742922347554473, - "NDCG": 0.3150874972343445, - "Documents_evaluated": 1311 -} \ No newline at end of file diff --git a/reports/mllm-fi-kirjastonhoitaja.json b/reports/mllm-fi-kirjastonhoitaja.json deleted file mode 100644 index 089b0b2896a624aa772be098416b8dc60ed080f1..0000000000000000000000000000000000000000 --- a/reports/mllm-fi-kirjastonhoitaja.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.33001418097571944, - "NDCG": 0.44591477513313293, - "Documents_evaluated": 312 -} \ No newline at end of file diff --git a/reports/mllm-fi-satakunnan-kansa.json b/reports/mllm-fi-satakunnan-kansa.json deleted file mode 100644 index 1ce470047f6adf8465bed7da932ad74b023876ff..0000000000000000000000000000000000000000 --- a/reports/mllm-fi-satakunnan-kansa.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2138675621763857, - "NDCG": 0.28605687618255615, - "Documents_evaluated": 200 -} \ No newline at end of file diff --git a/reports/mllm-fi-vapaakappaleet-orig.json b/reports/mllm-fi-vapaakappaleet-orig.json deleted file mode 100644 index ce4df82e23777cddf9b66cc19d52e3d04af0c3b4..0000000000000000000000000000000000000000 --- a/reports/mllm-fi-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3680668274891286, - "NDCG": 0.4940447211265564, - "Documents_evaluated": 1040 -} \ No newline at end of file diff --git a/reports/mllm-sv-abo-theses.json b/reports/mllm-sv-abo-theses.json deleted file mode 100644 index b1bdb542aa1c5cfa1f0a69f2d10397762275e617..0000000000000000000000000000000000000000 --- a/reports/mllm-sv-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3747733912276001, - "NDCG": 0.5160552859306335, - "Documents_evaluated": 107 -} \ No newline at end of file diff --git a/reports/mllm-sv-jyu-theses.json b/reports/mllm-sv-jyu-theses.json deleted file mode 100644 index d38034c280e9a7ca68dfb25ba3a47e243938a87d..0000000000000000000000000000000000000000 --- a/reports/mllm-sv-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.40063929126429126, - "NDCG": 0.562961220741272, - "Documents_evaluated": 32 -} \ No newline at end of file diff --git a/reports/mllm-sv-kirjaesittelyt2021.json b/reports/mllm-sv-kirjaesittelyt2021.json deleted file mode 100644 index 500463b1b2ff3261699c6ef17fbbab7507864d74..0000000000000000000000000000000000000000 --- a/reports/mllm-sv-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.20049768837489312, - "NDCG": 0.2875151038169861, - "Documents_evaluated": 560 -} \ No newline at end of file diff --git a/reports/mllm-sv-vapaakappaleet-orig.json b/reports/mllm-sv-vapaakappaleet-orig.json deleted file mode 100644 index 90b3dc2bca8f1c0be639ba0d204b62dd86451440..0000000000000000000000000000000000000000 --- a/reports/mllm-sv-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.2869693440553272, - "NDCG": 0.3894149959087372, - "Documents_evaluated": 176 -} \ No newline at end of file diff --git a/reports/sv-abo-theses.json b/reports/sv-abo-theses.json deleted file mode 100644 index b5c893dd2209c95b05916f90f7973bd6acf4b664..0000000000000000000000000000000000000000 --- a/reports/sv-abo-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.3815695135373134, - "NDCG": 0.5198650360107422, - "Documents_evaluated": 107 -} \ No newline at end of file diff --git a/reports/sv-jyu-theses.json b/reports/sv-jyu-theses.json deleted file mode 100644 index 1adcac9de72db93768d2bffffa16256928edcee1..0000000000000000000000000000000000000000 --- a/reports/sv-jyu-theses.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.5470708111333111, - "NDCG": 0.7341756820678711, - "Documents_evaluated": 32 -} \ No newline at end of file diff --git a/reports/sv-kirjaesittelyt2021.json b/reports/sv-kirjaesittelyt2021.json deleted file mode 100644 index a14df1f0ee27493360e622be4e30d0fe46fb7b9d..0000000000000000000000000000000000000000 --- a/reports/sv-kirjaesittelyt2021.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.4858890352163123, - "NDCG": 0.6765456199645996, - "Documents_evaluated": 560 -} \ No newline at end of file diff --git a/reports/sv-vapaakappaleet-orig.json b/reports/sv-vapaakappaleet-orig.json deleted file mode 100644 index 2a43afc3d3aa188e76f91a6351c916a2cdbf877a..0000000000000000000000000000000000000000 --- a/reports/sv-vapaakappaleet-orig.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "F1@5": 0.5059074567785757, - "NDCG": 0.6866861581802368, - "Documents_evaluated": 176 -} \ No newline at end of file diff --git a/repository-metrics-analysis/analyse-theseus-tietolinja.ipynb b/repository-metrics-analysis/analyse-theseus-tietolinja.ipynb deleted file mode 100644 index e20e3c9347d22714c0bc2957856fd415db013e4e..0000000000000000000000000000000000000000 --- a/repository-metrics-analysis/analyse-theseus-tietolinja.ipynb +++ /dev/null @@ -1,2263 +0,0 @@ -{ - "cells": [ - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Analysis of Annif suggestions in Theseus repository \n", - "\n", - "## Install and import packages\n", - "In VS Code the simplest way to install packages in correct venv is to do it in the notebook." - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Requirement already satisfied: pip in ./.venv/lib/python3.8/site-packages (23.1)\n", - "Collecting pip\n", - " Using cached pip-23.1.2-py3-none-any.whl (2.1 MB)\n", - "Requirement already satisfied: setuptools in ./.venv/lib/python3.8/site-packages (67.6.1)\n", - "Collecting setuptools\n", - " Downloading setuptools-68.0.0-py3-none-any.whl (804 kB)\n", - "\u001b[2K \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m804.0/804.0 kB\u001b[0m \u001b[31m9.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mm eta \u001b[36m0:00:01\u001b[0m0:01\u001b[0m\n", - "\u001b[?25hRequirement already satisfied: wheel in ./.venv/lib/python3.8/site-packages (0.40.0)\n", - "Installing collected packages: setuptools, pip\n", - " Attempting uninstall: setuptools\n", - " Found existing installation: setuptools 67.6.1\n", - " Uninstalling setuptools-67.6.1:\n", - " Successfully uninstalled setuptools-67.6.1\n", - " Attempting uninstall: pip\n", - " Found existing installation: pip 23.1\n", - " Uninstalling pip-23.1:\n", - " Successfully uninstalled pip-23.1\n", - "Successfully installed pip-23.1.2 setuptools-68.0.0\n", - "Requirement already satisfied: ndjson in ./.venv/lib/python3.8/site-packages (0.3.1)\n", - "Requirement already satisfied: pandas in ./.venv/lib/python3.8/site-packages (1.5.3)\n", - "Requirement already satisfied: rdflib in ./.venv/lib/python3.8/site-packages (6.3.1)\n", - "Requirement already satisfied: matplotlib in ./.venv/lib/python3.8/site-packages (3.7.1)\n", - "Requirement already satisfied: jinja2 in ./.venv/lib/python3.8/site-packages (3.1.2)\n", - "Requirement already satisfied: python-dateutil>=2.8.1 in ./.venv/lib/python3.8/site-packages (from pandas) (2.8.2)\n", - "Requirement already satisfied: pytz>=2020.1 in ./.venv/lib/python3.8/site-packages (from pandas) (2022.7.1)\n", - "Requirement already satisfied: numpy>=1.20.3 in ./.venv/lib/python3.8/site-packages (from pandas) (1.24.2)\n", - "Requirement already satisfied: isodate<0.7.0,>=0.6.0 in ./.venv/lib/python3.8/site-packages (from rdflib) (0.6.1)\n", - "Requirement already satisfied: pyparsing<4,>=2.1.0 in ./.venv/lib/python3.8/site-packages (from rdflib) (3.0.9)\n", - "Requirement already satisfied: contourpy>=1.0.1 in ./.venv/lib/python3.8/site-packages (from matplotlib) (1.0.7)\n", - "Requirement already satisfied: cycler>=0.10 in ./.venv/lib/python3.8/site-packages (from matplotlib) (0.11.0)\n", - "Requirement already satisfied: fonttools>=4.22.0 in ./.venv/lib/python3.8/site-packages (from matplotlib) (4.39.2)\n", - "Requirement already satisfied: kiwisolver>=1.0.1 in ./.venv/lib/python3.8/site-packages (from matplotlib) (1.4.4)\n", - "Requirement already satisfied: packaging>=20.0 in ./.venv/lib/python3.8/site-packages (from matplotlib) (23.0)\n", - "Requirement already satisfied: pillow>=6.2.0 in ./.venv/lib/python3.8/site-packages (from matplotlib) (9.4.0)\n", - "Requirement already satisfied: importlib-resources>=3.2.0 in ./.venv/lib/python3.8/site-packages (from matplotlib) (5.12.0)\n", - "Requirement already satisfied: MarkupSafe>=2.0 in ./.venv/lib/python3.8/site-packages (from jinja2) (2.1.2)\n", - "Requirement already satisfied: zipp>=3.1.0 in ./.venv/lib/python3.8/site-packages (from importlib-resources>=3.2.0->matplotlib) (3.15.0)\n", - "Requirement already satisfied: six in ./.venv/lib/python3.8/site-packages (from isodate<0.7.0,>=0.6.0->rdflib) (1.16.0)\n" - ] - } - ], - "source": [ - "! pip install --upgrade pip setuptools wheel\n", - "! pip install ndjson pandas rdflib matplotlib jinja2\n", - "\n", - "import ndjson\n", - "import pandas as pd\n", - "from rdflib import Graph, Literal, Namespace\n", - "from rdflib.namespace import OWL, SKOS\n", - "import unicodedata\n", - "import matplotlib.pyplot as plt\n", - "import matplotlib\n", - "import datetime" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Failed to convert Literal lexical form to value. Datatype=http://www.w3.org/2001/XMLSchema#date, Converter=\n", - "Traceback (most recent call last):\n", - " File \"/home/local/jmminkin/git/FintoAI-data-YSO/repository-metrics-analysis/.venv/lib/python3.8/site-packages/rdflib/term.py\", line 2084, in _castLexicalToPython\n", - " return conv_func(lexical) # type: ignore[arg-type]\n", - " File \"/home/local/jmminkin/git/FintoAI-data-YSO/repository-metrics-analysis/.venv/lib/python3.8/site-packages/isodate/isodates.py\", line 201, in parse_date\n", - " return date(sign * int(groups['year']),\n", - "ValueError: month must be in 1..12\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Number of triples in YSA: 433048\n", - "Number of triples in Allärs: 412097\n" - ] - } - ], - "source": [ - "# load current YSA\n", - "ysa = Graph()\n", - "ysa.parse('http://finto.fi/rest/v1/ysa/data?format=text/turtle')\n", - "print('Number of triples in YSA: ', len(ysa))\n", - "\n", - "# load current Allärs\n", - "allars = Graph()\n", - "allars.parse('http://finto.fi/rest/v1/allars/data?format=text/turtle')\n", - "print('Number of triples in Allärs: ', len(allars))\n" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Number of triples in YSO+YSO Places: 830887\n" - ] - } - ], - "source": [ - "# load YSO and YSO Places\n", - "yso = Graph()\n", - "yso.parse(\"../Annif-corpora/vocab/yso-skos.ttl\", format='turtle')\n", - "print('Number of triples in YSO+YSO Places: ', len(yso))" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Helper functions" - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": {}, - "outputs": [], - "source": [ - "YSO = Namespace('http://www.yso.fi/onto/yso/')\n", - "COMPLAIN = False # whether to complain about unknown labels\n", - "\n", - "\n", - "def is_deprecated(ysouri):\n", - " return (ysouri, OWL.deprecated, True) in yso\n", - "\n", - "\n", - "def label_to_yso_uris(label, voc, lang, complain=COMPLAIN):\n", - " # based on https://github.com/NatLibFi/Annif-corpora/blob/6521d0357f3d93019f6d2838b960e80d9131735c/tools/finna-metadata-collect-scripts/create-corpus.py#L54\n", - " \n", - " # Remove trailing \".\" present in labels of some records\n", - " try:\n", - " value = Literal(unicodedata.normalize('NFC', label.rstrip('.')), lang)\n", - " except Exception as e:\n", - " print(\"Error normalizing label '{}'\".format(label))\n", - " return []\n", - "\n", - " for prop in (SKOS.prefLabel, SKOS.altLabel):\n", - " vocuri = voc.value(None, prop, value, any=True)\n", - " if vocuri is not None:\n", - " if vocuri.startswith(YSO):\n", - " return [vocuri]\n", - " for matchprop in (SKOS.exactMatch, SKOS.closeMatch):\n", - " matches = [match for match in voc.objects(vocuri, matchprop)\n", - " if match.startswith(YSO)]\n", - " if matches:\n", - " return matches\n", - "\n", - " # hackish fallbacks for cases like \"kulttuuri\", where YSO Cicero is out of\n", - " # date: look up via ysa/allars\n", - " if lang == 'fi' and voc == yso:\n", - " matches = label_to_yso_uris(label, ysa, lang)\n", - " if matches:\n", - " #print(\"missing fi label '{}' found via ysa\".format(label))\n", - " return matches\n", - "\n", - " if lang == 'sv' and voc == yso:\n", - " matches = label_to_yso_uris(label, allars, lang)\n", - " if matches:\n", - " #print(\"missing sv label '{}' found via allars\".format(label))\n", - " return matches\n", - "\n", - " if complain:\n", - " print(\"Unknown label '{}'\".format(label))\n", - " return []\n", - "\n", - "\n", - "def add_subjects_uris(df, lang, source_col):\n", - " df[\"subjects_uris\"] = df[source_col].apply(lambda x: [label_to_yso_uris(label, yso, lang, COMPLAIN) for label in x])\n", - " df[\"subjects_uris\"] = df[\"subjects_uris\"].apply(lambda x: [str(item) for sublist in x for item in sublist])\n", - " return df\n", - "\n", - "\n", - "def calc_scores(df):\n", - " # calculate precision, recall, f1\n", - " df[\"precision\"] = df.apply(lambda x: len(set(x.subjects_uris) & set(x.suggestions)) / len(x.suggestions) if len(x.suggestions) > 0 else pd.np.nan, axis=1)\n", - " df[\"recall\"] = df.apply(lambda x: len(set(x.subjects_uris) & set(x.suggestions)) / len(x.subjects_uris) if len(x.subjects_uris) > 0 else pd.np.nan, axis=1)\n", - " df[\"f1 score\"] = df.apply(lambda x: 2 * x.precision * x.recall / (x.precision + x.recall) if (x.precision + x.recall) != 0 else 0, axis=1)\n", - " df['f1 score'] = pd.to_numeric(df['f1 score'])\n", - "\n", - "\n", - "def plot_scores(df):\n", - " calc_scores(df)\n", - "\n", - " # mark annif_updates in the plot\n", - " for upd in annif_updates:\n", - " plt.axvline(upd, color='red', linestyle='--')\n", - "\n", - " # plot precision, recall, f1 as a function of time. Label with month start\n", - " df_monthly = df.set_index(\"date_accessioned\").resample('MS').mean()\n", - " df_monthly.plot(marker='o', ax=plt.gca())\n", - " # labels = [x.strftime(\"%Y-%m\") for x in df_monthly.index]\n", - " # plt.gca().set_xticklabels(labels, rotation=45)\n", - " # plt.gca().xaxis.set_major_locator(plt.MaxNLocator(10))\n", - " plt.grid()\n", - "\n", - "\n", - "def plot_counts(df):\n", - " # How many records there are with suggestions each month? Label with month start\n", - " df_monthly = df.set_index('date_accessioned').resample(\"MS\").id.count()\n", - " # df_monthly.plot.bar()\n", - " plt.bar(df_monthly.index, df_monthly, width=20)\n", - " # plt.ylabel(\"Number of records\")\n", - " plt.grid()\n", - "\n", - "\n", - "def get_number_of_rows(df):\n", - " print(\"rows: \", len(df))\n", - " print(\"rows with subjects_yso: \", len(df[df.subjects_yso.apply(lambda x: len(x) > 0)]))\n", - " print(\"rows with subjects_all: \", len(df[df.subjects_all.apply(lambda x: len(x) > 0)]))\n", - " print(\"rows with suggestions: \", len(df[df.suggestions.apply(lambda x: len(x) > 0)]))\n", - " print(\"rows with subjects_uris: \", len(df[df.subjects_uris.apply(lambda x: len(x) > 0)]))\n", - " print(\"rows with both suggestions & subject_uris: \", len(retain_relevant(df)))\n", - "\n", - "\n", - "def retain_relevant(df):\n", - " df = df.copy()\n", - " return df[df.suggestions.apply(lambda x: len([e for e in x if e != '']) > 0) & df.subjects_uris.apply(lambda x: len([e for e in x if e != '']) > 0)]\n", - "\n", - "\n", - "# Dates when Annif models were updated in Finto AI\n", - "annif_updates = [\n", - " \"2020-03-26\",\n", - " \"2020-12-09\",\n", - " \"2021-04-27\",\n", - " \"2021-11-11\",\n", - " \"2022-06-21\",\n", - " \"2022-11-22\",\n", - "]\n", - "# convert to datetime with utc timezone\n", - "annif_updates = [datetime.datetime.strptime(x, \"%Y-%m-%d\").replace(tzinfo=datetime.timezone.utc) for x in annif_updates]\n", - "\n", - "def get_annif_update(date):\n", - " return max([upd for upd in annif_updates if upd starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "/tmp/ipykernel_184606/68295397.py:10: UserWarning: The label '_' of starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "/tmp/ipykernel_184606/68295397.py:10: UserWarning: The label '_' of starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "/tmp/ipykernel_184606/68295397.py:10: UserWarning: The label '_' of starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "/tmp/ipykernel_184606/68295397.py:10: UserWarning: The label '_' of starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "/tmp/ipykernel_184606/68295397.py:10: UserWarning: The label '_' of starts with '_'. It is thus excluded from the legend.