Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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
+translate-ru_vep-1_0.argosmodel filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1 @@
----
-license: cc-by-sa-4.0
----
+# Russian - Veps version 1.0

config.yml

@@ -0,0 +1,131 @@
+accum_count: 100
+accum_steps: 0
+adam_beta2: 0.998
+attention_dropout: 0.1
+batch_size: 1500
+batch_type: tokens
+bucket_size: 49304
+data:
+  27e6308e0faf12cd1b4bed73708fad6f:
+    path_src: dataset/vep-ru/et-synt/target.txt
+    path_tgt: dataset/vep-ru/et-synt/source.txt
+    src_prefix: et_s_Latn
+    tgt_prefix: ''
+    transforms: &id001
+    - sentencepiece
+    - filtertoolong
+    - prefix
+    weight: 13
+  2caaf1b1e5e15d4a06b5b77c438ba0bc:
+    path_src: dataset/vep-ru/fin-original/target.txt
+    path_tgt: dataset/vep-ru/fin-original/source.txt
+    src_prefix: fi_Latn
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 13
+  3def7622ce4622202126c8828c6fae3f:
+    path_src: dataset/vep-ru/fin-synt/target.txt
+    path_tgt: dataset/vep-ru/fin-synt/source.txt
+    src_prefix: fi_s_Latn
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 13
+  517216e92cd45eb876c79dc3bafaad18:
+    path_src: dataset/vep-ru/var/target.txt
+    path_tgt: dataset/vep-ru/var/source.txt
+    src_prefix: ''
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 6
+  815e8e066d03025e82529ef5c5ad232c:
+    path_src: dataset/vep-ru/vep/target.txt
+    path_tgt: dataset/vep-ru/vep/source.txt
+    src_prefix: ''
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 5
+  8e6d221db9fc70bb3fd1104c24c8f25c:
+    path_src: dataset/vep-ru/vep-dic/target.txt
+    path_tgt: dataset/vep-ru/vep-dic/source.txt
+    src_prefix: ''
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 1
+  f080464d79eb04a44e9947db1b54c17b:
+    path_src: dataset/vep-ru/et-original/target.txt
+    path_tgt: dataset/vep-ru/et-original/source.txt
+    src_prefix: et_Latn
+    tgt_prefix: ''
+    transforms: *id001
+    weight: 13
+  valid:
+    path_src: run/ru_vep-1.0/src-val.txt
+    path_tgt: run/ru_vep-1.0/tgt-val.txt
+    transforms: *id001
+dec_layers: 20
+decay_method: rsqrt
+decoder_type: transformer
+dropout: 0.1
+dropout_steps: 0
+early_stopping: 0
+enc_layers: 20
+encoder_type: transformer
+gpu_ranks:
+- 0
+- 1
+heads: 8
+hidden_size: 512
+keep_checkpoint: 15
+label_smoothing: 0.1
+learning_rate: 0.0015
+max_generator_batches: 2
+max_grad_norm: 0
+max_relative_positions: 20
+model_dtype: fp16
+normalization: tokens
+num_workers: 6
+optim: pagedadamw8bit
+param_init: 0
+param_init_glorot: true
+pos_ffn_activation_fn: gated-gelu
+position_encoding: false
+queue_size: 10000
+reset_optim: all
+rnn_size: 512
+save_checkpoint_steps: 100
+save_data: run/ru_vep-1.0/opennmt
+save_model: run/ru_vep-1.0/opennmt/openmt.model
+share_decoder_embeddings: true
+share_embeddings: true
+share_vocab: true
+skip_empty_level: silent
+src_onmttok_kwargs:
+  lang: ru
+  mode: none
+src_seq_length: 185
+src_subword_alpha: 0.0
+src_subword_model: run/ru_vep-1.0/sentencepiece.model
+src_subword_nbest: 1
+src_subword_type: sentencepiece
+src_vocab: run/ru_vep-1.0/opennmt/openmt.vocab
+src_vocab_size: 32000
+tgt_onmttok_kwargs:
+  lang: vep
+  mode: none
+tgt_seq_length: 185
+tgt_subword_alpha: 0.0
+tgt_subword_model: run/ru_vep-1.0/sentencepiece.model
+tgt_subword_nbest: 1
+tgt_subword_type: sentencepiece
+tgt_vocab: run/ru_vep-1.0/opennmt/openmt.vocab
+tgt_vocab_size: 32000
+train_steps: 1500
+transformer_ff: 6144
+update_vocab: 'False'
+valid_batch_size: 64
+valid_metrics:
+- BLEU
+valid_steps: 100
+warmup_steps: 200
+word_vec_size: 512
+world_size: 2

ctranslate2-model/config.json

@@ -0,0 +1,9 @@
+{
+  "add_source_bos": false,
+  "add_source_eos": false,
+  "bos_token": "<s>",
+  "decoder_start_token": "<s>",
+  "eos_token": "</s>",
+  "layer_norm_epsilon": null,
+  "unk_token": "<unk>"
+}

ctranslate2-model/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b83cbd1910ea8adf5cfb164d7e26413996f65af435f8240b5bb5b9ab9ed3abf
+size 460469083

metadata.json

@@ -0,0 +1 @@
+{"package_version": "1.0", "argos_version": "1.9.0", "from_code": "ru", "from_name": "Russian", "to_code": "vep", "to_name": "Veps"}

opennmt/openmt.model_averaged.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45e107a0a87c5f5794f38bb958686dffa1dcfd1f94decf4514b6d5f046e0fbb7
+size 1960760662

sentencepiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a602803c84fef522a000aa6eedfb440360cf2ffc9b5996cc57c8376be84eb6c6
+size 883660

stanza/ru/tokenize/syntagrus.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a28c7770bd33220f61b5ab060c9aaebe62354771f557683875602d1c2a38fefc
+size 638146

translate-ru_vep-1_0.argosmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20738ffc11bb4b3a97710b3758e7979f309c6a6f4b8527c6a590a9c996226ecd
+size 463123318