\n", - " plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n" - ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsAAAAFKCAYAAAAT7OnMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABBUklEQVR4nO3deXxM9/oH8M9kFSERIbHWGkTFXoS6qCWtKL1aXIpY26rSJK6tLbXV0hZV1bq1V2lRWjvJTYkittjFLsQtEZJIrMlk5vn94ZdT06jmxExm5szn/Xrd182cczKeT05P5plvvvM9OhEREBERERE5CCdrF0BEREREVJjYABMRERGRQ2EDTEREREQOhQ0wERERETkUNsBERERE5FDYABMRERGRQ2EDTEREREQOxcXaBdgDo9GIa9euoXjx4tDpdNYuh4iIiIj+RERw584dlCtXDk5OTx/jZQOcD9euXUPFihWtXQYRERER/Y2rV6+iQoUKTz2GDXA+FC9eHMCjH6iXl5eVqzEPvV6PqKgodOjQAa6urtYux6yYzT5pORug7XzMZp+0nA3Qdj4tZ3sWmZmZqFixotK3PQ0b4HzInfbg5eWlqQa4aNGi8PLy0tzFw2z2ScvZAG3nYzb7pOVsgLbzaTmbOeRnuio/BEdEREREDoUNMBERERE5FDbARERERORQ2AATERERkUNhA0xEREREDoUNMBERERE5FDbARERERORQ2AATERERkUPhjTCIiIjMqPKYzaqOvzw91EKVENFf4QgwERERETkUNsBERERE5FDYABMRERGRQ2EDTEREREQOhQ0wERERETkUNsBERERE5FDYABMRERGRQ2EDTEREREQOhQ0wERERETkUNsBERERE5FDYABMRERGRQ2EDTEREREQOhQ0wERERETkUNsBERERE5FDYABMRERGRQ2EDTEREREQOhQ0wERERETkUNsBERERE5FDYABMRERGRQ7GZBnj69OnQ6XQIDw9Xtj18+BBDhw6Fr68vihUrhtdffx03btww+b6kpCSEhoaiaNGi8PPzw8iRI5GTk2NyzM6dO9GwYUO4u7ujevXqWLp0aSEkIiIiIiJbZBMN8MGDB/Gf//wHdevWNdkeERGBjRs3Ys2aNYiNjcW1a9fQtWtXZb/BYEBoaCiys7Oxd+9eLFu2DEuXLsX48eOVYxITExEaGoo2bdrg6NGjCA8Px6BBg7B9+/ZCy0dEREREtsPqDfDdu3fx5ptvYsGCBfDx8VG2Z2RkYNGiRZg1axZeeuklNGrUCEuWLMHevXuxb98+AEBUVBQSEhLw/fffo379+njllVcwefJkzJs3D9nZ2QCA+fPno0qVKpg5cyYCAwPx3nvv4Y033sDs2bOtkpeIiIiIrMvF2gUMHToUoaGhaNeuHaZMmaJsj4+Ph16vR7t27ZRttWrVwnPPPYe4uDg0a9YMcXFxCAoKgr+/v3JMSEgIhgwZglOnTqFBgwaIi4szeY7cYx6favFnWVlZyMrKUh5nZmYCAPR6PfR6/bNGtgm5ObSS53HMZp+0nA3Qdj5mM8+/Udi0fN4AbefTcrZnoebnYdUG+Mcff8Thw4dx8ODBPPuSk5Ph5uaGEiVKmGz39/dHcnKycszjzW/u/tx9TzsmMzMTDx48gIeHR55/e9q0aZg4cWKe7VFRUShatGj+A9qB6Ohoa5dgMcxmn7ScDdB2PmbLpe6ldcuWLeqKMTMtnzdA2/m0nK0g7t+/n+9jrdYAX716Fe+//z6io6NRpEgRa5XxRGPHjkVkZKTyODMzExUrVkSHDh3g5eVlxcrMR6/XIzo6Gu3bt4erq6u1yzErZrNPWs4GaDsfs5l6Py5K1b/RsWPHgpT2zLR83gBt59NytmeR+xf7/LBaAxwfH4+UlBQ0bNhQ2WYwGLBr1y589dVX2L59O7Kzs3H79m2TUeAbN26gTJkyAIAyZcrgwIEDJs+bu0rE48f8eeWIGzduwMvL64mjvwDg7u4Od3f3PNtdXV019x+aFjPlYjb7pOVsgLbzMVvBn9uatHzeAG3n03K2glDzs7Dah+Datm2LEydO4OjRo8r/GjdujDfffFP52tXVFTExMcr3nD17FklJSQgODgYABAcH48SJE0hJSVGOiY6OhpeXF2rXrq0c8/hz5B6T+xxERERE5FisNgJcvHhx1KlTx2Sbp6cnfH19le0DBw5EZGQkSpYsCS8vLwwbNgzBwcFo1qwZAKBDhw6oXbs2+vTpg08//RTJycn46KOPMHToUGUE95133sFXX32FUaNGYcCAAfj111+xevVqbN68uXADExEREZFNsPoqEE8ze/ZsODk54fXXX0dWVhZCQkLw9ddfK/udnZ2xadMmDBkyBMHBwfD09ERYWBgmTZqkHFOlShVs3rwZERERmDNnDipUqICFCxciJCTEGpGIiIiIyMpsqgHeuXOnyeMiRYpg3rx5mDdv3l9+T6VKlf72E7StW7fGkSNHzFEiEREREdk5q98Ig4iIiIioMNnUCDARmV/lMermu1+eHmqhSoiIiGwDR4CJiIiIyKGwASYiIiIih6J6CoTBYMDs2bOxevVqJCUlITs722R/Wlqa2YojIiIiIjK3fI0Ad+3aVbm93MSJEzFr1iz06NEDGRkZiIyMRNeuXeHk5IQJEyZYslYiIiIiomeWrwbY29sbOp0OALBixQp8++23GDFiBFxcXNCzZ08sXLgQ48ePx759+yxaLBERERHRs8pXA7xkyRIUL14cAHD9+nXUrVsXAFCsWDFkZGQAADp16sS7qxERERGRzctXA+zs7IyUlBQAQIUKFZCcnAwAqFatGqKiogAABw8eVG4/TERERERkq/LVAK9btw4+Pj4AgFdffRXbt28HAAwbNgzjxo1DQEAA+vbtiwEDBliuUiIiIiIiM8jXKhBdunRRvp45c6bydY8ePVCpUiXs3bsXAQEBePXVV81fIRERERGRGalaBk2v1+Ptt9/GuHHjUKVKFQBAs2bN0KxZM4sUR0RERERkbqpuhOHq6oq1a9daqhYiIiIiIotTfSe41157Db/88osFSiEiIiIisjzVd4ILCAjApEmTsGfPHjRq1Aienp4m+4cPH2624oiIiIiIzE11A7xo0SKUKFEC8fHxiI+PN9mn0+nYABMRERGRTVPdACcmJlqiDiIiIiKiQqF6DjARERERkT1TPQIMAP/73/+wYcMGJCUlITs722TfrFmzzFIYEREREZEl5KsB3rBhA1555RW4uroiJiYGnTt3RtWqVXHmzBnUqVMHly9fhoigYcOGlq6XiIiIyOZVHrNZ1fGXp4daqBJ6knxNgXjttdeQnp4OAPjggw8wYsQInDhxAkWKFMHatWtx9epVtGrVCt26dbNosUREREREzypfDbDRaISfnx8AICEhAX379gUAuLi44MGDByhWrBgmTZqEGTNmWK5SIiIiIiIzyFcDPGDAANy5cwcA4OnpCb1eDwAoW7YsLl68qBx369YtC5RIRERERGQ++ZoDnJ6eDhEBADRp0gR79+5FYGAgOnbsqEyHWLduHZo1a2bRYomIiIiInlW+GuCff/5Z+fqzzz7D3bt3AQATJ07E3bt3sWrVKgQEBHAFCCIiIiKyeaqXQatZs6bytaenJ+bPn2/WgoiIiIiILIk3wiAiIiIih6J6BNhgMGD27NlYvXr1E2+EkZaWZrbiiIiIiIjMLV8jwF27dkVmZiaAR/N+Z82ahR49eiAjIwORkZHo2rUrnJycMGHCBEvWSkRERET0zPLVAHt7e0On0wEAVqxYgW+//RYjRoyAi4sLevbsiYULF2L8+PHYt2+fRYslIiIiInpW+WqAlyxZguLFiwMArl+/jrp16wIAihUrhoyMDABAp06dsHmzutv+EREREREVtnw1wM7OzkhJSQEAVKhQAcnJyQCAatWqISoqCgBw8OBBuLu7W6hMIiIiIiLzyFcDvG7dOvj4+AAAXn31VWzfvh0AMGzYMIwbNw4BAQHo27cvBgwYYLlKiYiIiIjMIF+rQHTp0kX5eubMmcrXPXr0wHPPPYe4uDgEBATg1VdfNX+FRERERERmpHoZtD8LDg5GcHCwOWohIiILqTxG3Wc0Lk8PtVAlRETWV6AG+Nq1a9i9ezdSUlJgNBpN9g0fPtwshRERERERWYLqBnjp0qV4++234ebmBl9fX2V5NADQ6XRsgImIiIjIpqlugMeNG4fx48dj7NixcHLinZSJiIiIyL6o7mDv37+Pf/3rX2x+iYiIiMguqe5iBw4ciDVr1liiFiIiIiIii1M9BWLatGno1KkTtm3bhqCgILi6uprsnzVrltmKI+3iJ9KJiIjIWgrUAG/fvh01a9YEgDwfgiMiIiIismWqG+CZM2di8eLF6NevnwXKISIiIiKyLNVzgN3d3dGiRQtL1EJEREREZHGqG+D3338fc+fOtUQtREREREQWp3oKxIEDB/Drr79i06ZNeP755/N8CG7dunVmK46IiMiR8QPDRJahugEuUaIEunbtaolaiIiIiIgsTnUDvGTJEkvUQURERERUKHg7NyIiIiJyKGyAiYiIiMihsAEmIiIiIofCBpiIiIiIHAobYCIiIiJyKKpXgQCAmJgYxMTEICUlBUaj0WTf4sWLzVIYEREREZElqG6AJ06ciEmTJqFx48YoW7YsdDqdJeoiIiIiIrII1Q3w/PnzsXTpUvTp08cS9RARERERWZTqOcDZ2dlo3ry5JWohIiIiIrI41Q3woEGDsHLlSkvUQkRERERkcaqnQDx8+BDffvst/vvf/6Ju3bpwdXU12T9r1iyzFUdEREREZG6qG+Djx4+jfv36AICTJ0+a7OMH4oiIiIjI1qlugHfs2GGJOojIDlQes1nV8Zenh1qoEiIiooIr8I0wLly4gO3bt+PBgwcAABExW1FERERERJaiugFOTU1F27ZtUaNGDXTs2BHXr18HAAwcOBAjRowwe4FEREREROakugGOiIiAq6srkpKSULRoUWV7jx49sG3bNrMWR0RERERkbqrnAEdFRWH79u2oUKGCyfaAgABcuXLFbIUREREREVmC6hHge/fumYz85kpLS4O7u7tZiiIiIiIishTVDXDLli3x3XffKY91Oh2MRiM+/fRTtGnTxqzFERERERGZm+opEJ9++inatm2LQ4cOITs7G6NGjcKpU6eQlpaGPXv2WKJGIiIiIiKzUT0CXKdOHZw7dw4vvvgiunTpgnv37qFr1644cuQIqlWrZokaiYiIiIjMRvUIMAB4e3vjww8/NHctREREREQWl68G+Pjx46hTpw6cnJxw/Pjxpx5bt25dsxRGRERERGQJ+WqA69evj+TkZPj5+aF+/frQ6XRPvPObTqeDwWAwe5FEREREROaSrwY4MTERpUuXVr4mIiIiIrJX+WqAK1WqpHx95coVNG/eHC4upt+ak5ODvXv3mhxLRERERGRrVK8C0aZNG6SlpeXZnpGRwXWAiYiIiMjmqW6ARQQ6nS7P9tTUVHh6epqlKCIiIiIiS8n3Mmhdu3YF8OiDbv369TO57bHBYMDx48fRvHlz81dIRERERGRG+W6Avb29ATwaAS5evDg8PDyUfW5ubmjWrBkGDx5s/gqJiIiIiMwo3w3wkiVLAACVK1fGyJEjUbRoUYsVRURERERkKarvBBcbG4v3338/TwOcmZmJ1157Db/++qvZiiMiIiIqLJXHbFZ1/OXpoRaqhCxN9YfgYmNjkZ2dnWf7w4cP8dtvv5mlKCIiIiIiS8n3CHDuLZBFBAkJCUhOTlb2GQwGbNu2DeXLlzd/hUREREREZpTvBjj3Fsg6nQ4vvfRSnv0eHh6YO3euWYsjIiIiIjK3fDfAiYmJEBFUrVoVBw4cUG6NDDxaBcLPzw/Ozs4WKZKIiIiIyFzy3QDn3uLYaDRarBgiIiIiIktTvQoEAJw/fx47duxASkpKnoZ4/PjxZimMiIiIiMgSVDfACxYswJAhQ1CqVCmUKVPG5LbIOp2ODTARERER2TTVDfCUKVPwySefYPTo0c/8j0+bNg3r1q3DmTNn4OHhgebNm2PGjBmoWbOmcszDhw8xYsQI/Pjjj8jKykJISAi+/vpr+Pv7K8ckJSVhyJAh2LFjB4oVK4awsDBMmzYNLi5/xNu5cyciIyNx6tQpVKxYER999BH69ev3zBmIiCyFa5ISEVmG6nWA09PT0a1bN7P847GxsRg6dCj27duH6Oho6PV6dOjQAffu3VOOiYiIwMaNG7FmzRrExsbi2rVr6Nq1q7LfYDAgNDQU2dnZ2Lt3L5YtW4alS5eajEQnJiYiNDQUbdq0wdGjRxEeHo5BgwZh+/btZslBRERERPZD9Qhwt27dEBUVhXfeeeeZ//Ft27aZPF66dCn8/PwQHx+Pf/zjH8jIyMCiRYuwcuVKZem1JUuWIDAwEPv27UOzZs0QFRWFhIQE/Pe//4W/vz/q16+PyZMnY/To0ZgwYQLc3Nwwf/58VKlSBTNnzgQABAYGYvfu3Zg9ezZCQkKeOQcRERER2Q/VDXD16tUxbtw47Nu3D0FBQXB1dTXZP3z48AIXk5GRAQAoWbIkACA+Ph56vR7t2rVTjqlVqxaee+45xMXFoVmzZoiLi0NQUJDJlIiQkBAMGTIEp06dQoMGDRAXF2fyHLnHhIeHP7GOrKwsZGVlKY8zMzMBAHq9Hnq9vsD5bEluDnvJo6ZOe8umRmFkM+dz87z9gefOPvG82S+eO8ej5uehugH+9ttvUaxYMcTGxiI2NtZkn06nK3ADbDQaER4ejhYtWqBOnToAgOTkZLi5uaFEiRImx/r7+yt3oktOTjZpfnP35+572jGZmZl48OABPDw8TPZNmzYNEydOzFNjVFQUihYtWqB8tio6OtpK/7K6//S2bNmi+l+wXjbLU5fNnD9rnrdnxXNnn3je7BfPneO4f/9+vo9V3QAnJiaq/ZZ8GTp0KE6ePIndu3db5PnVGDt2LCIjI5XHmZmZqFixIjp06AAvLy8rVmY+er0e0dHRaN++fZ5R/MLwflyUquM7duyY72Otnc2SCpLNnD9rnreC47mzTzxv9ovnzvHk/sU+Pwq0DjAAZGdnIzExEdWqVTNZbaEg3nvvPWzatAm7du1ChQoVlO1lypRBdnY2bt++bTIKfOPGDZQpU0Y55sCBAybPd+PGDWVf7v/nbnv8GC8vrzyjvwDg7u4Od3f3PNtdXV019x+avWQqSI32kq0gLJnNnM/L85YXz5194nmzXzx3jkPNz0J153r//n0MGzYMy5YtAwCcO3cOVatWxbBhw1C+fHmMGTMm388lIhg2bBh+/vln7Ny5E1WqVDHZ36hRI7i6uiImJgavv/46AODs2bNISkpCcHAwACA4OBiffPIJUlJS4OfnB+DRnwS8vLxQu3Zt5Zg//2khOjpaeQ7KHy7JRERERFqgehm0sWPH4tixY9i5cyeKFCmibG/Xrh1WrVql6rmGDh2K77//HitXrkTx4sWRnJyM5ORkPHjwAADg7e2NgQMHIjIyEjt27EB8fDz69++P4OBgNGvWDADQoUMH1K5dG3369MGxY8ewfft2fPTRRxg6dKgyivvOO+/g0qVLGDVqFM6cOYOvv/4aq1evRkREhNr4RERERGTnVI8A//LLL1i1ahWaNWtmche4559/HhcvXlT1XN988w0AoHXr1ibblyxZotykYvbs2XBycsLrr79uciOMXM7Ozti0aROGDBmC4OBgeHp6IiwsDJMmTVKOqVKlCjZv3oyIiAjMmTMHFSpUwMKFC7kEGhEREZEDUt0A37x5U5lq8Lh79+6ZNMT5ISJ/e0yRIkUwb948zJs37y+PqVSp0t9+erJ169Y4cuSIqvqIiIiISHtUT4Fo3LgxNm/+Yy5obtO7cOFCzqklIiIiIpunegR46tSpeOWVV5CQkICcnBzMmTMHCQkJ2Lt3b551gYmIiIiIbI3qEeAXX3wRR48eRU5ODoKCghAVFQU/Pz/ExcWhUaNGlqiRiIiIiMhsCrSAb7Vq1bBgwQJz10JEREREZHEFvoNFSkoKUlJSYDQaTbbXrVv3mYsiIiIiIrIU1Q1wfHw8wsLCcPr06TyrOOh0OhgMBrMVR0RERERkbqob4AEDBqBGjRpYtGgR/P39VS99RkRERERkTaob4EuXLmHt2rWoXr26JeohIiIiIrIo1atAtG3bFseOHbNELUREREREFqd6BHjhwoUICwvDyZMnUadOHbi6uprs79y5s9mKIyIiIiIyN9UNcFxcHPbs2YOtW7fm2ccPwRERERGRrVPdAA8bNgy9e/fGuHHj4O/vb4maiAhA5TGb/2KPC96Pi8qz9fL0UMsWREREpBGq5wCnpqYiIiKCzS8RERER2SXVDXDXrl2xY8cOS9RCRERERGRxqqdA1KhRA2PHjsXu3bsRFBSU50Nww4cPN1txREREhYXTjogcR4FWgShWrBhiY2MRGxtrsk+n07EBJiIiIiKbproBTkxMtEQdRERERESFQvUcYCIiIiIie6Z6BHjAgAFP3b948eICF0NEREREZGmqG+D09HSTx3q9HidPnsTt27fx0ksvma0wIiIiIiJLUN0A//zzz3m2GY1GDBkyBNWqVTNLUURERERElmKWOcBOTk6IjIzE7NmzzfF0REREREQWY7YPwV28eBE5OTnmejoiIiIiIotQPQUiMjLS5LGI4Pr169i8eTPCwsLMVhgRERERkSWoboCPHDli8tjJyQmlS5fGzJkz/3aFCCIiIiIia1PdAG/fvh1ubm5P3Hfr1i2UKlXqmYsiIiIiIrIU1XOAe/bsCRHJs/3GjRto3bq1OWoiIiIiIrIY1Q1wUlISBg0aZLLt+vXraN26NWrVqmW2woiIiIiILEF1A7xlyxbs3btX+TDctWvX0Lp1awQFBWH16tVmL5CIiIiIyJxUzwEuXbo0oqKi8OKLLwIANm3ahIYNG2LFihVwcjLbqmpERERERBahugEGgIoVKyI6OhotW7ZE+/btsXz5cuh0OnPXRkRERERkdvlqgH18fJ7Y4N6/fx8bN26Er6+vsi0tLc181RERERERmVm+GuAvvvjCwmUQERERERWOfDXAvMMbEREREWlFgeYAGwwG/PLLLzh9+jQA4Pnnn0fnzp3h7Oxs1uKIiIiIiMxNdQN84cIFdOzYEb///jtq1qwJAJg2bRoqVqyIzZs3o1q1amYvkoiIiIjIXFSvWzZ8+HBUq1YNV69exeHDh3H48GEkJSWhSpUqGD58uCVqJCIiIiIyG9UjwLGxsdi3bx9KliypbPP19cX06dPRokULsxZHRERERGRuqhtgd3d33LlzJ8/2u3fvws3NzSxFEREREZHtqTxms6rjL08PtVAlz0b1FIhOnTrhrbfewv79+yEiEBHs27cP77zzDjp37myJGomIiIiIzEZ1A/zll1+iWrVqCA4ORpEiRVCkSBG0aNEC1atXx5w5cyxRIxERERGR2aieAlGiRAmsX78e58+fx5kzZwAAgYGBqF69utmLIyIiIiIytwKtAwwAAQEBCAgIMGctREREREQWl68GODIyMt9POGvWrAIXQ0RERERkaflqgI8cOWLy+PDhw8jJyVFuhHHu3Dk4OzujUaNG5q+QyI5o5dOxREREWpavBnjHjh3K17NmzULx4sWxbNky+Pj4AADS09PRv39/tGzZ0jJVEhERERGZieo5wDNnzkRUVJTS/AKAj48PpkyZgg4dOmDEiBFmLZCIiOiv8K8uRFQQqpdBy8zMxM2bN/Nsv3nz5hNvkEFEREREZEtUN8D//Oc/0b9/f6xbtw7/+9//8L///Q9r167FwIED0bVrV0vUSERERERkNqqnQMyfPx///ve/0atXL+j1+kdP4uKCgQMH4rPPPjN7gURERERE5qS6AS5atCi+/vprfPbZZ7h48SIAoFq1avD09DR7cURERERE5lbgG2F4enqibt265qyFiIiIiMjiVM8BJiIiIiKyZwUeASYiIiIi+8FlA//ABpiIiKyCL8ZEZC1sgIn+H1+MiYiIHAPnABMRERGRQ2EDTEREREQOhVMgHMBf/2nfBe/HReXZyj/tExERkZZxBJiIiIiIHApHgImIiIhsFP+KaxlsgImIoH4VEOCvX2i0/IKl5Wxkv/jfJanFKRBERERE5FA4Akx2je/6iYjskzn/6kKkFkeAiYiIiMihsAEmIiIiIofCBpiIiIiIHAobYCIiIiJyKGyAiYiIiMihsAEmIiIiIofCBpiIiIiIHAobYCIiIiJyKGyAiYiIiMihsAEmIiIiIofCBpiIiIiIHAobYCIiIiJyKC7WLoCIiIjsR+Uxm1Udf3l6qIUqISo4jgATERERkUPhCLCNUvsOG+C7bHJMHI0iIiK1OAJMRERERA6FDTARERERORQ2wERERETkUNgAExEREZFDYQNMRERERA6FDTARERERORQ2wERERETkUNgAExEREZFDcagGeN68eahcuTKKFCmCpk2b4sCBA9YuiYiIiIgKmcPcCW7VqlWIjIzE/Pnz0bRpU3zxxRcICQnB2bNn4efnZ+3yiBzKX9+9zQXvx0Xl2cq7txERkTk5TAM8a9YsDB48GP379wcAzJ8/H5s3b8bixYsxZswYK1dHRERkOXzTSWTKIRrg7OxsxMfHY+zYsco2JycntGvXDnFxcXmOz8rKQlZWlvI4IyMDAJCWlga9Xm/5ggEYs+6r/p7U1FSzPJeln0fLNdlaPazJcjWZ87m0/HPiNceaeM1pqyZz1mNud+7cAQCIyN8eq5P8HGXnrl27hvLly2Pv3r0IDg5Wto8aNQqxsbHYv3+/yfETJkzAxIkTC7tMIiIiInpGV69eRYUKFZ56jEOMAKs1duxYREZGKo+NRiPS0tLg6+sLnU5nxcrMJzMzExUrVsTVq1fh5eVl7XLMitnsk5azAdrOx2z2ScvZAG3n03K2ZyEiuHPnDsqVK/e3xzpEA1yqVCk4Ozvjxo0bJttv3LiBMmXK5Dne3d0d7u7uJttKlChhyRKtxsvLS7MXD7PZJy1nA7Sdj9nsk5azAdrOp+VsBeXt7Z2v4xxiGTQ3Nzc0atQIMTExyjaj0YiYmBiTKRFEREREpH0OMQIMAJGRkQgLC0Pjxo3RpEkTfPHFF7h3756yKgQREREROQaHaYB79OiBmzdvYvz48UhOTkb9+vWxbds2+Pv7W7s0q3B3d8fHH3+cZ6qHFjCbfdJyNkDb+ZjNPmk5G6DtfFrOVlgcYhUIIiIiIqJcDjEHmIiIiIgoFxtgIiIiInIobICJiIiIyKGwASYiIiIih8IGmIiIiIgcChtgIrIJWVlZ1i6ByKEYjUZrl0BkNWyASXO0urLf3bt38eDBA2uXYRFnz57F+PHjkZOTY+1SqAC02khp+U3ZmTNnMGfOHGuXQQWk1WuuMDnMjTDokYsXL+Knn35CZmYm6tWrh9DQUHh6elq7LLNIS0tDyZIlodPpICLQ6XTWLslszp07h+7duyMiIgLdu3eHh4eHtUsym+PHj6Np06bIyspCmzZt8PLLL1u7JLM6d+4clixZgpSUFDRo0AAvv/wyqlevbu2yzCIjIwPe3t5wcnKC0WiEk5N2xlQSEhIwePBgzJgxAy+++KK1yzGrEydO4IUXXkB2djaaN2+Opk2bWrsks9Ly65yWr7nCxp+cAzl58iQaN26MLVu2YNeuXejVqxf69++P6Ohoa5f2zBISEuDv74/w8HAAUJpgrViyZAmOHz+ODz/8EL/88otmRqaOHTuGZs2aYcCAAejWrRtWrlyJBw8eaObcJSQkoEmTJjhy5AjS09PxwQcfYOjQoVi8eLG1S3tmCQkJqFSpEqZOnQoAyguyVnz++eeIi4tDv379EBcXZ+1yzObYsWNo0qQJevTogVatWmHTpk0AtDOiqPXXOS1fc4VOyCHcv39fOnbsKO+9956ybf/+/dKoUSNp3769/PLLL1as7tn8/vvv0qRJE2nYsKF4enpKeHi4ss9oNFqxMvPZvn27fPjhhzJixAhxd3eX5cuXW7ukZxYfHy/FixeXDz/8UEREZs6cKSVKlJBLly6JiP2fu6ysLHnzzTflrbfeUradOXNGunfvLsHBwfLVV19Zsbpnc/XqVWnQoIHUqFFDSpYsKdOmTVP2GQwGK1ZmPosXL5bRo0fLwIEDxdfXV3bt2mXtkp7Z4cOHTa65kSNHSunSpeX27dsiYv/XnJZf5xzhmitsHAF2EB4eHkhPT4efnx+AR+/2mzRpgmXLliErKwv/+c9/cPz4cStXqZ6IYMeOHahUqRLmzp2LBQsW4JtvvkFkZCQAbY0Er1+/Hp9//jl69+6Nt99+G5s2bcKQIUPw1VdfWbs01W7fvo0XX3wRb731FqZMmQIAGDp0KKpWrYrJkydrYgqLm5sbkpOTlT9Righq1qyJGTNmoEaNGli9ejU2b95s5SrVMxqNWLt2LapUqYL58+dj1KhRmDZtGqZPnw5AO6NSRYsWxW+//Ya5c+ciODgYb7zxBk6fPo1x48Zh1apV1i5PtZSUFLRo0QJvv/22cs0NGzYMJUuWVOYC2/s1p9XXOUe55gqddftvKgxGo1Hu3LkjrVq1Ut4Z6/V6ycnJERGR48ePS7ly5SQyMtKaZRbYlStXZP369crjFStWiLu7u6ZGgjMyMuQf//iHPHjwQEREhg8fLi4uLlKiRAk5ePCglasrmCNHjihfG41GycnJkTFjxkhQUJDcvHlT2W6PjEajZGdnS1hYmHTv3l0ePnwoBoNBGak5f/68NG3aVHr16mXlSgvm3LlzsnLlShERSUtLk2nTpomXl5emRqXOnTsnrVu3Vh736NFD3N3dxdfXV86dO2fFygomLS1NYmNjTbZlZWXJv/71L2nRooWyzZ6vOS2/zjnCNVfY2AA7kB9//FF0Op3SLBoMBsnOzhYRkeXLl4uPj48kJSVZs8QCe/yXdk5OjqxcuVLc3d0lIiJCRB79Ivz+++/lxIkT1irxmTVq1EhiYmJERGTQoEFSrFgxKVKkiKxZs0ZpjO1N7nnL/f9r166Jp6enTJ061ZplmU1MTIw4OTnJvHnzROSPRv/xfQkJCdYsscAev+Zu3rwp06dPN3lBzsnJkQ0bNihvZuxRvXr15OzZsyIi0qtXL/H09BQfHx+7fdP5uNxm6eTJk+Lu7i6LFi2yckXm4Sivc1q95goTV4HQKL1eD1dXVwB/LAv2xhtvYPfu3ejRowd+/vlnvPzyy8qfZ318fFC2bFm7+KTs1atXcfr0ady8eRPt27dHiRIl4ObmhpycHLi4uMDZ2RndunUDAPTv3x8AYDAY8M033+DChQvWLP1v/VU2g8GAGjVqwGAwYPjw4diyZQtOnjyJzz77DN27d8fq1avxxhtvWLv8p3paNmdnZxgMBpQtWxZvvfUWNm/ejN69e6NixYrWLjvf7t+/D1dXV7i6ukIeDS7gpZdewrRp0zBs2DB4eHigf//+cHZ2BgB4e3ujZs2aKFq0qJUr/3vXrl3D77//jtTUVLRr1w5OTk5wcnJSrrlSpUphwIABAICpU6dCRJCamoo5c+YgKSnJytU/3ePZ2rdvD51OBycnJzx48AA+Pj64c+cOhg8fjp07d+LXX3/FzJkz0axZM+zduxdNmjSxdvlP9VfnLXf1ABFBlSpV0KlTJ2zduhW9evWCu7u73UyF0PLrnJavOZthxeabLOTkyZPSuXNnOXXqVJ59iYmJMnDgQHFzc5OFCxdKcnKyPHz4UEaPHi316tWTtLQ0K1Scf8eOHRN/f39p2LChuLm5yfPPPy8jR46U9PR0ERFldC336+XLl4tOp7OLUZu/ypaamioiIp988onodDopW7asSZbw8HA5ffq0tcrOFzXnbevWrVK8eHHZuHGjlapV79SpU/LSSy/Jxo0bRa/Xi8gfozX379+X8ePHi06nk48//liOHj0q6enpMmbMGAkICJCUlBRrlv63jh07JhUrVpTatWuLi4uLNGjQQL755hu5c+eOiJieu5s3b8q0adPs6pp7UraMjAwRERkxYoR4eHhIuXLlJD4+XkQeTRvo3bu3MjJsq/7uvD3+5/LcaWMHDhywVrmqaf11TqvXnC1hA6wxiYmJUrVqVdHpdFK/fv0n/pK+fv26TJo0SVxdXaVatWpSr149KVWqlBw+fNgKFeff7du3pWHDhjJixAhJTU2VBw8eyNixY6V58+bSpUsXpVHM/eVgMBhk4MCB4uXlZfN/Zn5ats6dO0t6erocO3ZM+vXrp8ydffyXoC1Te95ERDp06CCtWrUSg8Fg83MSL1++LLVq1RJnZ2epXLmyREVFKU3w4xYvXiz+/v5Svnx5CQwMlHLlytn8NXfz5k0JDAyU0aNHS2JioqSkpEjPnj2ladOmEh4eLpmZmSJi2kz16dNHvLy8ntiY2JKnZXv//ffl/v37sn79egkNDTWZr24P8nveHr/mGjRoIH369LGLa07Lr3NavuZsDRtgDXn48KFMmDBB/vnPf8rBgwelSZMmEhgY+JcjFYcPH5YffvhBVq5cKYmJiYVbbAHk/tLbuXOnsi0rK0sWL14swcHB8uabbyq/HIxGo2zZskWqVKliF++In5atadOm0rdvX8nKynpiY2Xr1Jy33F/qa9eulQsXLlilXjX0er3Mnj1bXnvtNbly5Yq8/PLLUqZMmb9sgi9evCg7d+6UrVu3ytWrV61QsTonTpyQypUry7Fjx5RtWVlZMn78eGnSpIl8+OGHyvxzo9Eoy5cvF39/f2W01JY9LVvjxo1l4sSJIiLKqJs9UXPecs2ZM0fOnz9f2KWqpvXXOS1fc7aGDbCGGAwGWbt2raxZs0ZERNLT0//yl4Otv8N/kps3b0qdOnVk7ty5IvJHBoPBIPPmzZOGDRvKd999pxyfnJws169ft0qtav1dtvr16ytr/9rbuVNz3uxlVDuX0WiU3bt3y9q1a5VtISEhUrZsWYmKilI+fGM0Gu3uvImInD17VqpUqaJMR8lt6vV6vYwcOVLq169vsj7upUuX5PLly1apVa2/y1a3bl357bffRMT+rjk1583e3lRr/XVOy9ecrWEDrDF/biBu3bql/HLIXbpHr9fLnj175OHDh9YoscCys7Pl9ddfl+bNmz/xgu/QoYOEhoZaobJnx2z2mU3kyUsPhYSEKCPBudfkxo0b5e7du4Vd3jN5+PChNG7cWDp16qTkeHyOc1BQkPTt21d5bE/UZLM3Ws4mou3XOS1fc7aGN8LQmNxPmMv/fyLW19cXmzdvRvHixdGlSxecOnUKw4YNQ0REBO7evWvNUlUREbi6uuLrr7/GxYsXMXz4cKSkpJjc5OLVV1/FrVu38PDhQytWqh6z2We2XLmfMAeAnJwcAMC2bdtQv359hIWFISoqCm+//TYiIiKQmZlprTJVMxqNcHd3x5IlS7Br1y4MGTIEAODi4qLcqKRz585ISUkBYF83UVCbzZ5oOVsurb7Oafmas0VsgDUm9xdC7oUhIihVqhS2bNmCEiVKoG7duli2bBnmzZsHX19fa5aqik6nQ3Z2Nvz8/LBt2zbs378fvXv3xqFDh2AwGAAAR48eha+vr0lDYg+YzT6z5Xq8mXdxcYFerwcAbN26FfXr10doaChWrFiBH3/8EWXLlrVWmao5OTnBYDCgTp06WLZsGX744Qf07dsXN27cUI5JTEyEj4+Pci7tBbPZZ7ZcWn2dc4RzZ0t0Ihq5Tywp66lmZmbCaDSiRIkSJvsHDBiADRs2YNeuXahdu7Z1iiyg3GypqanIzs7GgwcP8Morr6BYsWLIyclB1apVERMTg927d6Nu3brWLlcVZrPPbMAf+dLT03Hr1i0EBAQAgLJWZ0REBL777jv89ttvdnfN5Wa4e/cusrKycPToUfTq1QuVKlVCyZIl4evri/Xr1yMuLg5BQUHWLlcVZrPPbIC2XufkT7d81/q5szX2OeTi4P78nkVEkJOTA2dnZ1y+fBmBgYGIi4sz2T937lwsXboU0dHRNv9L4c9yf+FdvnwZdevWRUxMDKpWrYqDBw8iPDwc7du3xwsvvICDBw/aXRPFbPaZDfgj35UrV1CnTh3Ex8cr+1xcXLBw4ULMmTMHUVFRdnfN5b4QX758GTVq1MDBgwfRtm1bnDp1Ch07dkT58uXh5+eHAwcO2N0LMbPZZzYAmnmdyx29zX0tz30N1/K5s0mFNdmYzOPMmTMybtw4CQsLkwULFpjcAOHKlSvi6+srgwYNMpkcbzQaZceOHTa/xE1ycvJfLmVz9epVKVGihAwePFiMRqPd3fOc2ewzm4i6fI87f/68zS+7dOnSJfnmm28kIiJCoqKiTG6hmpSUJKVKlZKBAwea3ML58VU8bBmz2Wc2kfzls9fXubNnz0p4eLh07dpVJk6cKJcuXVL2aeHc2RM2wHbk1KlT4u3trXyivmnTplKhQgWJjo4WkUfrOIaHh9vlJ0MTEhLkueeek+7du8vJkyfz7P/5559lxIgRzGZjtJxNRNv5jh8/LuXKlZNXXnlFAgICpGbNmjJjxgzJycmR7Oxs+eqrryQiIiJPttzHtpyZ2ewzm0jB89mD48ePi6+vr4SFhclrr70mzZo1k08++USMRqNkZ2fL3Llzn/gabi/nzt6wAbYTOTk50rt3b3nzzTeVbUeOHJFBgwaJs7OzREVFKcfZm99//12aN28u9erVkyZNmsjAgQPlxIkTJsfkrqdqb5jNPrOJaDvf5cuXJSAgQD744AMlw5gxY6R69erKIvu3b9+2ZokFxmz2mU1E2/kuXrwolSpVkg8//FDZNnDgQBk+fLjJcfb4Gm6vOAfYThiNRly9ehUVK1ZUttWvXx9Tp07F4MGD0aVLF+zbt09ZHsaenDlzBsWLF8eyZcvw7rvv4siRI/jiiy9w8uRJ5RhXV1crVlhwzGaf2QDt5jMYDFi/fj0aNGiAYcOGKatvhIeHIzs7G+fOnQMAeHt7W7PMAmE2+8wGaDufwWBAdHQ02rZtixEjRihzfz08PHDy5Em0atUKffv2xd69e+Hs7Jzncz5kGWyA7YSrqyvq1KmD2NhYpKenK9tLly6NDz74AB07dsTkyZPtap3RXM2bN8fHH3+MevXqISwsDO+9957SbJw4cUI5LveXgtFotFapqjGbfWYDtJvP2dkZ3t7eaNGiBcqUKaO8adbpdMjMzERaWlqe77GXF2RmM2Uv2QBt53N2dkaHDh0QGRkJHx8f6HQ6TJo0CQsXLkS7du3QunVrZGdno0+fPkhMTOT6voXFWkPPpN6qVaukQYMGMnPmTMnMzDTZt3TpUilXrpwkJSVZqbpn8+e5TUuXLpWGDRua/Nl54sSJJvdHtxfMZp/ZRLSfT+SPjA8ePJBatWrJ/v37lX3r16+3298pIsxmz7SYLzfTw4cPpWPHjrJp0yZl32+//SZ+fn7KdEayPBdrN+D0ZNeuXcPhw4eRnZ2N5557Do0bN0b37t2xc+dOLFiwAB4eHujRowdKliwJAHjhhRdQtGhR3Llzx8qV/73Hs1WqVAmNGjWCTqeDPJqTDicnJ4SFhQEAvvzyS8yZMweZmZn46aef8MYbb1i5+qdjNvvMBmg735N+nwB/LOUGPFqE38nJSRl9+uCDD7BkyRLs37/fanXnB7PZZzZA2/n+6veJwWCAu7s7Nm7cCCcnJxiNRjg5OaFkyZLw9/dXXtOpEFiv96a/cvz4calatao0adJESpUqJY0bN5YffvhB2d+vXz8JCgqS8PBwuXDhgty8eVNGjRolNWrUkFu3blmx8r/3pGxr1qwxOebxpV4WLVokrq6u4u3tLUeOHCnkatVhNvvMJqLtfPnJJiKSnp4upUuXlj179sjkyZOlSJEicvDgQStUnH/MZp/ZRLSdLz/Z/vzXpTFjxsgLL7xgsuQbWRYbYBtz4cIFqVChgowaNUpu374thw4dkrCwMBkwYIA8fPhQOW7ixInSsmVL0el00qhRIylTpowcPnzYipX/vadly8nJybOmY05OjgwfPlx8fHyeuASVLWG2R+wtm4i286nJdufOHWnQoIG0bt1aihQpIocOHbJi5X+P2R6xt2wi2s6nJpvIo/X7R44cKT4+PnY9lcoesQG2IVlZWRIZGSndu3eXrKwsZfuiRYvE19c3z+jurVu3ZOvWrbJ79265evVqYZeritpsIiIHDhwQnU5n8+/2mc2UvWQT0XY+tdlu374tlSpVkpIlS8rRo0cLu1xVmO0P9pRNRNv51GY7ePCgvPvuu1KvXj2bz6ZFnANsQ4xGIypUqIDAwEC4ubkp9wlv3rw5ihUrBr1erxzn5OQEX19fvPzyy1auOn/ym+1xL7zwAtLS0vLc693WMJspe8kGaDuf2mze3t4YPHgwXn/9ddSqVctKVecPs/3BnrIB2s6nNlvjxo3x4MEDfPTRRyhbtqyVqnZg1uq86ckevy1i7p9Krl+/LtWrVzf51KutT3d4koJks5c73zCbfWYT0Xa+/Gaz9dHsJ2E2+8wmou18+c1m61M5HAHXAbay69ev48CBA9i2bRuMRiOqVKkC4NGnYHM/9ZqRkWGy9u/48ePRtm1bpKam2vQ6iObIZqvrITKbfWYDtJ2voNk6dOig2d8nzGZ9Ws5X0Gzt27e3+WyaZ63Om0SOHTsmlSpVkho1aoi3t7fUqlVLVq5cKampqSLyx7vHs2fPSunSpSUtLU0mT54sHh4eNv/ukdmYzRZpOR+zMZst0nI+LWdzBGyArSQlJUVq1aolH3zwgVy8eFF+//136dGjhwQGBsrHH38sKSkpyrE3btyQBg0aSI8ePcTNzc3mLxxme4TZbIuW8zHbI8xmW7ScT8vZHAUbYCs5deqUVK5cOc+FMHr0aAkKCpJPP/1U7t27JyIiCQkJotPpxMPDw+bXHBVhNmazTVrOx2zMZou0nE/L2RwF5wBbiV6vR05ODu7fvw8AePDgAQBg+vTpaNOmDb755htcuHABAODj44N3330Xhw8fRv369a1Vcr4xG7PZIi3nYzZms0VazqflbI5CJ8IZ2NbSpEkTFCtWDL/++isAICsrC+7u7gAeLbVUvXp1/PDDDwCAhw8fokiRIlarVS1mYzZbpOV8zMZstkjL+bSczRFwBLiQ3Lt3D3fu3EFmZqay7T//+Q9OnTqFXr16AQDc3d2Rk5MDAPjHP/6Be/fuKcfa8oXDbMxmi7Scj9mYzRZpOZ+WszkqNsCFICEhAV27dkWrVq0QGBiIFStWAAACAwMxZ84cREdHo1u3btDr9XByenRKUlJS4OnpiZycHJteJoXZmM0WaTkfszGbLdJyPi1nc2jWmXrsOE6dOiW+vr4SEREhK1askMjISHF1dVUW1b93755s2LBBKlSoILVq1ZLXXntNunfvLp6ennLixAkrV/90zMZstkjL+ZiN2WyRlvNpOZuj4xxgC0pLS0PPnj1Rq1YtzJkzR9nepk0bBAUF4csvv1S23blzB1OmTEFaWhqKFCmCIUOGoHbt2tYoO1+Y7RFmsy1azsdsjzCbbdFyPi1nI8DF2gVomV6vx+3bt/HGG28AeHSfcCcnJ1SpUgVpaWkAAHm0FB2KFy+OGTNmmBxny5iN2WyRlvMxG7PZIi3n03I24hxgi/L398f333+Pli1bAnh0a0QAKF++vHJx6HQ6ODk5mUyst9XbrD6O2ZjNFmk5H7Mxmy3Scj4tZyM2wBYXEBAA4NE7QldXVwCP3jGmpKQox0ybNg0LFy5UPj1qLxcPszGbLdJyPmZjNluk5XxazuboOAWikDg5OUFElAsj993j+PHjMWXKFBw5cgQuLvZ5OpiN2WyRlvMxG7PZIi3n03I2R8UR4EKU+3lDFxcXVKxYEZ9//jk+/fRTHDp0CPXq1bNydc+G2eyTlrMB2s7HbPZJy9kAbefTcjZHxLcrhSj3HaOrqysWLFgALy8v7N69Gw0bNrRyZc+O2eyTlrMB2s7HbPZJy9kAbefTcjZHxBFgKwgJCQEA7N27F40bN7ZyNebFbPZJy9kAbedjNvuk5WyAtvNpOZsj4TrAVnLv3j14enpauwyLYDb7pOVsgLbzMZt90nI2QNv5tJzNUbABJiIiIiKHwikQRERERORQ2AATERERkUNhA0xEREREDoUNMBERERE5FDbARERERORQ2AATERERkUNhA0xEZEeWLl2KEiVKAABat26N8PDwZ3q+ypUr44svvsDOnTuh0+lw+/ZtVd//5xpyn4+IyJbxVshERHakR48e6NixIwBg3bp1cHV1NdlfuXJlhIeHmzSlS5cuRXh4+BOb24MHD8LT0xMuLi64fv06vL29VdXzpBqIiGwdG2AiIjvi4eEBDw8PAEDJkiWf+flKly6tfF2mTBnV32+OGoiIChunQBARFbKffvoJQUFB8PDwgK+vL9q1a4fY2Fi4uroiOTnZ5Njw8HC0bNlSebx27Vo8//zzcHd3R+XKlTFz5kxlX+vWrXHlyhVERERAp9NBp9Nh586d6N+/PzIyMpRtEyZMAACkp6ejb9++8PHxQdGiRfHKK6/g/PnzyvOlpqaiZ8+eKF++PIoWLYqgoCD88MMPJvX93TSMhQsXokSJEoiJiQEAzJo1C0FBQfD09ETFihXx7rvv4u7duwX9URIRFQgbYCKiQnT9+nX07NkTAwYMwOnTp7Fz50507doVjRo1QtWqVbF8+XLlWL1ejxUrVmDAgAEAgPj4eHTv3h3/+te/cOLECUyYMAHjxo3D0qVLATyajlChQgVMmjQJ169fx/Xr19G8eXN88cUX8PLyUrb9+9//BgD069cPhw4dwoYNGxAXFwcRQceOHaHX6wEADx8+RKNGjbB582acPHkSb731Fvr06YMDBw7kK+unn36KMWPGICoqCm3btgUAODk54csvv8SpU6ewbNky/Prrrxg1apS5frxERPkjRERUaOLj4wWAXL58Oc++GTNmSGBgoPJ47dq1UqxYMbl7966IiPTq1Uvat29v8j0jR46U2rVrK48rVaoks2fPNjlmyZIl4u3tbbLt3LlzAkD27NmjbLt165Z4eHjI6tWr/7L+0NBQGTFihPK4VatW8v777+f590eNGiVly5aVkydP/uVziYisWbNGfH19n3oMEZG5cQSYiKgQ1atXD23btkVQUBC6deuGBQsWID09HcCjEdkLFy5g3759AB59eK179+7w9PQEAJw+fRotWrQweb4WLVrg/PnzMBgMquo4ffo0XFxc0LRpU2Wbr68vatasidOnTwMADAYDJk+ejKCgIJQsWRLFihXD9u3bkZSU9NTnnjlzJhYsWIDdu3fj+eefN9n33//+F23btkX58uVRvHhx9OnTB6mpqbh//76q+omIngUbYCKiQuTs7Izo6Ghs3boVtWvXxty5c1GzZk0kJibCz88Pr776KpYsWYIbN25g69atyvQHa/jss88wZ84cjB49Gjt27MDRo0cREhKC7Ozsp35fy5YtYTAYsHr1apPtly9fRqdOnVC3bl2sXbsW8fHxmDdvHgD87XMSEZkTG2AiokKm0+nQokULTJw4EUeOHIGbmxt+/vlnAMCgQYOwatUqfPvtt6hWrZrJiG9gYCD27Nlj8lx79uxBjRo14OzsDABwc3PLMxr8pG2BgYHIycnB/v37lW2pqak4e/YsateurTx3ly5d0Lt3b9SrVw9Vq1bFuXPn/jZfkyZNsHXrVkydOhWff/65sj0+Ph5GoxEzZ85Es2bNUKNGDVy7di0/PzIiIrNiA0xEVIj279+PqVOn4tChQ0hKSsK6detw8+ZNBAYGAgBCQkLg5eWFKVOmoH///ibfO2LECMTExGDy5Mk4d+4cli1bhq+++kr5UBvwaB3gXbt24ffff8etW7eUbXfv3kVMTAxu3bqF+/fvIyAgAF26dMHgwYOxe/duHDt2DL1790b58uXRpUsXAEBAQACio6Oxd+9enD59Gm+//TZu3LiRr5zNmzfHli1bMHHiROXGGNWrV4der8fcuXNx6dIlLF++HPPnz3/WHykRkXrWnoRMRORIEhISJCQkREqXLi3u7u5So0YNmTt3rskx48aNE2dnZ7l27Vqe7//pp5+kdu3a4urqKs8995x89tlnJvvj4uKkbt264u7uLo//in/nnXfE19dXAMjHH38sIiJpaWnSp08f8fb2Fg8PDwkJCZFz584p35OamipdunSRYsWKiZ+fn3z00UfSt29f6dKli3LMX30ILldsbKx4enrKl19+KSIis2bNkrJlyyr/3nfffScAJD09XeVPkoio4HQiIlbuwYmI6DEDBw7EzZs3sWHDBmuXQkSkSbwTHBGRjcjIyMCJEyewcuVKNr9ERBbEBpiIyEZ06dIFBw4cwDvvvIP27dtbuxwiIs3iFAgiIiIicihcBYKIiIiIHAobYCIiIiJyKGyAiYiIiMihsAEmIiIiIofCBpiIiIiIHAobYCIiIiJyKGyAiYiIiMihsAEmIiIiIofyfwXsP2I2rIRlAAAAAElFTkSuQmCC", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "plt.figure(figsize=(8, 5))\n", - "plot_scores(df_fi)\n", - "plt.xlabel(\"syöttöaika\")\n", - "plt.ylabel(\"mittarin arvo\")\n", - "# set start of time range to 2020-09-01\n", - "plt.xlim(datetime.datetime(2020, 8, 1, tzinfo=datetime.timezone.utc))\n", - "plt.ylim(0.34, 1)\n", - "\n", - "# hide legend labels of vertical lines and change other labels to finnish\n", - "plt.legend([\"_\", \"_\", \"_\", \"_\", \"_\", \"_\", \"tarkkuus\", \"saanti\", \"F1-arvo\"]) #, loc='center right')\n", - "# set x ticks to every 3 months starting from 2020-10, ticks to start of month\n", - "plt.gca().set_xticks(pd.date_range(\"2020-10-01\", \"2023-03-31\", freq=\"3MS\"))\n", - "# format x ticks to month-year\n", - "plt.gca().xaxis.set_major_formatter(matplotlib.dates.DateFormatter(\"%Y-%m\"))\n", - "plt.savefig('kuva-1.svg', format='svg', bbox_inches='tight')\n", - "\n", - "plt.figure(figsize=(8, 3))\n", - "plot_counts(df_fi)\n", - "plt.xlabel(\"syöttöaika\")\n", - "plt.ylabel(\"dokumenttien määrä\")\n", - "plt.yticks([0, 2000, 4000, ])\n", - "# set x ticks to every 3 months starting from 2020-10, ticks to start of month\n", - "plt.gca().set_xticks(pd.date_range(\"2020-10-01\", \"2023-03-31\", freq=\"3MS\"))\n", - "# format x ticks to month-year\n", - "plt.gca().xaxis.set_major_formatter(matplotlib.dates.DateFormatter(\"%Y-%m\"))\n", - "# rotate x tick labels\n", - "_ = plt.setp(plt.gca().get_xticklabels(), rotation=45, ha=\"right\", rotation_mode=\"anchor\")\n", - "plt.gca().set_axisbelow(True)\n", - "plt.savefig('kuva-2.svg', format='svg', bbox_inches='tight')" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/3516352905.py:2: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n", - " df_fi.groupby(\"annif_update\")['precision', 'recall', 'f1 score'].agg(['mean', 'count']).style.set_caption(title_str)\n" - ] - }, - { - "data": { - "text/html": [ - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
Theseus suomi
 precisionrecallf1 score
 meancountmeancountmeancount
annif_update      
2020-03-26 00:00:00+00:000.42228640840.84579040840.5400594084
2020-12-09 00:00:00+00:000.42528752670.85786652670.5446065267
2021-04-27 00:00:00+00:000.43937188920.87736188920.5607568892
2021-11-11 00:00:00+00:000.484445159730.907410159730.60577015973
2022-06-21 00:00:00+00:000.51008926960.91523926960.6267862696
2022-11-22 00:00:00+00:000.50514272230.91476472230.6233267223
\n" - ], - "text/plain": [ - "" - ] - }, - "execution_count": 8, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# Group by annif_update and give mean precision, recall, f1\n", - "df_fi.groupby(\"annif_update\")['precision', 'recall', 'f1 score'].agg(['mean', 'count']).style.set_caption(title_str)" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Distributions\n", - "Number of records with each score value" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.000000 431\n", - "0.100000 1681\n", - "0.111111 7\n", - "0.125000 1\n", - "0.142857 3\n", - "0.166667 1\n", - "0.200000 4675\n", - "0.222222 6\n", - "0.250000 2\n", - "0.285714 1\n", - "0.300000 8535\n", - "0.333333 11\n", - "0.375000 2\n", - "0.400000 8371\n", - "0.428571 2\n", - "0.444444 8\n", - "0.500000 6567\n", - "0.555556 4\n", - "0.571429 1\n", - "0.600000 4752\n", - "0.625000 4\n", - "0.666667 2\n", - "0.700000 3328\n", - "0.750000 2\n", - "0.777778 2\n", - "0.800000 2249\n", - "0.833333 1\n", - "0.857143 1\n", - "0.875000 2\n", - "0.900000 1507\n", - "1.000000 1976\n", - "Name: precision, dtype: int64" - ] - }, - "execution_count": 9, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi['precision'].value_counts(sort=False).sort_index()" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.000000 431\n", - "0.100000 1\n", - "0.111111 1\n", - "0.125000 6\n", - "0.142857 9\n", - " ... \n", - "0.875000 374\n", - "0.888889 255\n", - "0.900000 170\n", - "0.909091 140\n", - "1.000000 30541\n", - "Name: recall, Length: 81, dtype: int64" - ] - }, - "execution_count": 10, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi['recall'].value_counts(sort=False).sort_index()" - ] - }, - { - "cell_type": "code", - "execution_count": 42, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - " # 11 bins cover all most common values of precision\n", - "df_fi[['precision', 'recall']].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.gca().set_title(\"saanti\")\n", - "# plt.ylabel(\"dokumenttien määrä\")\n", - "plt.xlabel(\"mittarin arvo\")\n", - "plt.yticks([0, 10000, 20000, 30000])\n", - "# put grid behind bars\n", - "plt.gca().set_axisbelow(True)\n", - "\n", - "# select axis of first subplot\n", - "plt.sca(plt.gcf().axes[0])\n", - "plt.yticks([0, 4000, 8000, 12000])\n", - "plt.gca().set_title(\"tarkkuus\")\n", - "plt.ylabel(\"dokumenttien määrä\")\n", - "plt.xlabel(\"mittarin arvo\")\n", - "plt.gca().set_axisbelow(True)\n", - "plt.savefig('kuva-3.svg', format='svg', bbox_inches='tight')" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The histogram of recall above indicates that nearly never users assigns subjects that\n", - "are not suggested by Annif." - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.691990483743061" - ] - }, - "execution_count": 12, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# Records without self added subjects\n", - "len(df_fi[(df_fi.recall == 1.0)]) / len(df_fi)" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### How many Annif suggestions have been used without any manual corrections?" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.277778 1\n", - "0.294118 1\n", - "0.322581 1\n", - "0.333333 2\n", - "0.400000 1\n", - "0.416667 1\n", - "0.454545 1\n", - "0.476190 4\n", - "0.500000 3\n", - "0.526316 3\n", - "0.555556 5\n", - "0.588235 5\n", - "0.625000 6\n", - "0.666667 14\n", - "0.714286 15\n", - "0.769231 48\n", - "0.833333 82\n", - "0.909091 140\n", - "1.000000 1643\n", - "Name: recall, dtype: int64" - ] - }, - "execution_count": 13, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# select rows with precision = 1.0 and describe recall\n", - "df_fi[df_fi.precision == 1.0].recall.value_counts(sort=False).sort_index()" - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.03722669083493826" - ] - }, - "execution_count": 14, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# percentage of rows with precision = 1.0 and recall = 1.0\n", - "len(df_fi[(df_fi.precision == 1.0) & (df_fi.recall == 1.0)]) / len(df_fi)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### How many documents with all suggestions rejected?" - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.0 431\n", - "Name: precision, dtype: int64" - ] - }, - "execution_count": 15, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi[df_fi.recall == 0.0].precision.value_counts(sort=False).sort_index()" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.009765492239719044" - ] - }, - "execution_count": 16, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "len(df_fi[df_fi.recall == 0.0]) / len(df_fi)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "## How many suggestions before and after Annif integration?" - ] - }, - { - "cell_type": "code", - "execution_count": 17, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Error normalizing label 'None'\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/1982199275.py:50: SettingWithCopyWarning: \n", - "A value is trying to be set on a copy of a slice from a DataFrame.\n", - "Try using .loc[row_indexer,col_indexer] = value instead\n", - "\n", - "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", - " df[\"subjects_uris\"] = df[source_col].apply(lambda x: [label_to_yso_uris(label, yso, lang, COMPLAIN) for label in x])\n", - "/tmp/ipykernel_184606/1982199275.py:51: SettingWithCopyWarning: \n", - "A value is trying to be set on a copy of a slice from a DataFrame.\n", - "Try using .loc[row_indexer,col_indexer] = value instead\n", - "\n", - "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", - " df[\"subjects_uris\"] = df[\"subjects_uris\"].apply(lambda x: [str(item) for sublist in x for item in sublist])\n" - ] - }, - { - "data": { - "text/plain": [ - "59339" - ] - }, - "execution_count": 17, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi_before = df[((df.language == 'fin') | (df.language == 'fi')) & (df.date_accessioned < \"2020-09-01\") & (df.date_accessioned >= \"2016-01-01\")]\n", - "df_fi_before = add_subjects_uris(df_fi_before, 'fi', source_col='subjects_all')\n", - "df_fi_before = df_fi_before[df_fi_before.subjects_uris.apply(lambda x: len([e for e in x if e != '']) > 0)]\n", - "len(df_fi_before)" - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "count 59339.000000\n", - "mean 3.325334\n", - "std 2.126571\n", - "min 1.000000\n", - "25% 2.000000\n", - "50% 3.000000\n", - "75% 4.000000\n", - "max 64.000000\n", - "Name: subjects_uris, dtype: float64" - ] - }, - "execution_count": 18, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi_before.subjects_uris.apply(lambda x: len(x)).describe()" - ] - }, - { - "cell_type": "code", - "execution_count": 19, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "count 44135.000000\n", - "mean 5.263963\n", - "std 2.401788\n", - "min 1.000000\n", - "25% 4.000000\n", - "50% 5.000000\n", - "75% 7.000000\n", - "max 36.000000\n", - "Name: subjects_uris, dtype: float64" - ] - }, - "execution_count": 19, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi.subjects_uris.apply(lambda x: len(x)).describe()" - ] - }, - { - "cell_type": "code", - "execution_count": 20, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "([,\n", - " ,\n", - " ,\n", - " ,\n", - " ],\n", - " [Text(0, 0.0, '0.00'),\n", - " Text(0, 0.05, '0.05'),\n", - " Text(0, 0.1, '0.10'),\n", - " Text(0, 0.15, '0.15'),\n", - " Text(0, 0.2, '0.20')])" - ] - }, - "execution_count": 20, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "df_tmp = pd.DataFrame({\"before\": df_fi_before.subjects_uris.apply(lambda x: len(x)), \"after\": df_fi.subjects_uris.apply(lambda x: len(x))})\n", - "df_tmp = df_tmp.apply(pd.value_counts)\n", - "# normalize each column to sum to 1\n", - "df_tmp = df_tmp/ df_tmp.sum()\n", - "df_tmp.plot.bar()\n", - "\n", - "plt.xlim(-1, 12.5)\n", - "plt.ylabel(\"osuus\")\n", - "plt.xlabel(\"käsitteiden määrä\")\n", - "plt.legend([\"ennen\", \"jälkeen\"])\n", - "plt.yticks([0, 0.05, 0.1, 0.15, 0.2])\n", - "# plt.gca().xaxis.set_major_formatter(matplotlib.ticker.FormatStrFormatter(\"%d\"))" - ] - }, - { - "cell_type": "code", - "execution_count": 21, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n", - "Error normalizing label 'None'\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/1982199275.py:50: SettingWithCopyWarning: \n", - "A value is trying to be set on a copy of a slice from a DataFrame.\n", - "Try using .loc[row_indexer,col_indexer] = value instead\n", - "\n", - "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", - " df[\"subjects_uris\"] = df[source_col].apply(lambda x: [label_to_yso_uris(label, yso, lang, COMPLAIN) for label in x])\n", - "/tmp/ipykernel_184606/1982199275.py:51: SettingWithCopyWarning: \n", - "A value is trying to be set on a copy of a slice from a DataFrame.\n", - "Try using .loc[row_indexer,col_indexer] = value instead\n", - "\n", - "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", - " df[\"subjects_uris\"] = df[\"subjects_uris\"].apply(lambda x: [str(item) for sublist in x for item in sublist])\n" - ] - } - ], - "source": [ - "df_fi_both = add_subjects_uris(df[(df.language == 'fin') | (df.language == 'fi')], 'fi', source_col='subjects_all')[[\"subjects_uris\", \"date_accessioned\", \"suggestions\"]]" - ] - }, - { - "cell_type": "code", - "execution_count": 22, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "6686" - ] - }, - "execution_count": 22, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi_just_before = df_fi_before[(df_fi_before.date_accessioned >= \"2020-05-01\") & (df_fi_before.date_accessioned < \"2020-09-01\")]\n", - "len(df_fi_just_before)\n" - ] - }, - { - "cell_type": "code", - "execution_count": 23, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "" - ] - }, - "execution_count": 23, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "# plot number of subject_uris per document over time grouped by week\n", - "df_fi_both.groupby(pd.Grouper(key='date_accessioned', freq='W'))['subjects_uris'].apply(lambda x: x.apply(lambda y: len([e for e in y if e != ''])).mean()).plot(figsize=(8, 3))\n", - "df_fi_both.groupby(pd.Grouper(key='date_accessioned', freq='W'))['suggestions'].apply(lambda x: x.apply(lambda y: len([e for e in y if e != ''])).mean()).plot(figsize=(8, 3))\n", - "df_fi_just_before.groupby(pd.Grouper(key='date_accessioned', freq='W'))['subjects_uris'].apply(lambda x: x.apply(lambda y: len([e for e in y if e != ''])).mean()).plot(figsize=(8, 3))\n", - "df_fi.groupby(pd.Grouper(key='date_accessioned', freq='W'))['subjects_uris'].apply(lambda x: x.apply(lambda y: len([e for e in y if e != ''])).mean()).plot(figsize=(8, 3))\n", - "\n", - "plt.grid()\n", - "# draw vertical line at 2020-05-01\n", - "plt.axvline(x=\"2020-05-01\", color='black', linestyle='--')" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "For some reason the number of subjects per document started to increase before\n", - "introducing the suggestions, probably because instructions for users changed." - ] - }, - { - "cell_type": "code", - "execution_count": 24, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "count 6686.000000\n", - "mean 3.730033\n", - "std 1.944853\n", - "min 1.000000\n", - "25% 3.000000\n", - "50% 3.000000\n", - "75% 5.000000\n", - "max 64.000000\n", - "Name: subjects_uris, dtype: float64" - ] - }, - "execution_count": 24, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi_just_before.subjects_uris.apply(lambda x: len(x)).describe()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Results per degree programme" - ] - }, - { - "cell_type": "code", - "execution_count": 25, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "array(['fi=Rakennus- ja yhdyskuntatekniikka|sv=Byggnads- och samhällsteknik|en=Civil and Construction Engineering|',\n", - " 'fi=Liiketalous, hallinto ja markkinointi|sv=Företagsekonomi, förvaltning och marknadsföring|en=Business Management, Administration and Marketing|',\n", - " 'fi=Sosiaaliala|sv=Sociala området|en=Social Sciences|',\n", - " 'fi=Fysioterapia|sv=Fysioterapi|en=Physiotherapy|',\n", - " 'fi=Energia- ja ympäristötekniikka|sv=Energi- och miljöteknik|en=Energy and Enviromental Engineering|',\n", - " 'fi=Teknologiaosaamisen johtaminen|sv=Teknologibaserat ledarskap|en=Technology Management|',\n", - " 'fi=Automaatiotekniikka|sv=Automationsteknik|en=Automation Engineering|',\n", - " 'fi=Rakennusmestarit|sv=Byggmästare|en=Construction Managers|', '',\n", - " 'fi=Tietojenkäsittely|sv=Informationsbehandling|en=Business Information Technology|',\n", - " 'fi=Luonnonvara- ja ympäristöala|sv=Bioekonomi och Miljöbranschen|en=Natural Resources and Environment|',\n", - " 'fi=Sosiaali- ja terveysalan johtaminen|sv=Ledarskap inom social- och hälsovårdsbranschen|en=Management of Health and Welfare|',\n", - " 'fi=Tuotantotalous|sv=Produktionsekonomi|en=Industrial Management|',\n", - " 'fi=Poliisi|sv=Polis|en=Police|',\n", - " 'fi=Sähkötekniikka|sv=Elektroteknik|en=Electrical Engineering|',\n", - " 'fi=Hoitotyö|sv=Vård|en=Nursing|',\n", - " 'fi=Kulttuurituotanto|sv=Kulturproduktion|en=Cultural Management|',\n", - " 'fi=Turvallisuusala|sv=Säkerhetsbranschen|en=Security Management|',\n", - " 'fi=Ajoneuvo- ja kuljetustekniikka|sv=Fordons- och transportteknik|en=Vehicle and Transportation Engineering|',\n", - " 'fi=Terveydenhoitotyö|sv=Hälsovård|en=Public Health Nursing|',\n", - " 'fi=Merenkulku|sv=Sjöfart|en=Marine technology|',\n", - " 'fi=Majoitus- ja ravitsemisala|sv=Inkvarterings- och kosthållsbranschen|en=Hotel and Restaurant|',\n", - " 'fi=Esittävä taide ja musiikki|sv=Scenkonst och musik|en=Performance art and music|',\n", - " 'fi=Johdon assistenttityö ja kielet|sv=Ledningsassistentarbete och språk|en=Multilingual Management Assistant|',\n", - " 'fi=Tieto- ja viestintätekniikka|sv=Informations- och kommunikationsteknik|en=Information and Communications Technology|',\n", - " 'fi=Konetekniikka|sv=Maskinteknik|en=Mechanical Engineering|',\n", - " 'fi=Yhteisöpedagogiikka|sv=Samhällspedagogik|en=Community Education|',\n", - " 'fi=Bio- ja elintarviketekniikka|sv=Bio- och livsmedelsteknik|en=Biotechnology, Food Industry|',\n", - " 'fi=LVI-tekniikka|sv=VVS-teknik|en=Heating and Ventilation Engineering|',\n", - " 'fi=Liikunta-ala|sv=Idrottsbranschen|en=Sports studies|',\n", - " 'fi=Muotoilu|sv=Formgivning|en=Design|',\n", - " 'fi=Kauneudenhoitoala|sv=Skönhetsbranschen|en=Beauty and Cosmetics|',\n", - " 'fi=Laboratorioala|sv=Laboratoriebranschen|en=Laboratory Services|',\n", - " 'fi=Kasvatus- ja opetusala|sv=Pedagogiska området|en=Education|',\n", - " 'fi=Kemiantekniikka|sv=Kemi|en=Chemical Engineering|',\n", - " 'fi=Ensihoito ja akuutti hoitotyö|sv=Första- och akutvård|en=Emergency and Acute Care|',\n", - " 'fi=Logistiikka|sv=Logistik|en=Logistics|',\n", - " 'fi=Kirjasto- ja informaatiopalvelut|sv=Biblioteks- och informationstjänster|en=Library and Information Services|',\n", - " 'fi=Suuhygienia|sv=Tandhygieni|en=Dental Hygiene|',\n", - " 'fi=Prosessitekniikka|sv=Processteknik|en=Process Engineering|',\n", - " 'fi=Matkailu|sv=Turism|en=Tourism|',\n", - " 'fi=Maanmittaustekniikka|sv=Lantmäteriteknik|en=Surveying Technology|',\n", - " 'fi=Liikenneala|sv=Transport|en=Traffic|',\n", - " 'fi=Radiografia ja sädehoito|sv=Radiografi och strålbehandling|en=Radiography and Radiotherapy|',\n", - " 'fi=Vanhustyö|sv=Äldrevård|en=Elderly Care|',\n", - " 'fi=Palo- ja pelastusala|sv=Räddningsbranschen|en=Emergency Services|',\n", - " 'fi=Media-ala|sv=Mediebranschen|en=Media|',\n", - " 'fi=Kätilötyö|sv=Förlossningsvård|en=Midwifery|',\n", - " 'fi=Kuvataide|sv=Bildkonst|en=Fine arts|',\n", - " 'fi=Palvelumuotoilu|sv=Service Design|en=Service Design|',\n", - " 'fi=Toimintaterapia|sv=Ergoterapi|en=Occupational Therapy|',\n", - " 'fi=Materiaalitekniikka|sv=Materialteknik|en=Materials Engineering|',\n", - " 'fi=Paperi-, tekstiili- ja kemiantekniikka|sv=Pappers-, textil- och kemiateknik|en=Paper, Textiles, Chemistry|',\n", - " 'fi=Bioanalytiikka|sv=Bioanalytik|en=Bionalytics|',\n", - " 'fi=Rakennusarkkitehtuuri|sv=Byggnadsarkitektur|en=Construction Architecture|',\n", - " 'fi=Naprapatia|sv=Naprapatia|en=Naprapathy|',\n", - " 'fi=Kuntoutus|sv=Rehabilitering|en=Rehabilitation|',\n", - " 'fi=Jalkaterapia|sv=Fotterapi|en=Podiatry|',\n", - " 'fi=Projekti-ja myyntijohtaminen|sv=Projekt- och säljledning|en=project and sales management|',\n", - " 'fi=Hyvinvointiteknologia|sv=Välfärdsteknologi|en=Welfare technology|',\n", - " 'fi=Tulkkaus|sv=Tolkning|en=Interpreting|',\n", - " 'fi=Tekstiili- ja vaatetussuunnittelu|sv=Textil- och beklädnadsdesign|en=Textiles and Clothing Design|',\n", - " 'fi=Vaatetusala|sv=Konfektionbranschen|en=Textile and Clothing Industry|',\n", - " 'fi=Apuvälinetekniikka|sv=Hjälpmedelsteknik|en=Orthotics and Prosthetics|',\n", - " 'fi=Rikosseuraamusala|sv=Brottspåföljdsområdet|en=Correctional Services|',\n", - " 'fi=Optometria|sv=Optometri|en=Optometry|',\n", - " 'fi=Konservointi, restaurointi|sv=Konservering, restauration|en=Conservation, restoration|',\n", - " 'fi=Puunjalostustekniikka|sv=Trädförädlingsteknik|en=Wood Processing Technology|',\n", - " 'fi=Elektroniikka|sv=Elektronik|en=Electronic Engineering|',\n", - " 'fi=Toimitilapalvelut ja talousala|sv=Rengöringsservice och ekonomibranschen|en=Facilities Management|',\n", - " 'fi=Hammastekniikka|sv=Tandteknik|en=Dental Technology|',\n", - " 'fi=Osteopatia|sv=Osteopathy|en=Osteopathy|'], dtype=object)" - ] - }, - "execution_count": 25, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi.degreeprogram.unique()" - ] - }, - { - "cell_type": "code", - "execution_count": 26, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "array(['publication',\n", - " \"fi=Publisher's version|sv=Publisher's version|en=Publisher's version|\"],\n", - " dtype=object)" - ] - }, - "execution_count": 26, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_fi[df_fi.degreeprogram == ''].type.unique()" - ] - }, - { - "cell_type": "code", - "execution_count": 27, - "metadata": {}, - "outputs": [], - "source": [ - "# assign special value to empty degreeprogram\n", - "df_fi.loc[df_fi.degreeprogram == '', 'degreeprogram'] = '(Muu julkaisu)'" - ] - }, - { - "cell_type": "code", - "execution_count": 28, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/2085424453.py:2: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n", - " df_per_dp = df_fi.groupby(\"degreeprogram\")['precision', 'recall', 'f1 score'].agg(['mean', 'count']).sort_values(by=('f1 score', 'mean'), ascending=False)\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
precisionrecallf1 score
meancountmeancountmeancount
degreeprogram
Esittävä taide ja musiikki0.5675495670.8581925670.660578567
(Muu julkaisu)0.5754339050.8150699050.648030905
Kulttuurituotanto0.5380254050.8619024050.638240405
Liikunta-ala0.5211274260.9054434260.633980426
Poliisi0.5053196580.9198946580.626285658
Yhteisöpedagogiikka0.5175746430.8657376430.623167643
Fysioterapia0.5085198100.8790628100.621604810
Terveydenhoitotyö0.50113810940.89759510940.6168521094
Muotoilu0.5122156140.8526756140.614132614
Hoitotyö0.48623735930.89893635930.6073163593
Majoitus- ja ravitsemisala0.47890810620.90567210620.6033651062
Media-ala0.5004978050.8329548050.600034805
Sosiaaliala0.48154739570.88458439570.5986563957
Matkailu0.4731786860.8984096860.596961686
Sosiaali- ja terveysalan johtaminen0.47163513160.89371713160.5932291316
Energia- ja ympäristötekniikka0.4666288890.9006898890.589380889
Luonnonvara- ja ympäristöala0.46738110270.89029010270.5883951027
Teknologiaosaamisen johtaminen0.4609774140.9041284140.583037414
Liiketalous, hallinto ja markkinointi0.45548392900.90887992900.5817019290
Rakennusmestarit0.4519486160.9248616160.578945616
Rakennus- ja yhdyskuntatekniikka0.44851318870.90945818870.5746341887
Sähkötekniikka0.4349768930.9213818930.566544893
Konetekniikka0.42838615410.91158915410.5574221541
Automaatiotekniikka0.4147775820.8947215820.544105582
Tietojenkäsittely0.40044517730.86368817730.5247091773
Tieto- ja viestintätekniikka0.38768015320.85126415320.5089181532
\n", - "
" - ], - "text/plain": [ - " precision recall \\\n", - " mean count mean count \n", - "degreeprogram \n", - "Esittävä taide ja musiikki 0.567549 567 0.858192 567 \n", - "(Muu julkaisu) 0.575433 905 0.815069 905 \n", - "Kulttuurituotanto 0.538025 405 0.861902 405 \n", - "Liikunta-ala 0.521127 426 0.905443 426 \n", - "Poliisi 0.505319 658 0.919894 658 \n", - "Yhteisöpedagogiikka 0.517574 643 0.865737 643 \n", - "Fysioterapia 0.508519 810 0.879062 810 \n", - "Terveydenhoitotyö 0.501138 1094 0.897595 1094 \n", - "Muotoilu 0.512215 614 0.852675 614 \n", - "Hoitotyö 0.486237 3593 0.898936 3593 \n", - "Majoitus- ja ravitsemisala 0.478908 1062 0.905672 1062 \n", - "Media-ala 0.500497 805 0.832954 805 \n", - "Sosiaaliala 0.481547 3957 0.884584 3957 \n", - "Matkailu 0.473178 686 0.898409 686 \n", - "Sosiaali- ja terveysalan johtaminen 0.471635 1316 0.893717 1316 \n", - "Energia- ja ympäristötekniikka 0.466628 889 0.900689 889 \n", - "Luonnonvara- ja ympäristöala 0.467381 1027 0.890290 1027 \n", - "Teknologiaosaamisen johtaminen 0.460977 414 0.904128 414 \n", - "Liiketalous, hallinto ja markkinointi 0.455483 9290 0.908879 9290 \n", - "Rakennusmestarit 0.451948 616 0.924861 616 \n", - "Rakennus- ja yhdyskuntatekniikka 0.448513 1887 0.909458 1887 \n", - "Sähkötekniikka 0.434976 893 0.921381 893 \n", - "Konetekniikka 0.428386 1541 0.911589 1541 \n", - "Automaatiotekniikka 0.414777 582 0.894721 582 \n", - "Tietojenkäsittely 0.400445 1773 0.863688 1773 \n", - "Tieto- ja viestintätekniikka 0.387680 1532 0.851264 1532 \n", - "\n", - " f1 score \n", - " mean count \n", - "degreeprogram \n", - "Esittävä taide ja musiikki 0.660578 567 \n", - "(Muu julkaisu) 0.648030 905 \n", - "Kulttuurituotanto 0.638240 405 \n", - "Liikunta-ala 0.633980 426 \n", - "Poliisi 0.626285 658 \n", - "Yhteisöpedagogiikka 0.623167 643 \n", - "Fysioterapia 0.621604 810 \n", - "Terveydenhoitotyö 0.616852 1094 \n", - "Muotoilu 0.614132 614 \n", - "Hoitotyö 0.607316 3593 \n", - "Majoitus- ja ravitsemisala 0.603365 1062 \n", - "Media-ala 0.600034 805 \n", - "Sosiaaliala 0.598656 3957 \n", - "Matkailu 0.596961 686 \n", - "Sosiaali- ja terveysalan johtaminen 0.593229 1316 \n", - "Energia- ja ympäristötekniikka 0.589380 889 \n", - "Luonnonvara- ja ympäristöala 0.588395 1027 \n", - "Teknologiaosaamisen johtaminen 0.583037 414 \n", - "Liiketalous, hallinto ja markkinointi 0.581701 9290 \n", - "Rakennusmestarit 0.578945 616 \n", - "Rakennus- ja yhdyskuntatekniikka 0.574634 1887 \n", - "Sähkötekniikka 0.566544 893 \n", - "Konetekniikka 0.557422 1541 \n", - "Automaatiotekniikka 0.544105 582 \n", - "Tietojenkäsittely 0.524709 1773 \n", - "Tieto- ja viestintätekniikka 0.508918 1532 " - ] - }, - "execution_count": 28, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "# mean precision, recall, f1 score for each degree program and sort by f1 score\n", - "df_per_dp = df_fi.groupby(\"degreeprogram\")['precision', 'recall', 'f1 score'].agg(['mean', 'count']).sort_values(by=('f1 score', 'mean'), ascending=False)\n", - "# rename degree program entries to take only the first part of the name\n", - "df_per_dp.index = df_per_dp.index.map(lambda x: x.split(\"|\")[0].lstrip(\"fi=\"))\n", - "\n", - "# copy to clipboard\n", - "df_per_dp[df_per_dp.precision['count'] > 400].to_clipboard()\n", - "df_per_dp[df_per_dp.precision['count'] > 400]" - ] - }, - { - "cell_type": "code", - "execution_count": 41, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/2215873696.py:4: FutureWarning: The pandas.np module is deprecated and will be removed from pandas in a future version. Import numpy directly instead.\n", - " plt.xticks(pd.np.arange(0.5, 0.75, 0.05))\n" - ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "df_per_dp[df_per_dp.precision['count'] > 500]['f1 score']['mean'].plot.barh(figsize=(6, 6))\n", - "plt.ylabel(None)\n", - "plt.xlim(0.5, 0.7)\n", - "plt.xticks(pd.np.arange(0.5, 0.75, 0.05))\n", - "plt.xlabel(\"F1-arvo\")\n", - "plt.grid(axis='x')\n", - "plt.gca().set_axisbelow(True)\n", - "plt.savefig('kuva-4.svg', format='svg', bbox_inches='tight')" - ] - }, - { - "cell_type": "code", - "execution_count": 30, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "df_per_dp[df_per_dp.precision['count'] > 500]['precision']['mean'].plot.barh(figsize=(6, 6))\n", - "plt.ylabel(None)\n", - "plt.xlim(0.3, 0.6)\n", - "# plt.xticks(pd.np.arange(0.5, 0.75, 0.05))\n", - "plt.xlabel(\"precision\")\n", - "plt.grid(axis='x')\n", - "plt.gca().set_axisbelow(True)" - ] - }, - { - "cell_type": "code", - "execution_count": 31, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "df_per_dp[df_per_dp.precision['count'] > 500]['recall']['mean'].plot.barh(figsize=(6, 6))\n", - "plt.ylabel(None)\n", - "plt.xlim(0.7, 1)\n", - "# plt.xticks(pd.np.arange(0.5, 0.75, 0.05))\n", - "plt.xlabel(\"recall\")\n", - "plt.grid(axis='x')\n", - "plt.gca().set_axisbelow(True)" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### (Distribution of precision and recall of some degree programs)" - ] - }, - { - "cell_type": "code", - "execution_count": 32, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "Text(0.5, 0.98, '(Muu julkaisu) precision mean: 0.57 recall mean: 0.81')" - ] - }, - "execution_count": 32, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAABCsAAAEwCAYAAACe4vuuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACAsklEQVR4nO3deVhU1f8H8PewzLCDqGyCSLjvhoW4pwgumeYW6lfRTE2xfmqpWW64llpZ5pJlauVSmpqZC7hvuCZlbqm55ALmBigwDDPn9wfNjWEGmIEZGKf363l4dO6ce+75zNx75s5n7j1HJoQQICIiIiIiIiKyEnbl3QAiIiIiIiIiovyYrCAiIiIiIiIiq8JkBRERERERERFZFSYriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhARERERERGRVWGygoiIiIiIiIisCpMVRERERERERGRVTE5WnDhxAs2bN4erqytkMhmSk5Mxbdo0yGQyS7TPoH379kEmk2Hfvn1lts382rZti7Zt21p0G9rX9N69e0WWu3fvHmQyGaZNmwYAqFatGl588cUSbTN/PZZ27do1yGQyrFy5ssTrltf7bympqano1asXKlasCJlMhgULFpR3k8xm0KBBqFatWnk3A23btkX9+vXNWqdMJsOoUaPMWmdpVatWDYMGDZIeG+ozy+o9mTZtmlW899amNH2grfjmm29Qu3ZtODo6wsvLq7ybYzb/lfe2PPu+sjgPM9bTfE6ycuVKyGQyXLt2zaT1yvJ8kchYhvZna+orqGRMSlaoVCr07t0bDx48wMcff4xvvvkGwcHBZmnIoEGDIJPJiv3LfwJeEkeOHMG0adPw6NEjs7S7vLm7u+Obb75Bjx49AAALFizA22+/bZa6i3qtZs+ejc2bN5tlO7bC2ARTYcaMGYOdO3di4sSJ+Oabb9CxY0czt9Cybt++jWnTpiE5Obm8m0JktLZt2+p8xsjlcoSEhGDYsGH466+/yrt5Vq00X1YvXLiAQYMGITQ0FF988QWWLVtm5tZZ3po1a6wmqaxNSBrzR0TWT6lUYsKECQgICICzszPCw8ORmJhYoro6dOhQZH+9fPly1KlTB05OTqhRowYWLlxYmqaTlTt//jw6duwINzc3eHt7Y8CAAfj777+NWjc7Oxtz5sxB3bp14eLigipVqqB37944e/aswfK7du1Cu3bt4OnpCXd3d4SFheG7774zqb0OphS+cuUKrl+/ji+++AKvvfaatHzSpEl45513TNpwQcOHD0dkZKT0+OrVq5gyZQqGDRuGVq1aSctDQ0MRHh6OrKwsyOVyk7dz5MgRxMfHY9CgQSX+JSchIaFE61mCQqHA//73P+lx9+7dS1xXVlYWHBz+3SWKeq1mz56NXr16lWp7pGvPnj3o1q2b2ZJNZe327duIj49HtWrV0LhxY53nvvjiC2g0mvJpGBnE9+RfgYGBmDNnDgAgJycH586dw9KlS7Fz506cP38eLi4uZt9mcHAwsrKy4OjoaPa6nwb79u2DRqPBJ598gurVq5d3c0pkzZo1+P333zF69Gid5eXx3tapUwfffPONzrKJEyfCzc0N7733Xpm1g54uAwYMQExMDBQKhUnrFTxfJPMbNGgQNmzYgNGjR6NGjRpYuXIlOnfujL1796Jly5ZG17Nx40YkJSUV+vznn3+O119/HT179sTYsWNx8OBBvPnmm8jMzMSECRPMEQpZkZs3b6J169bw9PTE7Nmz8fjxY8yfPx9nzpzB8ePHi/1u3b9/f2zZsgVDhw7Fs88+i9u3b2PRokWIiIjAmTNndC5iWLFiBYYMGYIOHTpg9uzZsLe3x8WLF03+Iciknubu3bsAoPfF1cHBodSdVkREBCIiIqTHJ0+exJQpUxAREaHzZVzLycmpVNsrjZIkSZ4G5fmaUt7xZc5LobOzsyGXy2FnV/5D0/xXv5BZM74n//L09NT7nAkJCcGoUaNw+PBhdOjQwezblMlk/+k+t7DzidLIzMy0SGLJVOXx3vr6+urtw++//z4qVapk8ByKnj5PnjyBq6urWeu0t7eHvb29yev9l/uusnD8+HGsW7cO8+bNk37AGjhwIOrXr4/x48fjyJEjRtWTnZ2Nt956CxMmTMCUKVP0ns/KysJ7772HLl26YMOGDQCAoUOHQqPRYMaMGRg2bBgqVKhgUtstsZ+S+cyePRtPnjzBqVOnULVqVQDA888/jw4dOmDlypUYNmxYoeveunULGzduxNtvv4158+ZJy1u1aoV27dph48aNGDNmDIC82+Pi4uLwxhtv4JNPPilVm43+FjNo0CC0adMGANC7d2/IZDLpHiBrGbPi2LFj6NixIzw9PeHi4oI2bdrg8OHD0vPTpk3DuHHjAOSdiGovidTe25Sbm4sZM2YgNDQUCoUC1apVw7vvvgulUqmzHUP3PymVSkydOhXVq1eHQqFAUFAQxo8fr7eu9jKszZs3o379+lAoFKhXrx527NhRbNzXr19H9erVUb9+faSmpgLIy1q1a9cOPj4+UCgUqFu3LpYsWaK37smTJxEdHY1KlSrB2dkZISEhePXVV/Xapr0HsajXSiaT4cmTJ1i1apXB23Nu3bqFV199Fb6+vlJ8X331VZGxrVixAjKZDKdPn9Z7TpuNu3XrVpF1rFu3DmFhYXB3d4eHhwcaNGggHSAnT56ETCbDqlWr9NbbuXMnZDIZtm7dWmT9JaEdI+HcuXN44YUXpEum5s6dK5XR3mMnhMCiRYv0LtX9888/0bt3b3h7e8PFxQXNmjXDzz//rLMd7TGxbt06TJo0CVWqVIGLiwvS09MxaNAguLm54caNG3jxxRfh5uaGKlWqYNGiRQCAM2fOoF27dnB1dUVwcDDWrFmjU/eDBw/w9ttvo0GDBnBzc4OHhwc6deqEX3/9VWf7zz33HABg8ODBUgzae7YNjY/w5MkTvPXWWwgKCoJCoUCtWrUwf/58CCF0yhl7zGRkZGD06NGoVq0aFAoFfHx80KFDB/zyyy9678upU6fQvHlz6VhYunSpXhljj2mt4tp3/fp1jBw5ErVq1YKzszMqVqyI3r17690rrN0fDh8+jLFjx6Jy5cpwdXXFyy+/rHeZnhACM2fORGBgIFxcXPDCCy8UeileQQXfE+2xPX/+fCxbtkzqB5977jmcOHFCb/3169ejbt26cHJyQv369bFp0yaTxsEwpp/Q7tfff/89Zs2ahcDAQDg5OaF9+/a4fPmyUdspKT8/PwDQScQXFp+hz8DExES0bNkSXl5ecHNzQ61atfDuu+9Kzxsa10B7rN66dQvdu3eHm5sbKleujLfffhtqtVqnfo1GgwULFqBevXpwcnKCr68vhg8fjocPH+qUM6bvL6rvNIWx71e1atUwdepUAEDlypX17n9fvHgx6tWrB4VCgYCAAMTFxendjqjtW0+dOoXWrVvDxcUF7777rs5+vGjRIjzzzDNwcXFBVFQU/vrrLwghMGPGDAQGBsLZ2RndunXDgwcPdOr+8ccf0aVLFwQEBEChUCA0NBQzZszQeQ/atm2Ln3/+GdevX5f6O+2+UdiYFXv27EGrVq3g6uoKLy8vdOvWDefPn9cpo92XLl++LF3V6OnpicGDByMzM9Pk96SgR48eYfTo0VK/W716dXzwwQd6V1lpr3pp0KABnJycULlyZXTs2BEnT57Uq7O4vs+UmIw9DzPk7t27GDJkCHx9feHk5IRGjRoZ/My/f/8+BgwYAA8PD3h5eSE2Nha//vqrzntmjnMSQ7TjiiUkJKBx48ZwcnJC3bp1sXHjRp1y2s+B/fv3Y+TIkfDx8UFgYKD0/Pbt26V9yd3dHV26dDHY91+4cAF9+vRB5cqV4ezsjFq1aulcaWPoHn9Tzxe1Tp8+jU6dOsHDwwNubm5o3749jh49ajAuYz7fDCnt+Qxg/DEwf/58NG/eHBUrVoSzszPCwsKkL/QFXwtjz+svXLiAGzduFBvnhg0bYG9vr/PF0cnJCUOGDEFSUpLRv0zPnTsXGo2m0Ct29+7di/v372PkyJE6y+Pi4vDkyRO9c82CtMf2uXPn0K9fP1SoUEHnqo9vv/0WYWFhcHZ2hre3N2JiYgy2/dixY+jcuTMqVKgAV1dXNGzYUOcz6LfffsOgQYPwzDPPwMnJCX5+fnj11Vdx//59o14HY2jfR+25jbOzs3S1AJB3BUr16tXh5OSEtm3bGhznpbjvoYBlzgXT0tJw4cIFpKWlFRvnDz/8gBdffFFKVABAZGQkatasie+//77IdTMyMgDkJcfz8/f3BwA4OztLy5YuXQq1Wo3p06cDAB4/fqx3fm8soy+HGD58OKpUqYLZs2fjzTffxHPPPafX2PzS0tKgUqmKrdfJyQlubm7GNqNQe/bsQadOnRAWFoapU6fCzs5O+iJ/8OBBPP/88+jRowf++OMPrF27Fh9//DEqVaoEIO9kCQBee+01rFq1Cr169cJbb72FY8eOYc6cOTh//jw2bdpU6LY1Gg1eeuklHDp0CMOGDUOdOnVw5swZfPzxx/jjjz/0xnY4dOgQNm7ciJEjR8Ld3R2ffvopevbsiRs3bqBixYoGt3HlyhW0a9cO3t7eSExMlNq+ZMkS1K1bFy+99BIcHBzw008/YeTIkdBoNIiLiwOQ9wEeFRWFypUr45133oGXlxeuXbum9+GYX1Gv1TfffIPXXnsNzz//vNSRhoaGAsgbJLJZs2bSQV+5cmVs374dQ4YMQXp6ut7lslq9evVCXFwcVq9ejSZNmug8t3r1arRt2xZVqlQptL2JiYno27cv2rdvjw8++ABA3j1Zhw8fxv/93/+hadOmeOaZZ/D9998jNjZWZ93vvvsOFSpUQHR0NIC8sVmMOeABwNvbu9grFx4+fIiOHTuiR48e6NOnDzZs2IAJEyagQYMG6NSpE1q3bo1vvvkGAwYMQIcOHTBw4EBp3dTUVDRv3hyZmZl48803UbFiRaxatQovvfQSNmzYgJdffllnWzNmzIBcLsfbb78NpVIpXQWkVqulbc2dOxerV6/GqFGj4Orqivfeew/9+/dHjx49sHTpUgwcOBAREREICQkBkJcs2bx5M3r37o2QkBCkpqbi888/R5s2bXDu3DkEBASgTp06mD59ut6tW82bNzf4mggh8NJLL2Hv3r0YMmQIGjdujJ07d2LcuHG4desWPv74Y53yxhwzr7/+OjZs2IBRo0ahbt26uH//Pg4dOoTz58/j2Wef1Xk/OnfujD59+qBv3774/vvvMWLECMjlcumEzBLH9IkTJ3DkyBHExMQgMDAQ165dw5IlS9C2bVucO3dO7xfhN954AxUqVMDUqVNx7do1LFiwAKNGjdK512/KlCmYOXMmOnfujM6dO+OXX35BVFQUcnJyitgji7ZmzRpkZGRg+PDhkMlkmDt3Lnr06IE///xTuhrj559/xiuvvIIGDRpgzpw5ePjwIYYMGVLkMZqfqf3E+++/Dzs7O7z99ttIS0vD3Llz0b9/fxw7dkwqk5mZadSXOXt7e71fitRqtTTWjEqlwvnz56VEVYsWLYyKKb+zZ8/ixRdfRMOGDTF9+nQoFApcvnxZ76TFELVajejoaISHh2P+/PnYtWsXPvzwQ4SGhmLEiBFSueHDh2PlypUYPHgw3nzzTVy9ehWfffYZTp8+jcOHD8PR0dGovr+4vrMkinu/FixYgK+//hqbNm3CkiVL4ObmhoYNGwLIO/mNj49HZGQkRowYgYsXL2LJkiU4ceKEFJfW/fv30alTJ8TExOB///ufzjnJ6tWrkZOTgzfeeAMPHjzA3Llz0adPH7Rr1w779u3DhAkTcPnyZSxcuBBvv/22TqJs5cqVcHNzw9ixY+Hm5oY9e/ZgypQpSE9Pl35Neu+995CWloabN29K/VVR5zK7du1Cp06d8Mwzz2DatGnIysrCwoUL0aJFC/zyyy96SbA+ffogJCQEc+bMwS+//IIvv/wSPj4+0ntUEpmZmWjTpg1u3bqF4cOHo2rVqjhy5AgmTpyIO3fu6Iy/MWTIEKxcuRKdOnXCa6+9htzcXBw8eBBHjx5F06ZNpXKmnM8YE1NJz8OysrLQtm1bXL58GaNGjUJISAjWr1+PQYMG4dGjR9K+rNFo0LVrVxw/fhwjRoxA7dq18eOPP+qdF5T2nKQoly5dwiuvvILXX38dsbGxWLFiBXr37o0dO3boXcU1cuRIVK5cGVOmTMGTJ08A5A1MGxsbi+joaHzwwQfIzMzEkiVL0LJlS5w+fVral3777Te0atUKjo6OGDZsGKpVq4YrV67gp59+wqxZswy2rSTni0Ben9eqVSt4eHhg/PjxcHR0xOeff462bdti//79CA8P1ylvzOdbYUpzPmPKMfDJJ5/gpZdeQv/+/ZGTk4N169ahd+/e2Lp1K7p06aLTJmOPgzp16qBNmzbFDsZ6+vRp1KxZEx4eHjrLn3/+eQBAcnIygoKCiqzjxo0beP/99/HVV1/pfIksuB0AOsc0AISFhcHOzg6nT5826sqs3r17o0aNGpg9e7b0hXTWrFmYPHky+vTpg9deew1///03Fi5ciNatW+P06dPSVXWJiYl48cUX4e/vj//7v/+Dn58fzp8/j61bt0rHbWJiIv78808MHjwYfn5+OHv2LJYtW4azZ8/i6NGjZvvB/ODBg9iyZYv0/WnOnDl48cUXMX78eCxevBgjR47Ew4cPMXfuXLz66qvYs2ePtK4x30MBy5wLbtq0CYMHD8aKFSuKHNvx1q1buHv3rt77DeTtW9u2bSvy9QkNDUVgYCA+/PBD1KpVC02aNMHt27cxfvx4hISEICYmRiq7a9cu1K5dG9u2bZPO7StUqIC4uDjEx8ebdtW3MMHevXsFALF+/Xqd5VOnThUFq2rTpo0AUOxfbGyswW2dOHFCABArVqwotB179+4VQgih0WhEjRo1RHR0tNBoNFK5zMxMERISIjp06CAtmzdvngAgrl69qlNncnKyACBee+01neVvv/22ACD27NmjE1ubNm2kx998842ws7MTBw8e1Fl36dKlAoA4fPiwtAyAkMvl4vLly9KyX3/9VQAQCxculJZpX9O///5bnD9/XgQEBIjnnntOPHjwQGcbT5480Xt9oqOjxTPPPCM93rRpkwAgTpw4oVc2PwBi6tSp0uPCXishhHB1dTX43g0ZMkT4+/uLe/fu6SyPiYkRnp6eIjMzUwghxNWrV/Xe3759+4qAgAChVqulZb/88oteOe262vdfCCH+7//+T3h4eIjc3NxC45s4caJwdHTUeQ2VSqXw8vISr776qrRMu38Z85f/tcn/nmlpj4Ovv/5aZ5t+fn6iZ8+eOu0DIOLi4nSWjR49WgDQ2bcyMjJESEiIqFatmvRaadv8zDPPSK+xVmxsrAAgZs+eLS17+PChcHZ2FjKZTKxbt05afuHCBb39IDs7W+c9ESLvPVAoFGL69OnSsqKO2djYWBEcHCw93rx5swAgZs6cqVOuV69eQiaT6Rwfxh4znp6eeq9fQdr348MPP5SWKZVK0bhxY+Hj4yNycnKEEJY5pgu+L0IIkZSUpLd/rFixQgAQkZGROv3ZmDFjhL29vXj06JEQQoi7d+8KuVwuunTpolPu3Xff1etbC/aZQui/J9rjqmLFijrHyI8//igAiJ9++kla1qBBAxEYGCgyMjKkZfv27RMAdOoUIu+4KLjM2H5C2+46deoIpVIplfvkk08EAHHmzBmd7RhzzBZsS2GfVXXq1BF//vmnTtmCr1nBbWt9/PHHen1BQYb6QO2xmv+4EkKIJk2aiLCwMOnxwYMHBQCxevVqnXI7duzQWW5M329M31mYgn1WSd6v/K+Rdp+OiorS6XM+++wzAUB89dVX0jLt+7Z06VKdNmlf18qVK0vHihB5/T8A0ahRI6FSqaTlffv2FXK5XGRnZ0vLDB2rw4cPFy4uLjrlunTpYnB/MPTeavuY+/fvS8t+/fVXYWdnJwYOHKj3uuT/TBJCiJdffllUrFhRb1tFqVevns65yowZM4Srq6v4448/dMq98847wt7eXty4cUMIIcSePXsEAPHmm2/q1Zm/rzH1fKa4mEpzHrZgwQIBQHz77bfSspycHBERESHc3NxEenq6EEKIH374QQAQCxYskMqp1WrRrl07s56TFCY4OFgAED/88IO0LC0tTfj7+4smTZpIy7SfAy1bttQ5NjMyMoSXl5cYOnSoTr0pKSnC09NTZ3nr1q2Fu7u7uH79uk7Z/O+hdjvac5mSni92795dyOVyceXKFWnZ7du3hbu7u2jdurXe9or7fCtMac9njD0GhNDvB3JyckT9+vVFu3bt9F4LY44Dbdn8+21h6tWrp7cdIYQ4e/aswX7PkF69eonmzZvrbLvgOVJcXJywt7c3uH7lypVFTExMkdvQHtt9+/bVWX7t2jVhb28vZs2apbP8zJkzwsHBQVqem5srQkJCRHBwsHj48KFO2YLf5wpau3atACAOHDggLSu4Pwuh31cUBoBQKBQ6637++ecCgPDz85P6ECH+/TzRljXle6i5zwXzlzV0/p2f9jw9/3a0xo0bJwDofMYZcuzYMREaGqpzvhQWFibu3LmjU87Dw0NUqFBBKBQKMXnyZLFhwwbRr18/AUC88847RW6jIIvdzP7hhx8iMTGx2L/x48eXelvJycm4dOkS+vXrh/v37+PevXu4d+8enjx5gvbt2+PAgQPFDiSnzSaNHTtWZ/lbb70FAEVeCrV+/XrUqVMHtWvXlrZ97949tGvXDkDeZVb5RUZGSlciAEDDhg3h4eGBP//8U6/u33//HW3atEG1atWwa9cuvV8E82fg0tLScO/ePbRp0wZ//vmndHWANnu5detWo652KSkhBH744Qd07doVQgid1yI6OhppaWkGL8nXGjhwIG7fvq3zeq1evRrOzs7o2bNnkdv28vLCkydPihwp+ZVXXoFKpdL5hSAhIQGPHj3CK6+8Ii1r1KiRUftuYmKidKl4Udzc3HQy03K5HM8//7zB97ugbdu24fnnn9e5rM7NzQ3Dhg3DtWvXcO7cOZ3ysbGxhWbQ8w+K6+XlhVq1asHV1RV9+vSRlteqVQteXl46bVMoFFIGVK1W4/79+9Jl7UW9n8XFZW9vjzfffFNn+VtvvQUhBLZv366z3JhjxsvLC8eOHcPt27eL3LaDgwOGDx8uPZbL5Rg+fDju3r2LU6dOAbDMMZ3/fVGpVLh//z6qV68OLy8vg6/jsGHDdH4taNWqFdRqNa5fvw4gL2ut/eU4f7nCrl4y1iuvvKLTz2ivktHGcvv2bZw5cwYDBw7U+SW5TZs2aNCgQbH1l6SfGDx4sM5YQQXbBOT1H8Ycs6tXr9ZrU7Vq1aTnt2/fjgULFiAtLQ2dOnUyeoTs/LR97o8//liiQUxff/11ncetWrXSiXX9+vXw9PREhw4ddF6/sLAwuLm5SfunMX2/MX2nqYx5vwzR7tOjR4/W+dVl6NCh8PDw0PscVigUGDx4sMG6evfuDU9PT+mx9pfd//3vfzq39oSHhyMnJ0fnkv78x2pGRgbu3buHVq1aITMzExcuXCgyBkPu3LmD5ORkDBo0CN7e3tLyhg0bokOHDgZ/zTK0D9y/fx/p6ekmb19r/fr1aNWqFSpUqKCz30RGRkKtVuPAgQMA8i4Tlslk0q06+RX8BdOU85niYirNedi2bdvg5+eHvn37SsscHR3x5ptv4vHjx9i/fz8AYMeOHXB0dMTQoUOlcnZ2dtIvqfmV5pykKAEBATpXRXp4eGDgwIE4ffo0UlJSdMoOHTpUZ0yJxMREPHr0CH379tV5D+3t7REeHi619e+//8aBAwfw6quv6lzuDei/h/mV5HxRrVYjISEB3bt3xzPPPCMt9/f3R79+/XDo0CG9/ba4z7filPR8xthjANDtBx4+fIi0tDS0atXK4Oe1sceBEMKoKW6zsrIMDnqqHSskKyuryPX37t2LH374odjZioqarMDJyanY7WgVPLY3btwIjUaDPn366LzOfn5+qFGjhrSfnj59GlevXsXo0aP1xi/Kv3/kfy+ys7Nx7949NGvWDABKfB5qSPv27XWuctN+bvTs2RPu7u56y7XvrynfQ819Lgjk3R4lhCh2xkzt+1mafatChQpo3Lgx3nnnHWzevBnz58/HtWvX0Lt3b2RnZ0vlHj9+jIcPHyI+Ph7Tp09Hz549sXr1anTs2BGffPKJdEuJMSw2lG9YWJilqtZz6dIlANC7jC+/tLS0IgeJuX79Ouzs7PRGJffz84OXl1eRHeilS5dw/vx56XaSgrQDiWkV/OAA8t78gvcaA0DXrl3h6+uLnTt3GrzE9PDhw5g6dSqSkpL0LoFOS0uDp6cn2rRpg549eyI+Ph4ff/wx2rZti+7du6Nfv34mjwBdlL///huPHj3CsmXLCp2GruBrkV+HDh3g7++P1atXo3379tBoNFi7di26deum00kYMnLkSHz//ffo1KkTqlSpgqioKPTp00dn+s9GjRqhdu3a+O677zBkyBAAebeAVKpUSfoSCuS9F/lnpimtwMBAvZODChUq4Lfffit23evXr+tdPgnkXUqofb5+/frScu2ljgVp7znOz9PT02DbPD09dfZF7b3LixcvxtWrV3Xu2y7stqXiXL9+HQEBAXrva/648jPmmJk7dy5iY2MRFBSEsLAwdO7cGQMHDtQ5eQLyThQLDv5Us2ZNAHn3mjdr1swix3RWVhbmzJmDFStW4NatWzr37hm67ahgndr+S1un9jWqUaOGTrnKlSubPCBWSbZraAaH6tWrF3viUJJ+org2AcAzzzyj914by9XVVeeY79ixI1q2bImmTZvi/fffx4cffmhSfa+88gq+/PJLvPbaa3jnnXfQvn179OjRA7169Sr20kdDx2rBfenSpUtIS0uDj4+PwTq0r58xfb8xfaepjHm/DNHuW7Vq1dJZLpfL8cwzz+j1C1WqVCn0RLtgG7SJi4KXTmuX52/b2bNnMWnSJOzZs0fvS5axtwjmV1hcQF6ft3PnTr1B6Yp6DQteGm6sS5cu4bfffiu2X7ty5QoCAgJ0EiuFMeV8priYSnMedv36ddSoUUPv+Cr4mXL9+nX4+/vrXWptqD8rzTlJUapXr673uZv/Myj/jyAFP9O157v5z1ny0+4b2i9R+c8PjFGS88W///4bmZmZhe7fGo0Gf/31F+rVqyctL2kfAZTufMbYYwDIS9jMnDkTycnJOmOmGEr2mHIcGMPZ2dngOC3aL4OF/SgF5I378uabb2LAgAHSWGJFbaew20azs7OL3E5+hvZTIYTe+YmW9na+K1euACh+P33w4AHi4+Oxbt06vfODkvTJhSnp54Yp30PNfS5oCu37WdJ9S5uwGzdunJREBvJuI2rbti1WrFgh3a7q7OyMJ0+e6CSQAaBv377YsWMHTp8+jdatWxvVboslKx48eGDUfdPOzs46v36UhDZbNW/ePL0pE7WMHRejJPc9aTQaNGjQAB999JHB5wvu5IWNvJx/h9Xq2bMnVq1ahdWrV+v8GgzkHeTt27dH7dq18dFHHyEoKAhyuRzbtm3Dxx9/LL0uMpkMGzZswNGjR/HTTz9h586dePXVV/Hhhx/i6NGjZhkzBPj3ffjf//5X6AGrvS/ZEHt7e/Tr1w9ffPEFFi9ejMOHD+P27dtG3S/n4+OD5ORk7Ny5E9u3b8f27duxYsUKDBw4UGeArVdeeQWzZs3CvXv34O7uji1btqBv3746v7Tl5OToDbhWmMqVKxc7krYp73dpFdbJFNYGY9o2e/ZsTJ48Ga+++ipmzJghjdMxevToMpv60ph29unTB61atcKmTZuQkJCAefPm4YMPPsDGjRvRqVMnk7ZniWP6jTfewIoVKzB69GhERETA09MTMpkMMTExBl/HstxvynK7JeknjGnT48eP8fjx42K3b29vX+iJan5hYWHw9PTU+aWtsM+HgoNfOjs748CBA9i7dy9+/vln7NixA9999x3atWuHhISEIvsMY0bm12g08PHxMXiVCPDvOEzG9P3G9p2mKKt9t6iTqpL2eY8ePUKbNm3g4eGB6dOnIzQ0FE5OTvjll18wYcIEq+rzTKXRaNChQ4dCr2jVfmE2hSntNLZsWQ7YXpTSnJOYS8F9XLv/ffPNNwav7CztzHxldb5Ymv27NOczxh4DBw8exEsvvYTWrVtj8eLF8Pf3h6OjI1asWGFw0E5zH6/+/v4GB3C9c+cOgLwfXQrz9ddf4+LFi/j888/1Bm3MyMjAtWvX4OPjAxcXF/j7+0OtVuPu3bs6ye+cnBzcv3+/yO3kZ2g/lclk2L59u8HXxtT9qE+fPjhy5AjGjRuHxo0bw83NDRqNBh07djRrn1zSfcuU76HleS6oHQhTux/ld+fOHXh7exf5I/YPP/yA1NRUvPTSSzrLtZ+Zhw8flpIVAQEBuHTpkt74ltr9zJRki8WSFT169JAuuytKbGys3ojZptJeeuXh4VHsL+KFfQgGBwdDo9Hg0qVLUiYeyBsI7tGjRzrzxhra/q+//or27dub/UN23rx5cHBwkAbt6devn/TcTz/9BKVSiS1btuhk3gpeoq7VrFkzNGvWDLNmzcKaNWvQv39/rFu3TudyuvyKisXQc5UrV4a7uzvUanWJr0wYOHAgPvzwQ/z000/Yvn07KleuLA18WRy5XI6uXbuia9eu0Gg0GDlyJD7//HNMnjxZ+tXklVdeQXx8PH744Qf4+voiPT1dZ0AYADhy5AheeOEFo7Z59epVo2c/KIng4GBcvHhRb7n2UuSi9ktz2bBhA1544QUsX75cZ/mjR4+kgVcB004wg4ODsWvXLmRkZOj8QlXauPz9/TFy5EiMHDkSd+/exbPPPotZs2bpJCtu376t9yvmH3/8AQDSe2mJY3rDhg2IjY3V+ZU+Oztbb5YDY2lfo0uXLulcUfD333+X+NccU7ZraDYOY2boMEc/Ycj8+fMRHx9fbLng4GCDo3gbolardRIgFSpUMPh+GfrF187ODu3bt0f79u3x0UcfYfbs2Xjvvfewd+/eUscdGhqKXbt2oUWLFkb98lVc329M31kWtPvWxYsXdfbpnJwcXL161az7S2H27duH+/fvY+PGjTq/+ly9elWvrLF9Q/64Crpw4QIqVapUJlP9hYaG4vHjx8W+jqGhodi5cycePHhg1NUV5lKa87Dg4GD89ttv0Gg0OldXFPxMCQ4Oxt69e/WmuS2s7yrNOUlhLl++DCGEzv5T8DOoMNrzXR8fnyLfR+3x8/vvv5eojaacL1auXBkuLi6F7t92dnbFDgZZVow9Bn744Qc4OTlh586dOl/eVqxYYekmAgAaN26MvXv3Ij09XedKKu0gxYV9IQbyBtZUqVQGB4f++uuvpcGNu3fvLtVz8uRJdO7cWSp38uRJaDSaIrdTlNDQUAghEBISUmQSVLs///7774W+Jw8fPsTu3bsRHx+vM/2q9moGa2DK91BznwuaokqVKqhcubLBWZ2OHz9e7PutnYmy4A80Qgio1Wrk5uZKy8LCwnDp0iXcunVL5/Nce6u2MT8aadnEmBVhYWEIDQ3F/PnzDf6ylv+eY+0JQcGdQnuQFry/S/vLasGRf/Pr06cPbt26hS+++ELvuaysLGkE55KQyWRYtmwZevXqhdjYWGzZskV6TpttK3gJUcHO9OHDh3oZOO0OWdR0YIW9VtrnCi63t7dHz5498cMPPxj8gDTm3u+GDRuiYcOG+PLLL/HDDz8gJibGqF8KCk5fZGdnJ/06mz/GOnXqoEGDBvjuu+/w3Xffwd/fX+8yJHOPWVEanTt3xvHjx5GUlCQte/LkCZYtW4Zq1aqhbt26Ft0+kPe+Ftx/1q9fr5f1L2p/Kahz585Qq9X47LPPdJZ//PHHkMlkJl8JoVar9S6f8/HxQUBAgN4+npubi88//1x6nJOTg88//xyVK1eWbl+zxDFt6HVcuHChXqdvrMjISDg6OmLhwoU69RZ3j2ppBQQEoH79+vj66691+tv9+/dLU3wVxRz9hCGlGbPCkL179+Lx48do1KiRtCw0NBRpaWk6t3DduXNHb5YCQ1dmGdPnGqtPnz5Qq9WYMWOG3nO5ubnSMWhM329s31kWIiMjIZfL8emnn+q0e/ny5UhLSyvyc9hcDH2u5uTkYPHixXplXV1djboE2d/fH40bN8aqVat0+sfff/8dCQkJOl8SLKlPnz5ISkrCzp079Z579OiRdKLZs2dPCCEMJv8seWVXac7DOnfujJSUFJ0R8nNzc7Fw4UK4ubmhTZs2AIDo6GioVCqdvl2j0UhTXxZU0nOSoty+fVunz0hPT8fXX3+Nxo0bF3tOER0dDQ8PD8yePdvgmBLa/rNy5cpo3bo1vvrqK72pMot6D0tyvmhvb4+oqCj8+OOPOong1NRUrFmzBi1btizxrUvmZuwxYG9vD5lMpvP5fO3aNb2ZwExl7NSlvXr1glqt1rlVUqlUYsWKFQgPD9dJ/ty4cUNnLJ2YmBhs2rRJ7w/IO042bdok3V6snWVwyZIlOttfsmQJXFxcStzn9ujRA/b29oiPj9fbn4QQ0ufOs88+i5CQECxYsEDv3FG7nqE+GbD8uY4pTPkeau5zQcC0qUt79uyJrVu36kwhu3v3bvzxxx/o3bu3tEylUuHChQs6V2FoE0/r1q3TqXPLli148uSJzsxJ2rEA8//QqdFosGLFCnh7e5s0XIRNjFlhZ2eHL7/8Ep06dUK9evUwePBgVKlSBbdu3cLevXvh4eGBn376Sadd7733HmJiYuDo6IiuXbuiUaNGiI2NxbJly6TLQI8fP45Vq1ahe/fuRf7SPmDAAHz//fd4/fXXsXfvXrRo0QJqtRoXLlzA999/j507dxqcJsaU+L799lt0794dffr0wbZt29CuXTtERUVJv4gNHz4cjx8/xhdffAEfHx+dnWvVqlVYvHgxXn75ZYSGhiIjIwNffPEFPDw8ijxJKuy1cnV1RVhYGHbt2oWPPvoIAQEBCAkJQXh4ON5//33s3bsX4eHhGDp0KOrWrYsHDx7gl19+wa5du4y6vWLgwIHSnNDGXm752muv4cGDB2jXrh0CAwNx/fp1LFy4EI0bN9b5hQbIO4CmTJkizVld8B5Xc49ZURrvvPMO1q5di06dOuHNN9+Et7c3Vq1ahatXr+KHH34wbeqfEnrxxRcxffp0DB48GM2bN8eZM2ewevVqvfEBQkND4eXlhaVLl8Ld3R2urq4IDw83OI5G165d8cILL+C9997DtWvX0KhRIyQkJODHH3/E6NGjdQaqMkZGRgYCAwPRq1cvNGrUCG5ubti1axdOnDihN95AQEAAPvjgA1y7dg01a9bEd999h+TkZCxbtky6j9ISx/SLL76Ib775Bp6enqhbty6SkpKwa9euEo/7UblyZbz99tvS1FqdO3fG6dOnsX37dp0rXixh9uzZ6NatG1q0aIHBgwfj4cOH+Oyzz1C/fn2jbsUwRz9RUGnGrEhLS8O3334LIO8Ljna6TGdnZ7zzzjtSuZiYGEyYMAEvv/wy3nzzTWnKwJo1a+qM1TF9+nQcOHAAXbp0QXBwMO7evYvFixcjMDBQZ7DckmrTpg2GDx+OOXPmIDk5GVFRUXB0dMSlS5ewfv16fPLJJ+jVq5dRfb8pfaelVa5cGRMnTkR8fDw6duyIl156CRcvXsTixYvx3HPPlcnl982bN0eFChUQGxuLN998EzKZDN98843BL3hhYWH47rvvMHbsWDz33HNwc3ND165dDdY7b948dOrUCRERERgyZIg0damnpyemTZtm4ajyjBs3Dlu2bMGLL76IQYMGISwsDE+ePMGZM2ewYcMGXLt2DZUqVcILL7yAAQMG4NNPP8WlS5ekS60PHjyIF154AaNGjbJI+0pzHjZs2DB8/vnnGDRoEE6dOoVq1aphw4YNOHz4MBYsWCBdwde9e3c8//zzeOutt3D58mXUrl0bW7ZskfocQ1fLlOScpCg1a9bEkCFDcOLECfj6+uKrr75CamqqUb/ae3h4YMmSJRgwYACeffZZxMTEoHLlyrhx4wZ+/vlntGjRQvoR4NNPP0XLli3x7LPPYtiwYQgJCcG1a9fw888/Izk52WD9JT1fnDlzJhITE9GyZUuMHDkSDg4O+Pzzz6FUKjF37twSvU6WYOwx0KVLF3z00Ufo2LEj+vXrh7t372LRokWoXr26UeONFcbYqUvDw8PRu3dvTJw4EXfv3kX16tWxatUqXLt2Te8q14EDB2L//v1SH1W7dm3Url3bYL0hISHo3r279NjZ2RkzZsxAXFwcevfujejoaBw8eBDffvstZs2aVeIrq0JDQzFz5kxMnDgR165dQ/fu3eHu7o6rV69i06ZNGDZsGN5++23Y2dlhyZIl6Nq1Kxo3bozBgwfD398fFy5cwNmzZ7Fz5054eHhI09SqVCpUqVIFCQkJBq92Ky+mfA8197kgYPzUpQDw7rvvYv369XjhhRfwf//3f3j8+DHmzZuHBg0a6AxYfevWLdSpU0fnDoiuXbuiXr16mD59Oq5fv45mzZrh8uXL+Oyzz+Dv7y+NBwgA3bp1Q/v27TFnzhzcu3cPjRo1wubNm3Ho0CF8/vnnpo2ZaMrUIaZMXVpapkxdqnX69GnRo0cPUbFiRaFQKERwcLDo06eP2L17t065GTNmiCpVqgg7OzudqWdUKpWIj48XISEhwtHRUQQFBYmJEyfqTeNiaBqcnJwc8cEHH4h69eoJhUIhKlSoIMLCwkR8fLxIS0uTysHA1EFC5E1nlX+qQUPTumVmZoo2bdoINzc3cfToUSGEEFu2bBENGzYUTk5Oolq1auKDDz4QX331lU5cv/zyi+jbt6+oWrWqUCgUwsfHR7z44ovi5MmTOm1AgSmeinqtLly4IFq3bi2cnZ0FCkyTmJqaKuLi4kRQUJBwdHQUfn5+on379mLZsmVSGUNTu2nduXNH2Nvbi5o1a+o9l3/d/O//hg0bRFRUlPDx8RFyuVxUrVpVDB8+XG8qHSGEuHTpkjTdzqFDhwxuoyQKm7q0Xr16emUNTYFY2L5x5coV0atXL+Hl5SWcnJzE888/L7Zu3apTprBjU7stV1dXveWFtS04OFh06dJFepydnS3eeust4e/vL5ydnUWLFi1EUlKSwePgxx9/FHXr1hUODg4676+heDMyMsSYMWNEQECAcHR0FDVq1BDz5s3TmaKpqNcl/zGjVCrFuHHjRKNGjYS7u7twdXUVjRo1EosXLzYY88mTJ0VERIRwcnISwcHB4rPPPtOr39zH9MOHD8XgwYNFpUqVhJubm4iOjhYXLlzQK6edgqrg1HGG+j21Wi3i4+Ol96Zt27bi999/16vTlKlL582bpxeLob5h3bp1onbt2kKhUIj69euLLVu2iJ49e4ratWvrlDM0dakQxvUThe3XRfUfpio4dalMJhPe3t7ipZdeEqdOndIrn5CQIOrXry/kcrmoVauW+Pbbb/U+A3fv3i26desmAgIChFwuFwEBAaJv37460+UVNnWpoWO1sM/YZcuWibCwMOHs7Czc3d1FgwYNxPjx48Xt27eFEMb1/ab0nQUV3PdNeb8M9Zdan332mahdu7ZwdHQUvr6+YsSIEXrT2hXWfxW2HxfWNkPH2+HDh0WzZs2Es7OzCAgIEOPHjxc7d+7UO4YeP34s+vXrJ7y8vATw77S4he2fu3btEi1atBDOzs7Cw8NDdO3aVZw7d06nTGGvi6Ep+YpTcOpSIfL63YkTJ4rq1asLuVwuKlWqJJo3by7mz58vTd0sRN6UgvPmzRO1a9cWcrlcVK5cWXTq1EnnmCjN+UxhMZXmPCw1NVXqY+VyuWjQoIHBPuLvv/8W/fr1E+7u7sLT01MMGjRIHD58WADQmfpSqyTnJIXRfr7u3LlTNGzYUCgUClG7dm2j9sv89u7dK6Kjo4Wnp6dwcnISoaGhYtCgQXrndb///rt4+eWXpfOHWrVqicmTJ+ttxxzni7/88ouIjo4Wbm5uwsXFRbzwwgviyJEjRsVV2Hl9QaU9nxHC+GNg+fLlokaNGtJ7tGLFCoN9sbHHgbasMdNoCiFEVlaWePvtt4Wfn59QKBTiueeeEzt27DAYuzHfwQprpxB5nyW1atUScrlchIaGio8//ljvXMyQovpxIfKmCm7ZsqVwdXUVrq6uonbt2iIuLk5cvHhRp9yhQ4dEhw4dpPO3hg0b6kz7evPmTWk/9vT0FL179xa3b9/W2w9LO3VpwdfH1M8TY76HWuJc0NipS7V+//13ERUVJVxcXISXl5fo37+/SElJMRh7wX34wYMHYsyYMaJmzZpCoVCISpUqiZiYGL2p3oXIO9b+7//+T/j5+Ul9cv7ppY0lE8LCo7XZoFatWkGhUGDXrl3l3RSbdO/ePfj7+2PKlCmYPHmy3vPXrl1DSEgI9u7di7Zt25Z9A4nIoMaNG6Ny5co602BOmzYNK1euNHqcCCKisrZ582a8/PLLOHTokN69/uY8J6lWrRrq16+PrVu3mrP5REQ2y/LXkNugO3fuWPwy6/+ylStXQq1WY8CAAeXdFCIyQKVS6QykBOQNTPjrr78ygUhEVi0rK0vnsVqtxsKFC+Hh4YFnn31WrzzPSYiIyo/FxqywRUeOHMHGjRtx5coVTJgwobybY3P27NmDc+fOYdasWejevbtFZ9kgopK7desWIiMj8b///Q8BAQG4cOECli5dCj8/P7z++uvl3TwiokK98cYbyMrKQkREBJRKJTZu3IgjR45g9uzZOrPr8JyEiKj8MVlhgi+++ALbt2/H6NGjdQYhIfOYPn06jhw5ghYtWmDhwoXl3RwiKkSFChUQFhaGL7/8En///TdcXV3RpUsXvP/++6UaJIqIyNLatWuHDz/8EFu3bkV2djaqV6+OhQsX6g0cynMSIqLyxzEriIiIiIiIiMiqcMwKIiIiIiIiIrIqTFYQERERERERkVVhsoKIiIiIiIiIrAqTFURERERERERkVZisICIiIiIiIiKrwmQFEREREREREVkVJiuIiIiIiIiIyKowWUFEREREREREVoXJCiIiIiIiIiKyKg7l3QAiKppGo8Ht27fh7u4OmUxW3s0hIhskhEBGRgYCAgJgZ2d7v2OwHyUiS7P1fpSoPDBZQWTlbt++jaCgoPJuBhH9B/z1118IDAws72aYHftRIiorttqPEpUHJiuIrJy7uzuAvA8/Dw+PYsurVCokJCQgKioKjo6Olm6exdlSPIzFOtlSLEDJ4klPT0dQUJDU39gaU/tRwLb2C8ZinWwpFsC24mE/SmQdmKwgsnLaS5Y9PDyMTla4uLjAw8PjqT9ZAGwrHsZinWwpFqB08djqLRKm9qOAbe0XjMU62VIsgG3Fw36UyDrwhioiIiIiIiIisipMVhARERERERGRVWGygoiIiIiIiIisCpMVREREVCIHDhxA165dERAQAJlMhs2bN0vPqVQqTJgwAQ0aNICrqysCAgIwcOBA3L59W6eOBw8eoH///vDw8ICXlxeGDBmCx48f65T57bff0KpVKzg5OSEoKAhz584ti/CIiIioHDFZQURERCXy5MkTNGrUCIsWLdJ7LjMzE7/88gsmT56MX375BRs3bsTFixfx0ksv6ZTr378/zp49i8TERGzduhUHDhzAsGHDpOfT09MRFRWF4OBgnDp1CvPmzcO0adOwbNkyi8dHRERE5YezgRAREVGJdOrUCZ06dTL4nKenJxITE3WWffbZZ3j++edx48YNVK1aFefPn8eOHTtw4sQJNG3aFACwcOFCdO7cGfPnz0dAQABWr16NnJwcfPXVV5DL5ahXrx6Sk5Px0Ucf6SQ1iIiIyLYwWUFERERlIi0tDTKZDF5eXgCApKQkeHl5SYkKAIiMjISdnR2OHTuGl19+GUlJSWjdujXkcrlUJjo6Gh988AEePnyIChUq6G1HqVRCqVRKj9PT0wHk3ZqiUqmMaqu2nLHlrRljsU62FAtgW/GUJBZbiJvI2jBZQUT/GdnZ2cjJybFY/XK5HE5OTharn+hplp2djQkTJqBv377w8PAAAKSkpMDHx0ennIODA7y9vZGSkiKVCQkJ0Snj6+srPWcoWTFnzhzEx8frLU9ISICLi4tJ7S54dcjTjLFYJ1uKBbCteEyJJTMz04ItIfpvYrKCiP4TsrOzEVg1GPf/vmuxbVSs7IObN64zYUFUgEqlQp8+fSCEwJIlSyy+vYkTJ2Ls2LHS4/T0dAQFBSEqKkpKlBRHpVIhMTERHTp0gKOjo6WaWiYYi3WypViAso9HqVSidt16eHDvb7PX7ezsjK+++gpt2rSBq6urUetor+AiIvNhsoKI/hNycnJw/++7qDJiJewUpv2yagyNMhO3lgxCTk4OkxVE+WgTFdevX8eePXt0kgV+fn64e1c3gZibm4sHDx7Az89PKpOamqpTRvtYW6YghUIBhUKht9zR0dHkL1ElWcdaMRbrZEuxAGUXT1ZWFm79dcMin+vy3LyrJDQajdGx2NJ7SGQtmKwgov8UO4WLRZIVRKRPm6i4dOkS9u7di4oVK+o8HxERgUePHuHUqVMICwsDAOzZswcajQbh4eFSmffeew8qlUr6MpCYmIhatWoZvAWEiP5bLPG5bmdv1uqIqIQ4dSkRERGVyOPHj5GcnIzk5GQAwNWrV5GcnIwbN25ApVKhV69eOHnyJFavXg21Wo2UlBSkpKRIY8fUqVMHHTt2xNChQ3H8+HEcPnwYo0aNQkxMDAICAgAA/fr1g1wux5AhQ3D27Fl89913+OSTT3Ru8yAiIiLbwysriIiIqEROnjyJF154QXqsTSDExsZi2rRp2LJlCwCgcePGOuvt3bsXbdu2BQCsXr0ao0aNQvv27WFnZ4eePXvi008/lcp6enoiISEBcXFxCAsLQ6VKlTBlyhROW0pERGTjmKwgIiKiEmnbti2EEIU+X9RzWt7e3lizZk2RZRo2bIiDBw+a3D4iIiJ6evE2ECIiIiIiIiKyKkxWEBEREREREZFVYbKCiIiIiIiIiKwKkxVEREREREREZFWYrCAiIiIiIiIiq8JkBRERERERERFZFSYriAqxZMkSNGzYEB4eHvDw8EBERAS2b98uPd+2bVvIZDKdv9dff12njhs3bqBLly5wcXGBj48Pxo0bh9zc3LIOhYiIiIiI6KniUN4NILJWgYGBeP/991GjRg0IIbBq1Sp069YNp0+fRr169QAAQ4cOxfTp06V1XFxcpP+r1Wp06dIFfn5+OHLkCO7cuYOBAwfC0dERs2fPLvN4iIiIiIiInhZMVhAVomvXrjqPZ82ahSVLluDo0aNSssLFxQV+fn4G109ISMC5c+ewa9cu+Pr6onHjxpgxYwYmTJiAadOmQS6XWzwGKnvp6emFPqe9qiYjIwMODqZ1v3K5HE5OTqVqGxERERHR04LJCiIjqNVqrF+/Hk+ePEFERIS0fPXq1fj222/h5+eHrl27YvLkydLVFUlJSWjQoAF8fX2l8tHR0RgxYgTOnj2LJk2aGNyWUqmEUqmUHmu//KpUKqhUqmLbqi1jTNmngbniyc3NhbOzM5zsAZm9MEfTdGhEDpxd3VGzZs1Cyzg7O+Orr75CjRo1kJWVZVL93pUq48K5s1AoFKVtqlnY0n5mS7EAJYvHVmInIiIi28FkBVERzpw5g4iICGRnZ8PNzQ2bNm1C3bp1AQD9+vVDcHAwAgIC8Ntvv2HChAm4ePEiNm7cCABISUnRSVQAkB6npKQUus05c+YgPj5eb3lCQoLObSbFSUxMNLrs08Ac8axdu/af/6lLXZc+N+CFb4wq+dVXX5VoC7t37y7RepZkS/uZLcUCmBZPZmamBVtCREREZDomK4iKUKtWLSQnJyMtLQ0bNmxAbGws9u/fj7p162LYsGFSuQYNGsDf3x/t27fHlStXEBoaWuJtTpw4EWPHjpUep6enIygoCFFRUfDw8Ch2fZVKhcTERHTo0AGOjo4lboe1MFc8GRkZCAwMRODIVZApnM3Ywjzq9Pu4vXxEkfUr7ARmNNVg8kk7KDUyo+sWyizcXByLmzdvwt3d3VxNLhVb2s9sKRagZPEUdfsSERERUXlgsoKoCHK5HNWrVwcAhIWF4cSJE/jkk0/w+eef65UNDw8HAFy+fBmhoaHw8/PD8ePHdcqkpqYCQKHjXACAQqEweKm/o6OjSV+kTC1v7Uobj4ODA7KyspCtBuzUxicKjJWrhtH1KzUyKE1og+afuh0cHKzuPbWl/cyWYgFMi8eW4iYiIiLbwKlLiUyg0Wh0xpPILzk5GQDg7+8PAIiIiMCZM2dw9+5dqUxiYiI8PDykW0mIiIiIiIhIH6+sICrExIkT0alTJ1StWhUZGRlYs2YN9u3bh507d+LKlStYs2YNOnfujIoVK+K3337DmDFj0Lp1azRs2BAAEBUVhbp162LAgAGYO3cuUlJSMGnSJMTFxVnNIIlERERERETWiMkKokLcvXsXAwcOxJ07d+Dp6YmGDRti586d6NChA/766y/s2rULCxYswJMnTxAUFISePXti0qRJ0vr29vbYunUrRowYgYiICLi6uiI2NhbTp08vx6iIiIiIiIisH5MVRIVYvnx5oc8FBQVh//79xdYRHByMbdu2mbNZRERERERENo9jVhARERERERGRVWGygoiIiIiIiIisCpMVRERERERERGRVmKwgIiIiIiIiIqvCZAURERERERERWRUmK4iIiIiIiIjIqjBZQURERERERERWhckKIiIiIiIiIrIqTFYQERERERERkVVhsoKIiIhK5MCBA+jatSsCAgIgk8mwefNmneeFEJgyZQr8/f3h7OyMyMhIXLp0SafMgwcP0L9/f3h4eMDLywtDhgzB48ePdcr89ttvaNWqFZycnBAUFIS5c+daOjQiIiIqZ0xWEBERUYk8efIEjRo1wqJFiww+P3fuXHz66adYunQpjh07BldXV0RHRyM7O1sq079/f5w9exaJiYnYunUrDhw4gGHDhknPp6enIyoqCsHBwTh16hTmzZuHadOmYdmyZRaPj4iIiMqPQ3k3gIiIiJ5OnTp1QqdOnQw+J4TAggULMGnSJHTr1g0A8PXXX8PX1xebN29GTEwMzp8/jx07duDEiRNo2rQpAGDhwoXo3Lkz5s+fj4CAAKxevRo5OTn46quvIJfLUa9ePSQnJ+Ojjz7SSWoQERGRbWGygoiIiMzu6tWrSElJQWRkpLTM09MT4eHhSEpKQkxMDJKSkuDl5SUlKgAgMjISdnZ2OHbsGF5++WUkJSWhdevWkMvlUpno6Gh88MEHePjwISpUqKC3baVSCaVSKT1OT08HAKhUKqhUKqPary1nbHlrxlisky3FApR9PLm5uXB2doaTPSCzF2atWy7+3YapfQYRmQ+TFURERGR2KSkpAABfX1+d5b6+vtJzKSkp8PHx0XnewcEB3t7eOmVCQkL06tA+ZyhZMWfOHMTHx+stT0hIgIuLi0lxJCYmmlTemjEW62RLsQBlG8/atWv/+Z/azDXnJUeTkpKMXiMzM9PMbSAiJiuIiIjIpkycOBFjx46VHqenpyMoKAhRUVHw8PAwqg6VSoXExER06NABjo6OlmpqmWAs1smWYgHKPp6MjAwEBgYicOQqyBTOZq1bnpuFmc3liIiIMJgQNUR7BRcRmQ+TFURERGR2fn5+AIDU1FT4+/tLy1NTU9G4cWOpzN27d3XWy83NxYMHD6T1/fz8kJqaqlNG+1hbpiCFQgGFQqG33NHR0eQvUSVZx1oxFutkS7EAZRePg4MDsrKykK0G7NQys9Yt1P9uw9hYbOk9JLIWnA2EiIiIzC4kJAR+fn7YvXu3tCw9PR3Hjh1DREQEACAiIgKPHj3CqVOnpDJ79uyBRqNBeHi4VObAgQM694MnJiaiVq1aRv/iSURERE8fJiuIiIioRB4/fozk5GQkJycDyBtUMzk5GTdu3IBMJsPo0aMxc+ZMbNmyBWfOnMHAgQMREBCA7t27AwDq1KmDjh07YujQoTh+/DgOHz6MUaNGISYmBgEBAQCAfv36QS6XY8iQITh79iy+++47fPLJJzq3eRAREZHtYbKCqBBLlixBw4YN4eHhAQ8PD0RERGD79u3S89nZ2YiLi0PFihXh5uaGnj176l2qfOPGDXTp0gUuLi7w8fHBuHHjkJubW9ahEBFZxMmTJ9GkSRM0adIEADB27Fg0adIEU6ZMAQCMHz8eb7zxBoYNG4bnnnsOjx8/xo4dO+Dk5CTVsXr1atSuXRvt27dH586d0bJlSyxbtkx63tPTEwkJCbh69SrCwsLw1ltvYcqUKZy2lIiIyMZxzAqiQgQGBuL9999HjRo1IITAqlWr0K1bN5w+fRr16tXDmDFj8PPPP2P9+vXw9PTEqFGj0KNHDxw+fBgAoFar0aVLF/j5+eHIkSO4c+cOBg4cCEdHR8yePbucoyMiKr22bdtCiMKnDJTJZJg+fTqmT59eaBlvb2+sWbOmyO00bNgQBw8eLHE7iYiI6OnDZAVRIbp27arzeNasWViyZAmOHj2KwMBALF++HGvWrEG7du0AACtWrECdOnVw9OhRNGvWDAkJCTh37hx27doFX19fNG7cGDNmzMCECRMwbdo0yOXy8giLiIiIiIjI6jFZQWQEtVqN9evX48mTJ4iIiMCpU6egUqkQGRkplalduzaqVq2KpKQkNGvWDElJSWjQoAF8fX2lMtHR0RgxYgTOnj0rXTZdkFKphFKplB5rp8JSqVQ6A8wVRlvGmLJPA3PFk5ubC2dnZzjZAzL7wn8JLikHexRbv8JO6PxrLPFP3bm5uVbzvtrSfmZLsQAli8dWYiciIiLbwWQFURHOnDmDiIgIZGdnw83NDZs2bULdunWRnJwMuVwOLy8vnfK+vr5ISUkBAKSkpOgkKrTPa58rzJw5cxAfH6+3PCEhAS4uLka3PTEx0eiyTwNzxLN27dp//qcudV36vIBI4+qf0VRjYt1yoNVaq7wM3pb2M1uKBTAtnszMTAu2hIiIiMh0TFYQFaFWrVpITk5GWloaNmzYgNjYWOzfv9+i25w4caLOKPfp6ekICgpCVFQUPDw8il1fpVIhMTERHTp0sIk5v80VT0ZGBgIDAxE4chVkCmcztjCPOv0+bi8fUWT9CjuBGU01mHzSDkqN8XPCC2UWbi6Oxc2bN+Hu7m6uJpeKLe1nthQLULJ4tFdwEREREVkLJiuIiiCXy1G9enUAQFhYGE6cOIFPPvkEr7zyCnJycvDo0SOdqytSU1Ph5+cHAPDz88Px48d16tPOFqItY4hCoYBCodBb7ujoaNIXKVPLW7vSxuPg4ICsrCxkqwE7tfGJAmPlqmF0/UqNDEoT2qD5p24HBwere09taT+zpVgA0+KxpbiJiIjINnDqUiITaDQaKJVKhIWFwdHREbt375aeu3jxIm7cuIGIiAgAQEREBM6cOYO7d+9KZRITE+Hh4YG6deuWeduJiIiIiIieFryygqgQEydORKdOnVC1alVkZGRgzZo12LdvH3bu3AlPT08MGTIEY8eOhbe3Nzw8PPDGG28gIiICzZo1AwBERUWhbt26GDBgAObOnYuUlBRMmjQJcXFxBq+cICIiIiIiojxMVhAV4u7duxg4cCDu3LkDT09PNGzYEDt37kSHDh0AAB9//DHs7OzQs2dPKJVKREdHY/HixdL69vb22Lp1K0aMGIGIiAi4uroiNjYW06dPL6+QiIiIiIiIngpMVhAVYvny5UU+7+TkhEWLFmHRokWFlgkODsa2bdvM3TQiIiIiIiKbxjEriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhARERERERGRVWGygoiIiIiIiIisCpMVRERERERERGRVmKwgIiIiIiIiIqvCZAURERERERERWRUmK4iIiIiIiIjIqjBZQURERERERERWhckKIiIiIiIiIrIqTFYQERERERERkVVhsoKIiIiIiIiIrAqTFURERERERERkVZisICIiIiIiIiKrwmQFEREREREREVkVJiuIiIiIiIiIyKowWUFEREQWoVarMXnyZISEhMDZ2RmhoaGYMWMGhBBSGSEEpkyZAn9/fzg7OyMyMhKXLl3SqefBgwfo378/PDw84OXlhSFDhuDx48dlHQ4RERGVIYfybgARUX7Z2dnIycmRHufm5gIAMjIy4OBQ8i4rPT291G0jItN88MEHWLJkCVatWoV69erh5MmTGDx4MDw9PfHmm28CAObOnYtPP/0Uq1atQkhICCZPnozo6GicO3cOTk5OAID+/fvjzp07SExMhEqlwuDBgzFs2DCsWbOmPMMjIiIiC+KVFUSFmDNnDp577jm4u7vDx8cH3bt3x8WLF3XKtG3bFjKZTOfv9ddf1ylz48YNdOnSBS4uLvDx8cG4ceOkL+CkKzs7G4FVg+Hp6Sn9BQYGAgACAwN1lpv6FxQUVM7REf33HDlyBN26dUOXLl1QrVo19OrVC1FRUTh+/DiAvKsqFixYgEmTJqFbt25o2LAhvv76a9y+fRubN28GAJw/fx47duzAl19+ifDwcLRs2RILFy7EunXrcPv27XKMjoiIiCyJV1YQFWL//v2Ii4vDc889h9zcXLz77ruIiorCuXPn4OrqKpUbOnQopk+fLj12cXGR/q9Wq9GlSxf4+fnhyJEjuHPnDgYOHAhHR0fMnj27TON5GuTk5OD+33dRZcRK2CnyXkcn+7znAkeuQra65HXnZtzHneUjzNBKIjJW8+bNsWzZMvzxxx+oWbMmfv31Vxw6dAgfffQRAODq1atISUlBZGSktI6npyfCw8ORlJSEmJgYJCUlwcvLC02bNpXKREZGws7ODseOHcPLL7+st12lUgmlUik91l5ZpVKpoFKpjGq7tpyx5a0ZY7FOthQLUPbx5ObmwtnZGU72gMxeFL+CCeTi322Y2mcQkfkwWUFUiB07dug8XrlyJXx8fHDq1Cm0bt1aWu7i4gI/Pz+DdSQkJODcuXPYtWsXfH190bhxY8yYMQMTJkzAtGnTIJfL9dYp7Un203zyoz3xcHFxgUzhDABQ2AkAGjg5O0GmkZW4bnWus8VOagDAwR7F1p8Xy7//Gkv8U7cpJ02W9jTvZwXZUixAyeKxVOzvvPMO0tPTUbt2bdjb20OtVmPWrFno378/ACAlJQUA4Ovrq7Oer6+v9FxKSgp8fHx0nndwcIC3t7dUpqA5c+YgPj5eb3lCQoJOQtkYiYmJJpW3ZozFOtlSLEDZxrN27dp//leKXzMMyjs/S0pKMnqNzMxMM7eBiJisIDJSWloaAMDb21tn+erVq/Htt9/Cz88PXbt2xeTJk6WT4aSkJDRo0EDnRDw6OhojRozA2bNn0aRJE73tmOsk+2k9+SnsxGNGU00pa/YCIi11UmNa/abHIgdarcXBgwdL0jCLelr3M0NsKRbAtHgsdZL9/fffY/Xq1VizZg3q1auH5ORkjB49GgEBAYiNjbXINgFg4sSJGDt2rPQ4PT0dQUFBiIqKgoeHh1F1qFQqJCYmokOHDnB0dLRUU8sEY7FOthQLUPbxZGRkIDAwEIEjV0k/cJiLPDcLM5vLERERgQoVKhi1DsfGIjI/JiuIjKDRaDB69Gi0aNEC9evXl5b369cPwcHBCAgIwG+//YYJEybg4sWL2LhxI4C8XwQN/WKofc6Q0p5kP80nP4ZOPBR2AjOaajD5pB2UpbmyIv0+bi8fYZGTGmPrL2ksQpmFm4tjcf78ebi7u5uryRK5XA6FQmHSOk/zflaQLcUClCweS51kjxs3Du+88w5iYmIAAA0aNMD169cxZ84cxMbGSlelpaamwt/fX1ovNTUVjRs3BgD4+fnh7t27OvXm5ubiwYMHhV7VplAoDO7Tjo6OJr/HJVnHWjEW62RLsQBlF4+DgwOysrKQrQbs1CU/PzBEqP/dhrGx2NJ7SGQtmKwgMkJcXBx+//13HDp0SGf5sGHDpP83aNAA/v7+aN++Pa5cuYLQ0NASbctcJ9lP48lPUSceSo0MylKcjOSqYbGTGlPrNzUWtTIHWUoVqlWrVspWGlaxsg9u3rguzbxgiqdxPyuMLcUCmBaPpeLOzMyEnZ3uWN729vbQaPKuLgoJCYGfnx92794tJSfS09Nx7NgxjBiRN8ZMREQEHj16hFOnTiEsLAwAsGfPHmg0GoSHh1uk3URERFT+mKwgKsaoUaOwdetWHDhwQJqZojDaE+fLly8jNDQUfn5+0qj3WqmpqQBQ6C+CRAUJdS6gydUZeNRcNMpM3FoyCDk5OSVKVhAVpWvXrpg1axaqVq2KevXq4fTp0/joo4/w6quvAgBkMhlGjx6NmTNnokaNGtLUpQEBAejevTsAoE6dOujYsSOGDh2KpUuXQqVSYdSoUYiJiUFAQEA5RkdERESWxGQFUSGEEHjjjTewadMm7Nu3DyEhIcWuk5ycDADS5cwRERGYNWsW7t69Kw0Ql5iYCA8PD9StW9dibSfbZKdwMXuygsiSFi5ciMmTJ2PkyJG4e/cuAgICMHz4cEyZMkUqM378eDx58gTDhg3Do0eP0LJlS+zYsUMnebZ69WqMGjUK7du3h52dHXr27IlPP/20PEIiIiKiMsJkBVEh4uLisGbNGvz4449wd3eXxpjw9PSEs7Mzrly5gjVr1qBz586oWLEifvvtN4wZMwatW7dGw4YNAQBRUVGoW7cuBgwYgLlz5yIlJQWTJk1CXFycyWMEEBE9bdzd3bFgwQIsWLCg0DIymQzTp0/XmQK6IG9vb6xZs8YCLSQiIiJrZVd8EaL/piVLliAtLQ1t27aFv7+/9Pfdd98ByBuUcNeuXYiKikLt2rXx1ltvoWfPnvjpp5+kOuzt7bF161bY29sjIiIC//vf/zBw4MAiT8qJiIiIiIj+63hlBVEhhBBFPh8UFIT9+/cXW09wcDC2bdtmrmYRERERERHZPF5ZQURERERERERWhckKIiIiIiIiIrIqTFYQERERERERkVVhsoKIiIiIiIiIrAqTFURERERERERkVZisICIiIiIiIiKrwmQFEREREREREVkVJiuIiIiIiIiIyKowWUFEREREREREVoXJCiIiIiIiIiKyKkxWEBEREREREZFVYbKCiIiIiIiIiKwKkxVEREREREREZFWYrCAiIiIiIiIiq8JkBRERERERERFZFSYriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhAVYs6cOXjuuefg7u4OHx8fdO/eHRcvXtQpk52djbi4OFSsWBFubm7o2bMnUlNTdcrcuHEDXbp0gYuLC3x8fDBu3Djk5uaWZShERERERERPFSYriAqxf/9+xMXF4ejRo0hMTIRKpUJUVBSePHkilRkzZgx++uknrF+/Hvv378ft27fRo0cP6Xm1Wo0uXbogJycHR44cwapVq7By5UpMmTKlPEIiIiIiIiJ6KjiUdwOIrNWOHTt0Hq9cuRI+Pj44deoUWrdujbS0NCxfvhxr1qxBu3btAAArVqxAnTp1cPToUTRr1gwJCQk4d+4cdu3aBV9fXzRu3BgzZszAhAkTMG3aNMjl8vIIjYiIiIiIyKoxWUFkpLS0NACAt7c3AODUqVNQqVSIjIyUytSuXRtVq1ZFUlISmjVrhqSkJDRo0AC+vr5SmejoaIwYMQJnz55FkyZN9LajVCqhVCqlx+np6QAAlUoFlUpVbDu1ZYwpa21yc3Ph7OwMJ3tAZi8AAAo73X9LysEeenWbkzH1lzQWS7Zd/FN3bm6uSfvM07yfFWRLsQAli8dWYiciIiLbwWQFkRE0Gg1Gjx6NFi1aoH79+gCAlJQUyOVyeHl56ZT19fVFSkqKVCZ/okL7vPY5Q+bMmYP4+Hi95QkJCXBxcTG6zYmJiUaXtSZr1679539qneUzmmpKWbMXEGm4bvMwvn7TYzG+btPJgVZrcfDgwRKt/bTuZ4bYUiyAafFkZmZasCVEREREpmOygsgIcXFx+P3333Ho0CGLb2vixIkYO3as9Dg9PR1BQUGIioqCh4dHseurVCokJiaiQ4cOcHR0tGRTzS4jIwOBgYEIHLkKMoUzgLyrEGY01WDySTsoNbIS161Ov4/by0fo1G1OxtRf0lgs2XahzMLNxbG4efMm3N3djV7vad7PCrKlWICSxaO9gouIiIjIWjBZQVSMUaNGYevWrThw4AACAwOl5X5+fsjJycGjR490rq5ITU2Fn5+fVOb48eM69WlnC9GWKUihUEChUOgtd3R0NOmLlKnlrYGDgwOysrKQrQbs1Lpf5pUaGZTqkicrctUotG5zMKV+U2OxZNs1/9Tt4OBQov3ladzPCmNLsQCmxWPJuG/duoUJEyZg+/btyMzMRPXq1bFixQo0bdoUACCEwNSpU/HFF1/g0aNHaNGiBZYsWYIaNWpIdTx48ABvvPEGfvrpJ9jZ2aFnz5745JNP4ObmZrF2ExERUfnibCBEhRBCYNSoUdi0aRP27NmDkJAQnefDwsLg6OiI3bt3S8suXryIGzduICIiAgAQERGBM2fO4O7du1KZxMREeHh4oG7dumUTCBFROXn48CFatGgBR0dHbN++HefOncOHH36IChUqSGXmzp2LTz/9FEuXLsWxY8fg6uqK6OhoZGdnS2X69++Ps2fPIjExUUoeDxs2rDxCIiIiojLCKyuIChEXF4c1a9bgxx9/hLu7uzTGhKenJ5ydneHp6YkhQ4Zg7Nix8Pb2hoeHB9544w1ERESgWbNmAICoqCjUrVsXAwYMwNy5c5GSkoJJkyYhLi7O4NUTRES25IMPPkBQUBBWrFghLcuf+BVCYMGCBZg0aRK6desGAPj666/h6+uLzZs3IyYmBufPn8eOHTtw4sQJ6WqMhQsXonPnzpg/fz4CAgL0tlvagYq1ZfP/+zRjLNbJlmIByj4eQ4Nym4tc/LsNU/sMIjIfJiuICrFkyRIAQNu2bXWWr1ixAoMGDQIAfPzxx9IlyUqlEtHR0Vi8eLFU1t7eHlu3bsWIESMQEREBV1dXxMbGYvr06WUVBhFRudmyZQuio6PRu3dv7N+/H1WqVMHIkSMxdOhQAMDVq1eRkpKiM6uSp6cnwsPDkZSUhJiYGCQlJcHLy0tKVABAZGQk7OzscOzYMbz88st62zXXQMWAbQ28ylisky3FApRtPIUNyl16eVPLJyUlGb0GByomMj8mK4gKIUTxWXonJycsWrQIixYtKrRMcHAwtm3bZs6mERE9Ff78808sWbIEY8eOxbvvvosTJ07gzTffhFwuR2xsrHTFmqFZk/LPquTj46PzvIODA7y9vQudVam0AxUDtjXwKmOxTrYUC1D28RgalNtc5LlZmNlcjoiICJ3b1orCgYqJzI/JCiIiIrIIjUaDpk2bYvbs2QCAJk2a4Pfff8fSpUsRGxtrse2aa6Dikq5jrRiLdbKlWICyi6eoQblLS6j/3YY1DFRM9F/FATaJiIjIIvz9/fUGE65Tpw5u3LgB4N9ZkbSzJGkVnFUp/yDFQN595A8ePCh0ViUiIiJ6+jFZQURERBbRokULXLx4UWfZH3/8geDgYAB5g236+fnpzKqUnp6OY8eO6cyq9OjRI5w6dUoqs2fPHmg0GoSHh5dBFERERFQeeBsIERERWcSYMWPQvHlzzJ49G3369MHx48exbNkyLFu2DAAgk8kwevRozJw5EzVq1EBISAgmT56MgIAAdO/eHUDelRgdO3bE0KFDsXTpUqhUKowaNQoxMTEGZwIhIiIi28BkBREREVnEc889h02bNmHixImYPn06QkJCsGDBAvTv318qM378eDx58gTDhg3Do0eP0LJlS+zYsQNOTk5SmdWrV2PUqFFo3769NAPTp59+Wh4hERERURlhsoKIiIgs5sUXX8SLL75Y6PMymQzTp08vckpnb29vrFmzxhLNIyIiIivFMSuIiIiIiIiIyKowWUFEREREREREVoXJCiIiIiIiIiKyKkxWEBEREREREZFVYbKCiIiIiIiIiKwKkxVEREREREREZFWYrCAiIiIiIiIiq8JkBRERERERERFZFSYriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhARERERERGRVWGygqgQBw4cQNeuXREQEACZTIbNmzfrPD9o0CDIZDKdv44dO+qUefDgAfr37w8PDw94eXlhyJAhePz4cRlGQURERERE9PRhsoKoEE+ePEGjRo2waNGiQst07NgRd+7ckf7Wrl2r83z//v1x9uxZJCYmYuvWrThw4ACGDRtm6aYTERERERE91RzKuwFE1qpTp07o1KlTkWUUCgX8/PwMPnf+/Hns2LEDJ06cQNOmTQEACxcuROfOnTF//nwEBASYvc1ERERERES2gMkKolLYt28ffHx8UKFCBbRr1w4zZ85ExYoVAQBJSUnw8vKSEhUAEBkZCTs7Oxw7dgwvv/yywTqVSiWUSqX0OD09HQCgUqmgUqmKbZO2jDFlrU1ubi6cnZ3hZA/I7AUAQGGn+29JOdhDr25zMqb+ksZiybaLf+p++PAhcnNzjV5PW/bhw4dwcCj6o0Qul0OhUJSqnZb0NB8zhpQkHluJnYiIiGwHkxVEJdSxY0f06NEDISEhuHLlCt5991106tQJSUlJsLe3R0pKCnx8fHTWcXBwgLe3N1JSUgqtd86cOYiPj9dbnpCQABcXF6Pbl5iYaHwwVuTfW2nUOstnNNWUsmYvINJw3eZhfP2mx2J83aaTA63WIjk5uURrJyUlmbc55ehpPWYKY0o8mZmZFmwJERERkemYrCAqoZiYGOn/DRo0QMOGDREaGop9+/ahffv2Ja534sSJGDt2rPQ4PT0dQUFBiIqKgoeHR7Hrq1QqJCYmokOHDnB0dCxxO8pDRkYGAgMDEThyFWQKZwB5VyHMaKrB5JN2UGpkJa5bnX4ft5eP0KnbnIypv6SxWLLt2roDhiyBncL4ZJjCHpjZXI5JR3KgLCJ/olFm4vbyEbh58ybc3d3N0GLze5qPGUNKEo/2Ci4iIiIia8FkBZGZPPPMM6hUqRIuX76M9u3bw8/PD3fv3tUpk5ubiwcPHhQ6zgWQNw6GoUvmHR0dTfoiZWp5a+Dg4ICsrCxkqwE7te6XeaVGBqW65MmKXDUKrdscTKnf1Fgs2XZt3TkOLrBzMD5ZkXc7iho5Ds7IkRXeJs0/9Ts4OFj9/vg0HjNFMSUeW4qbiIiIbANnAyEyk5s3b+L+/fvw9/cHAERERODRo0c4deqUVGbPnj3QaDQIDw8vr2YSERERERFZPV5ZQVSIx48f4/Lly9Ljq1evIjk5Gd7e3vD29kZ8fDx69uwJPz8/XLlyBePHj0f16tURHR0NAKhTpw46duyIoUOHYunSpVCpVBg1ahRiYmI4EwgREREREVEReGUFUSFOnjyJJk2aoEmTJgCAsWPHokmTJpgyZQrs7e3x22+/4aWXXkLNmjUxZMgQhIWF4eDBgzq3cKxevRq1a9dG+/bt0blzZ7Rs2RLLli0rr5CIiIiIiIieCryygqgQbdu2hRCFTxO5c+fOYuvw9vbGmjVrzNksIiIiIiIim8crK4iIiIiIiIjIqjBZQURERERERERWhckKIiIiIiIiIrIqTFYQERFRmXj//fchk8kwevRoaVl2djbi4uJQsWJFuLm5oWfPnkhNTdVZ78aNG+jSpQtcXFzg4+ODcePGITc3t4xbT0RERGWJyQoiIiKyuBMnTuDzzz9Hw4YNdZaPGTMGP/30E9avX4/9+/fj9u3b6NGjh/S8Wq1Gly5dkJOTgyNHjmDVqlVYuXIlpkyZUtYhEBERURnibCBERERkUY8fP0b//v3xxRdfYObMmdLytLQ0LF++HGvWrEG7du0AACtWrECdOnVw9OhRNGvWDAkJCTh37hx27doFX19fNG7cGDNmzMCECRMwbdo0yOVyve0plUoolUrpcXp6OgBApVJBpVIZ1WZtOWPLWzPGYp1sKRag7OPJzc2Fs7MznOwBmX3hs7eVhFz8uw1T+wwiMh8mK4iIiMii4uLi0KVLF0RGRuokK06dOgWVSoXIyEhpWe3atVG1alUkJSWhWbNmSEpKQoMGDeDr6yuViY6OxogRI3D27Fk0adJEb3tz5sxBfHy83vKEhAS4uLiY1PbExESTylszxmKdbCkWoGzjWbt27T//U5u55rwkaFJSktFrZGZmmrkNRMRkBREREVnMunXr8Msvv+DEiRN6z6WkpEAul8PLy0tnua+vL1JSUqQy+RMV2ue1zxkyceJEjB07Vnqcnp6OoKAgREVFwcPDw6h2q1QqJCYmokOHDnB0dDRqHWvFWKyTLcUClH08GRkZCAwMRODIVZApnM1atzw3CzObyxEREYEKFSoYtY72Ci4iMh8mK4iIiMgi/vrrL/zf//0fEhMT4eTkVGbbVSgUUCgUessdHR1N/hJVknWsFWOxTrYUC1B28Tg4OCArKwvZasBOLTNr3UL97zaMjcWW3kMia8FkBRGZLDs7Gzk5OWavl79KENmWU6dO4e7du3j22WelZWq1GgcOHMBnn32GnTt3IicnB48ePdK5uiI1NRV+fn4AAD8/Pxw/flynXu1sIdoyREREZHuYrCAik2RnZyOwajDu/323vJtCRFauffv2OHPmjM6ywYMHo3bt2pgwYQKCgoLg6OiI3bt3o2fPngCAixcv4saNG4iIiAAAREREYNasWbh79y58fHwA5N0T7+Hhgbp165ZtQERERFRmmKwgIpPk5OTg/t93UWXEStgpTBuorji5GfdxZ/kIs9ZJROXH3d0d9evX11nm6uqKihUrSsuHDBmCsWPHwtvbGx4eHnjjjTcQERGBZs2aAQCioqJQt25dDBgwAHPnzkVKSgomTZqEuLg4g7d6EBERkW1gsoKISsRO4WL2ZIWdkiNpE/3XfPzxx7Czs0PPnj2hVCoRHR2NxYsXS8/b29tj69atGDFiBCIiIuDq6orY2FhMnz69HFtNRERElsZkBREREZWZffv26Tx2cnLCokWLsGjRokLXCQ4OxrZt2yzcMiIiIrImduXdACIiIiIiIiKi/HhlBRERWZylZnqRy+VlOiUmEREREZUNJiuIiMhiRG4OYOeAoKAgi9RfsbIPbt64zoQFERERkY1hsoKIiCxGqHMBTa5FZo/RKDNxa8kg5OTkMFlBREREZGM4ZgVRIQ4cOICuXbsiICAAMpkMmzdv1nleCIEpU6bA398fzs7OiIyMxKVLl3TKPHjwAP3794eHhwe8vLwwZMgQPH78uAyjILIO2tljzP1HRERERLaJyQqiQjx58gSNGjUqdIT6uXPn4tNPP8XSpUtx7NgxuLq6Ijo6GtnZ2VKZ/v374+zZs0hMTMTWrVtx4MABDBs2rKxCICIiIiIieirxNhCiQnTq1AmdOnUy+JwQAgsWLMCkSZPQrVs3AMDXX38NX19fbN68GTExMTh//jx27NiBEydOoGnTpgCAhQsXonPnzpg/fz4CAgLKLBYiIiIiIqKnCZMVRCVw9epVpKSkIDIyUlrm6emJ8PBwJCUlISYmBklJSfDy8pISFQAQGRkJOzs7HDt2DC+//LLBupVKJZRKpfRYO4uCSqWCSqUqtm3aMsaULYnc3Fw4OzvDyR6Q2Quz1u1gD726FXa6/5qzbnMypv6SxmLJtpe0bmNjsWTbxT915+bmlmp/t/QxU9ZKEo+txE5ERES2g8kKohJISUkBAPj6+uos9/X1lZ5LSUmBj4+PzvMODg7w9vaWyhgyZ84cxMfH6y1PSEiAi4vx9+gnJiYaXdZUa9eu/ed/ajPX7AVEGq57RlONxeo2D+PrNz0W4+s2XenqLj6W0tVfNDnQai0OHjxoltosecyUB1PiyczMtGBLiIiIiEzHZAWRlZk4cSLGjh0rPU5PT0dQUBCioqLg4eFR7PoqlQqJiYno0KEDHB0dzd6+jIwMBAYGInDkKsgUzmatW51+H7eXj9CpW2EnMKOpBpNP2kGpkZm1bnMypv6SxmLJtpe0bmNjsWTbhTILNxfH4ubNm3B3dy9xPZY+ZspaSeLRXsFFREREZC2YrCAqAT8/PwBAamoq/P39peWpqalo3LixVObu3bs66+Xm5uLBgwfS+oYoFAooFAq95Y6OjiZ9kTK1vLEcHByQlZWFbDVgpy558sCQXDUKrVupkUFZiu0VVbc5mFK/qbFYsu2lrbu4WCzZds0/dTs4OJhlX7fUMVNeTInHluImIiIi28DZQIhKICQkBH5+fti9e7e0LD09HceOHUNERAQAICIiAo8ePcKpU6ekMnv27IFGo0F4eHiZt5mIiIiIiOhpwSsriArx+PFjXL58WXp89epVJCcnw9vbG1WrVsXo0aMxc+ZM1KhRAyEhIZg8eTICAgLQvXt3AECdOnXQsWNHDB06FEuXLoVKpcKoUaMQExPDmUCIiIiIiIiKwGQFUSFOnjyJF154QXqsHUciNjYWK1euxPjx4/HkyRMMGzYMjx49QsuWLbFjxw44OTlJ66xevRqjRo1C+/btYWdnh549e+LTTz8t81iIiIiIiIieJkxWEBWibdu2EKLwqRZlMhmmT5+O6dOnF1rG29sba9assUTziIiIiIiIbBbHrCAiIiIiIiIiq8JkBRERERERERFZFSYriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhARERERERGRVWGygoiIiIiIiIisCpMVRERERERERGRVmKwgIiIiIiIiIqvCZAURERERERERWRUmK4iIiIiIiIjIqjBZQURERBYxZ84cPPfcc3B3d4ePjw+6d++Oixcv6pTJzs5GXFwcKlasCDc3N/Ts2ROpqak6ZW7cuIEuXbrAxcUFPj4+GDduHHJzc8syFCIiIipjTFYQERGRRezfvx9xcXE4evQoEhMToVKpEBUVhSdPnkhlxowZg59++gnr16/H/v37cfv2bfTo0UN6Xq1Wo0uXLsjJycGRI0ewatUqrFy5ElOmTCmPkIiIiKiMOJR3A4iIiMg27dixQ+fxypUr4ePjg1OnTqF169ZIS0vD8uXLsWbNGrRr1w4AsGLFCtSpUwdHjx5Fs2bNkJCQgHPnzmHXrl3w9fVF48aNMWPGDEyYMAHTpk2DXC7X265SqYRSqZQep6enAwBUKhVUKpVRbdeWM7a8NWMs1smWYgHKPp7c3Fw4OzvDyR6Q2Quz1i0X/27D1D6DiMyHyQoiIiIqE2lpaQAAb29vAMCpU6egUqkQGRkplalduzaqVq2KpKQkNGvWDElJSWjQoAF8fX2lMtHR0RgxYgTOnj2LJk2a6G1nzpw5iI+P11uekJAAFxcXk9qcmJhoUnlrxlisky3FApRtPGvXrv3nf2oz15yXBE1KSjJ6jczMTDO3gYiYrCAiIiKL02g0GD16NFq0aIH69esDAFJSUiCXy+Hl5aVT1tfXFykpKVKZ/IkK7fPa5wyZOHEixo4dKz1OT09HUFAQoqKi4OHhYVR7VSoVEhMT0aFDBzg6Ohq1jrViLNbJlmIByj6ejIwMBAYGInDkKsgUzmatW56bhZnN5YiIiECFChWMWkd7BRcRmQ+TFURERGRxcXFx+P3333Ho0CGLb0uhUEChUOgtd3R0NPlLVEnWsVaMxTrZUixA2cXj4OCArKwsZKsBO7XMrHUL9b/bMDYWW3oPiawFB9gkIiIiixo1ahS2bt2KvXv3IjAwUFru5+eHnJwcPHr0SKd8amoq/Pz8pDIFZwfRPtaWISIiItvDZAVRKUybNg0ymUznr3bt2tLzxkzJR0Rkq4QQGDVqFDZt2oQ9e/YgJCRE5/mwsDA4Ojpi9+7d0rKLFy/ixo0biIiIAABERETgzJkzuHv3rlQmMTERHh4eqFu3btkEQkRERGWOt4EQlVK9evWwa9cu6bGDw7+H1ZgxY/Dzzz9j/fr18PT0xKhRo9CjRw8cPny4PJpKZJNKe59wbm4ugLz7n/Mfv3K5HE5OTqWq+78uLi4Oa9aswY8//gh3d3dpjAlPT084OzvD09MTQ4YMwdixY+Ht7Q0PDw+88cYbiIiIQLNmzQAAUVFRqFu3LgYMGIC5c+ciJSUFkyZNQlxcnMFbPYiIiMg2MFlBVEoODg4GL0U2Zko+Iio5kZsD2DkgKCioVPU4Oztj7dq1CAwMRFZWlrS8YmUf3LxxnQmLUliyZAkAoG3btjrLV6xYgUGDBgEAPv74Y9jZ2aFnz55QKpWIjo7G4sWLpbL29vbYunUrRowYgYiICLi6uiI2NhbTp08vqzCIiIioHDBZQVRKly5dQkBAAJycnBAREYE5c+agatWqRk3JZ4hSqYRSqZQea381VqlURs3hbel5zi05r7mDPfTqVtjp/mvOus3JmPpLGosl217Suo2NxZJtVyMXzgpHBAz5EnYK06ajzE9hn/dv6BuroPxnUDWNMhO3l49AZmYm7O3tzdDaslOSPsBS/YUQxb/nTk5OWLRoERYtWlRomeDgYGzbts2cTSOif2RnZyMnJ6fU9Ri6So1XqBFRaTBZQVQK4eHhWLlyJWrVqoU7d+4gPj4erVq1wu+//27UlHyGzJkzB/Hx8XrLExIS4OJi/BcyS85zbrl5zb2ASMN1z2iqsVjd5mF8/abHYnzdpitd3cXHUrr6ja+79GY2l+d7JAci1+LgwYNmq7+smdIHZGZmWrAlRGStsrOzEVg1GPf/vlt84WIYukqNV6gRUWkwWUFUCp06dZL+37BhQ4SHhyM4OBjff/89nJ1LNuf3xIkTMXbsWOlxeno6goKCEBUVBQ8Pj2LX185zHhERoXP/vblkZGSgTp06FpnXXJ1+H7eXj9CpW2EnMKOpBpNP2kGpKfnUZIbqNidj6i9pLJZse0nrNjYWa2x7QYZiEcos3Fwci5s3b8Ld3d1cTS4T2j6gQ4cORk+lV9pxP4jo6ZSTk4P7f99FlRErS3WFGgA4/XMRWuDIVchW512hdmvJIOTk5DBZQUQlwmQFkRl5eXmhZs2auHz5Mjp06CBNyZf/6or8U/IZolAoDA4aZ+y85dpbSKpVq6Zz/725WWJe81w1Cp0zXamRQVmK7RVVtzmYUr+psViy7aWtu7hYrLntBeWPRfNP3Q4ODkZ/4bc2xvYZ2rJE9N9lp3ApdbIi71Y/NWQKZ4t8zhLRfw+TFURm9PjxY1y5cgUDBgzQmZKvZ8+eAPSn5LME7X2nAUOWIMehdCcehuRm3Med5SPMXi8REREREZEWkxVEpfD222+ja9euCA4Oxu3btzF16lTY29ujb9++Rk3JZ0l2ChfYWSBZYafkve1ERERERGRZTFYQlcLNmzfRt29f3L9/H5UrV0bLli1x9OhRVK5cGUDxU/IRERERERGRPiYriEph3bp1RT5vzJR8REREREREpIvJCiIiokJYapYMuVzO0fGJiIiIisBkBRERUQEiNwewc0BQUJBF6q9Y2Qc3b1xnwoKIiIioEExWEBERFSDUuYAmF1VGrCz1dH4FaZSZuLVkEHJycpisICIiIioEkxVERESFsFO4mD1ZQURkquzsbGlqcnOy1K1uRETmwGQFEREREZGVys7ORmDVYNz/+255N4WIqEwxWUFERFQOLPWLZm5urkXqJaLykZOTg/t/37XIbWm5GfdxZ/kIs9ZJRGQuTFYQERGVIUsP3uns7Iy1a9dCqVTC0dHRItsgorJnidvS7JSZZq2PiMicmKwgIiIqQ5YcvBMA5Ll5Xz4scX87ERERUVlhsoKIiKgcWGrwTjt7s1dJRFRilrrljYODEtk+JiuIiIiIyOZZakYNAJDL5ZyKuABL3/JGRLaPyQoiIiIismmWnlGjYmUf3LxxnQmLfCx9yxsHByWyfUxWEBEREZFNs+SMGhplJm4tGYScnBwmKwyw2C1vHByUyOYxWUFERERE/wmW+uJMRETmx2QFERERUSEyMjLg4GD+0yWOcWB7TB3wMTc3F0Dx+xgHkiSi/yomK4iIiIgKUCqVAIDAwEBkZWWZvf6neYwDDlSpq6QDSTo7O2Pt2rUW28eIiJ52TFYQERERFaD9Mh4wZAlyHDjGgRYHqtRX0oEknf6ZZjhw5Cpkqwsvx4Ekiei/iskKIiIiokLYKVxgZ+ZkhaUVvPLB2NsNjJGens6BKgth6ngYMnsBQA2Zwhl2alnh9XIgSSL6j2KygoiIiMhGGLrywRK3G1hyoMqixmgoaeKF4z4QET19mKwgKgOLFi3CvHnzkJKSgkaNGmHhwoV4/vnny7tZRERPDfajxjE0RaextxsYw5K3JBgz9gPHeSAi+u9gsoLIwr777juMHTsWS5cuRXh4OBYsWIDo6GhcvHgRPj4+5d08IiKrx37UdPmvfDD2dgOj6rXgLQnGjP1Q0sQLx30gInr62JV3A4hs3UcffYShQ4di8ODBqFu3LpYuXQoXFxd89dVX5d00IqKnAvvR/xZtosXQn0zhDAB5iZciyun9yZ3LOSoiIjIVr6wgsqCcnBycOnUKEydOlJbZ2dkhMjISSUlJBtdRKpXSlHkAkJaWBgB48OABVCpVsdt89OgRMjMzYfckDfYOFvgFLPshnJyc4JB5H7JcM9dvoG57eyAzUw77JzmwL83ly5Zst5H1lziWMn7NjWF0LFbY9oIMxvIUtLswdrlZyMz0xMOHDyGEMGqdjIwMADC6fFkqj34UsGxfapeTBScnJ9y4cQNubm5mrfvx48eW60eBcj822I9aR9sN0YvHCj53S8rW+lGip5VM8Igispjbt2+jSpUqOHLkCCIiIqTl48ePx/79+3Hs2DG9daZNm4b4+PiybCYREQDgr7/+QmBgYHk3Qwf7USJ6mlhjP0r0tOKVFURWZuLEiRg7dqz0WKPR4MGDB6hYsSJksuLvNU5PT0dQUBD++usveHh4WLKpZcKW4mEs1smWYgFKFo8QAhkZGQgICLBw68pGaftRwLb2C8ZinWwpFsC24mE/SmQdmKwgsqBKlSrB3t4eqampOstTU1Ph5+dncB2FQgGFQqGzzMvLy+Rte3h4PPUnC/nZUjyMxTrZUiyA6fF4enpasDUlV579KGBb+wVjsU62FAtgW/HYSj9K9LTiAJtEFiSXyxEWFobdu3dLyzQaDXbv3q1zOTMRERnGfpSIiOi/iVdWEFnY2LFjERsbi6ZNm+L555/HggUL8OTJEwwePLi8m0ZE9FRgP0pERPTfw2QFkYW98sor+PvvvzFlyhSkpKSgcePG2LFjB3x9fS2yPYVCgalTp+pdAv20sqV4GIt1sqVYANuLByj7fhSwrdeRsVgnW4oFsK14bCkWoqcZZwMhIiIiIiIiIqvCMSuIiIiIiIiIyKowWUFEREREREREVoXJCiIiIiIiIiKyKkxWEBEREREREZFVYbKCiIiIiIiIiKwKkxVET6FFixahWrVqcHJyQnh4OI4fP15k+fXr16N27dpwcnJCgwYNsG3btjJqqXFMieeLL75Aq1atUKFCBVSoUAGRkZHFxl+WTH1vtNatWweZTIbu3btbtoEmMDWWR48eIS4uDv7+/lAoFKhZs6bV7GumxrJgwQLUqlULzs7OCAoKwpgxY5CdnV1GrS3cgQMH0LVrVwQEBEAmk2Hz5s3FrrNv3z48++yzUCgUqF69OlauXGnxdj4tbKkvZT/KfrQs2EJfyn6U6CkiiOipsm7dOiGXy8VXX30lzp49K4YOHSq8vLxEamqqwfKHDx8W9vb2Yu7cueLcuXNi0qRJwtHRUZw5c6aMW26YqfH069dPLFq0SJw+fVqcP39eDBo0SHh6eoqbN2+Wccv1mRqL1tWrV0WVKlVEq1atRLdu3cqmscUwNRalUimaNm0qOnfuLA4dOiSuXr0q9u3bJ5KTk8u45fpMjWX16tVCoVCI1atXi6tXr4qdO3cKf39/MWbMmDJuub5t27aJ9957T2zcuFEAEJs2bSqy/J9//ilcXFzE2LFjxblz58TChQuFvb292LFjR9k02IrZUl/KfpT9aFmwlb6U/SjR04PJCqKnzPPPPy/i4uKkx2q1WgQEBIg5c+YYLN+nTx/RpUsXnWXh4eFi+PDhFm2nsUyNp6Dc3Fzh7u4uVq1aZakmGq0kseTm5ormzZuLL7/8UsTGxlrNSbapsSxZskQ888wzIicnp6yaaDRTY4mLixPt2rXTWTZ27FjRokULi7bTVMacZI8fP17Uq1dPZ9krr7wioqOjLdiyp4Mt9aXsR9mPlgVb7EvZjxJZN94GQvQUycnJwalTpxAZGSkts7OzQ2RkJJKSkgyuk5SUpFMeAKKjowstX5ZKEk9BmZmZUKlU8Pb2tlQzjVLSWKZPnw4fHx8MGTKkLJpplJLEsmXLFkRERCAuLg6+vr6oX78+Zs+eDbVaXVbNNqgksTRv3hynTp2SLm/+888/sW3bNnTu3LlM2mxO1nz8lydb6kvZj7IfLQv/5b7UWo99ov8Ch/JuABEZ7969e1Cr1fD19dVZ7uvriwsXLhhcJyUlxWD5lJQUi7XTWCWJp6AJEyYgICBA70SirJUklkOHDmH58uVITk4ugxYarySx/Pnnn9izZw/69++Pbdu24fLlyxg5ciRUKhWmTp1aFs02qCSx9OvXD/fu3UPLli0hhEBubi5ef/11vPvuu2XRZLMq7PhPT09HVlYWnJ2dy6ll5cuW+lL2o+xHy8J/uS9lP0pUfnhlBRE9td5//32sW7cOmzZtgpOTU3k3xyQZGRkYMGAAvvjiC1SqVKm8m1NqGo0GPj4+WLZsGcLCwvDKK6/gvffew9KlS8u7aSbbt28fZs+ejcWLF+OXX37Bxo0b8fPPP2PGjBnl3TQis2M/aj1sqR8F2JcSUenxygqip0ilSpVgb2+P1NRUneWpqanw8/MzuI6fn59J5ctSSeLRmj9/Pt5//33s2rULDRs2tGQzjWJqLFeuXMG1a9fQtWtXaZlGowEAODg44OLFiwgNDbVsowtRkvfF398fjo6OsLe3l5bVqVMHKSkpyMnJgVwut2ibC1OSWCZPnowBAwbgtddeAwA0aNAAT548wbBhw/Dee+/Bzu7pyfMXdvx7eHj8p38NtKW+lP0o+9Gy8F/uS9mPEpWfp6OXICIAgFwuR1hYGHbv3i0t02g02L17NyIiIgyuExERoVMeABITEwstX5ZKEg8AzJ07FzNmzMCOHTvQtGnTsmhqsUyNpXbt2jhz5gySk5Olv5deegkvvPACkpOTERQUVJbN11GS96VFixa4fPmy9EUBAP744w/4+/uX6wl2SWLJzMzUO4nWfnkQQliusRZgzcd/ebKlvpT9KPvRsvBf7kut9dgn+k8o3/E9ichU69atEwqFQqxcuVKcO3dODBs2THh5eYmUlBQhhBADBgwQ77zzjlT+8OHDwsHBQcyfP1+cP39eTJ061Wqm2xPC9Hjef/99IZfLxYYNG8SdO3ekv4yMjPIKQWJqLAVZ0yj2psZy48YN4e7uLkaNGiUuXrwotm7dKnx8fMTMmTPLKwSJqbFMnTpVuLu7i7Vr14o///xTJCQkiNDQUNGnT5/yCkGSkZEhTp8+LU6fPi0AiI8++kicPn1aXL9+XQghxDvvvCMGDBgglddOuTdu3Dhx/vx5sWjRIk659w9b6kvZj/6L/ajl2Epfyn6U6OnBZAXRU2jhwoWiatWqQi6Xi+eff14cPXpUeq5NmzYiNjZWp/z3338vatasKeRyuahXr574+eefy7jFRTMlnuDgYAFA72/q1Kll33ADTH1v8rOmk2whTI/lyJEjIjw8XCgUCvHMM8+IWbNmidzc3DJutWGmxKJSqcS0adNEaGiocHJyEkFBQWLkyJHi4cOHZd/wAvbu3Wtw/9e2PzY2VrRp00ZvncaNGwu5XC6eeeYZsWLFijJvt7Wypb6U/Wge9qOWZQt9KftRoqeHTIin6DosIiIiIiIiIrJ5HLOCiIiIiIiIiKwKkxVEREREREREZFWYrCAiIiIiIiIiq8JkBRERERERERFZFSYriIiIiIiIiMiqMFlBRERERERERFaFyQoiIiIiIiIisipMVhARERERERGRVWGygoiIiIiIiIisCpMVRERERERERGRVmKwgIiIiIiIiIqvy/y5x7o5iE6XiAAAAAElFTkSuQmCC", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "plt.figure()\n", - "dp = 'fi=Tietojenkäsittely|sv=Informationsbehandling|en=Business Information Technology|'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n", - "\n", - "plt.figure()\n", - "dp = 'fi=Automaatiotekniikka|sv=Automationsteknik|en=Automation Engineering|'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n", - "\n", - "plt.figure()\n", - "dp = 'fi=Esittävä taide ja musiikki|sv=Scenkonst och musik|en=Performance art and music|'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n", - "\n", - "plt.figure()\n", - "dp = 'fi=Fysioterapia|sv=Fysioterapi|en=Physiotherapy|'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n", - "plt.figure()\n", - "\n", - "dp = 'fi=Poliisi|sv=Polis|en=Police|'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n", - "\n", - "plt.figure()\n", - "dp = '(Muu julkaisu)'\n", - "plt.subplot(1,2,1)\n", - "df_fi[df_fi.degreeprogram == dp]['precision'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.subplot(1,2,2)\n", - "df_fi[df_fi.degreeprogram == dp]['recall'].hist(bins=11, figsize=(8, 3), edgecolor = \"black\", linewidth = 0.8)\n", - "plt.suptitle(dp + ' precision mean: ' + str(df_fi[df_fi.degreeprogram == dp]['precision'].mean())[:4] + ' recall mean: ' + str(df_fi[df_fi.degreeprogram == dp]['recall'].mean())[:4]) \n" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# (Correlation of F1 score in JYX test set and in Theseus)" - ] - }, - { - "cell_type": "code", - "execution_count": 33, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/tmp/ipykernel_184606/2028142037.py:1: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n", - " df_theseus_updates = df_fi.groupby(\"annif_update\")['precision', 'recall', 'f1 score'].agg(['mean', 'count'])\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
precisionrecallf1 score
meancountmeancountmeancount
annif_update
2020-03-260.42228640840.84579040840.5400594084
2020-12-090.42528752670.85786652670.5446065267
2021-04-270.43937188920.87736188920.5607568892
2021-11-110.484445159730.907410159730.60577015973
2022-06-210.51008926960.91523926960.6267862696
2022-11-220.50514272230.91476472230.6233267223
\n", - "
" - ], - "text/plain": [ - " precision recall f1 score \n", - " mean count mean count mean count\n", - "annif_update \n", - "2020-03-26 0.422286 4084 0.845790 4084 0.540059 4084\n", - "2020-12-09 0.425287 5267 0.857866 5267 0.544606 5267\n", - "2021-04-27 0.439371 8892 0.877361 8892 0.560756 8892\n", - "2021-11-11 0.484445 15973 0.907410 15973 0.605770 15973\n", - "2022-06-21 0.510089 2696 0.915239 2696 0.626786 2696\n", - "2022-11-22 0.505142 7223 0.914764 7223 0.623326 7223" - ] - }, - "execution_count": 33, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_theseus_updates = df_fi.groupby(\"annif_update\")['precision', 'recall', 'f1 score'].agg(['mean', 'count'])\n", - "# index with month precision\n", - "df_theseus_updates.index = df_theseus_updates.index.map(lambda x: x.date())\n", - "df_theseus_updates" - ] - }, - { - "cell_type": "code", - "execution_count": 34, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
f1 score
annif_update
2020-03-260.4288
2020-12-090.4746
2021-04-270.4929
2021-11-110.5445
2022-06-210.5169
2022-11-220.5188
\n", - "
" - ], - "text/plain": [ - " f1 score\n", - "annif_update \n", - "2020-03-26 0.4288\n", - "2020-12-09 0.4746\n", - "2021-04-27 0.4929\n", - "2021-11-11 0.5445\n", - "2022-06-21 0.5169\n", - "2022-11-22 0.5188" - ] - }, - "execution_count": 34, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "df_jyx_updates = pd.DataFrame({\"f1 score\": \n", - " [0.4288, 0.4746, 0.4929, 0.5445, 0.5169, 0.5188]},\n", - " index=df_theseus_updates.index)\n", - " # pd.to_datetime(\n", - " # [\"2020-03-01\", \"2020-12-01\", \"2021-04-01\", \"2021-09-01\", \"2022-05-01\", \"2022-11-01\"]))\n", - "df_jyx_updates" - ] - }, - { - "cell_type": "code", - "execution_count": 35, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "Text(0, 0.5, 'JYX')" - ] - }, - "execution_count": 35, - "metadata": {}, - "output_type": "execute_result" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "plt.plot(df_theseus_updates.precision['mean'], df_jyx_updates, '-o')\n", - "plt.xlabel(\"Theseus\")\n", - "plt.ylabel(\"JYX\")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.8.10" - }, - "vscode": { - "interpreter": { - "hash": "186dc853cf67aa271b9921a97411607eafab6dd5e87641576ef5a5a9e3abec98" - } - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/repository-metrics-analysis/collect.py b/repository-metrics-analysis/collect.py deleted file mode 100644 index c093a4fde605d743e995c4eb9af10c6cd5ce57e9..0000000000000000000000000000000000000000 --- a/repository-metrics-analysis/collect.py +++ /dev/null @@ -1,109 +0,0 @@ -#!/usr/bin/env python3 - -import traceback -import time -import sys - -import requests -import json -from lxml import etree - - -# Script for collecting metadata from repositories with open-search interface -# and outputting it as newline-delimited JSON. See also instructions in -# https://www.kiwi.fi/pages/viewpage.action?pageId=45782169 -# Start page be can given in command line argument, default is 0. -# -# Usage example: ./collect.py https://trepo.tuni.fi/ 2000 > trepo.ndjson 2> trepo.log - - -url_base = sys.argv[1].strip("/") - - -def get_from_item(spec): - elem = item.xpath(spec) - return elem[0].text if (elem and elem[0].text is not None) else "" - - -headers = {"User-Agent": "Annif-analysis-collect"} -start = int(sys.argv[2]) if len(sys.argv) == 3 else 0 -rpp = 100 # results per page -cnt = 0 -while True: - url_query = ( - # f"/open-search/?query=*&sort_by=3&order=desc&start={start}&rpp={rpp}&format=kkf" - f"/open-search/?query=*&sort_by=3&order=asc&start={start}&rpp=100&format=kkf" - ) - url = url_base + url_query - print(f"Request {cnt}, performing call with url {url}", file=sys.stderr) - try: - resp = requests.get(url, headers=headers) - except Exception as e: - print(traceback.format_exc(), file=sys.stderr) - print("Waiting for 10 seconds before retrying...", file=sys.stderr) - time.sleep(10) - else: - try: - xml_string = resp.text - root = etree.fromstring(bytes(xml_string, encoding="utf8")) - if len(root.xpath("//item")) == 0: - print("No records received, all done.", file=sys.stderr) - break - for item in root.xpath("//item"): - my_dict = { - "title": get_from_item('metadata[@element="title"]'), - "id": get_from_item( - 'metadata[@element="identifier" and @qualifier="uri"]' - ), - "lang": get_from_item('metadata[@element="language"]'), - "type": get_from_item( - 'metadata[@element="type"]' - ), - "type_level": get_from_item( - 'metadata[@element="type" and @qualifier="ontasot"]' - ), - "faculty": get_from_item( - 'metadata[@element="contributor" and @qualifier="faculty"]' - ), - "discipline": get_from_item( - 'metadata[@element="subject" and @qualifier="discipline"]' - ), - "degreeprogram": get_from_item( - 'metadata[@element="subject" and @qualifier="degreeprogram"]' - ), - "date_accessioned": get_from_item( - 'metadata[@element="date" and @qualifier="accessioned"]' - ), - "date_issued": get_from_item( - 'metadata[@element="date" and @qualifier="issued"]' - ), - "suggestions": get_from_item('metadata[@element="suggestions"]').split( - "|" - ), - "subjects_yso": [ - s.text - for s in item.xpath( - 'metadata[@element="subject" and @qualifier="yso"]' - ) - ], - "subjects_none": [ - s.text - for s in item.xpath( - 'metadata[@element="subject" and @qualifier=""]' - ) - ], - "subjects_all": [ - s.text for s in item.xpath('metadata[@element="subject"]') - ], - } - # print data as json object - print(json.dumps(my_dict)) - except Exception as e: - print(traceback.format_exc(), file=sys.stderr) - print("Failed parsing, continue to next page.", file=sys.stderr) - - start += rpp - cnt += 1 - time.sleep(5) - # if cnt >= 0: - # break diff --git a/requirements.txt b/requirements.txt deleted file mode 100644 index 8f781938494dd6fea7dc28e78f2811e9e0c3c895..0000000000000000000000000000000000000000 --- a/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -annif[fasttext,omikuji,nn,voikko,pycld3]==1.0.* diff --git a/sync-model-data-ocp.sh b/sync-model-data-ocp.sh deleted file mode 100644 index 6fd3260c3cf0d758969d598d627672d60e10ca17..0000000000000000000000000000000000000000 --- a/sync-model-data-ocp.sh +++ /dev/null @@ -1,32 +0,0 @@ -#!/bin/bash - -# Runs rsync to transfer model data from the current directory to an OpenShift volume -# that is attached to a pod which is running Annif. The instance -# {api-annif-org,ai-finto-fi, etc.} to transfer to is given as the argument. -# You need to be logged to the cluster with the oc tool. - -set -e - -if [ $# -ne 1 ] - then - echo "Not enough arguments; argument 1: destination_instance" - exit 1 -fi - -pod=$(oc get pods -l app.kubernetes.io/instance=$1,app.kubernetes.io/name=annif -o name) - -if [[ $pod = *[[:space:]]* ]] - then - echo "Multiple pod exists; using first" - pod=(${pod//$'\n'/ }) -fi -echo "Target is "$pod -pod=${pod#pod/} -if [ -z "${pod}" ] - then - echo "No target pod found" - exit 1 -fi - -rsync --rsh='oc rsh' -avrL --exclude="*train*" --exclude="*zip" --inplace projects.d $pod:/annif-projects -rsync --rsh='oc rsh' -avrL --exclude="*train*" --exclude="*zip" --inplace data/{projects,vocabs} $pod:/annif-projects/data diff --git a/vocabs/yso.zip b/vocabs/yso.zip deleted file mode 100644 index 16a0ef0b5a2294a258c90871672d622af2dd6a52..0000000000000000000000000000000000000000 --- a/vocabs/yso.zip +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:55343674c8324f578e192e5219e2f69121924fa6f482abe7e14844852ff89be0 -size 61626689 diff --git a/yso-mllm-en.cfg b/yso-mllm-en.cfg deleted file mode 100644 index 248b40c663d09e342e9fef15935024e8b4bc27d2..0000000000000000000000000000000000000000 --- a/yso-mllm-en.cfg +++ /dev/null @@ -1,10 +0,0 @@ -[yso-mllm-en] -name = YSO MLLM English -language = en -backend = mllm -analyzer = snowball(english) -vocab = yso -limit = 1000 -transform = limit(2500000) -access = hidden -