Hugging Face's logo

kperkins411
/
msmarco-distilbert-base-v2_triplet_legal

Sentence Similarity
sentence-transformers
Safetensors
distilbert
feature-extraction
Generated from Trainer
dataset_size:88018
loss:TripletLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
msmarco-distilbert-base-v2_triplet_legal / README.md
kperkins411's picture
kperkins411
Add new SentenceTransformer model.
623a245 verified
preview code
|
raw
history blame
38.9 kB
---
base_model: sentence-transformers/msmarco-distilbert-base-v2
datasets: []
language: []
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
- dot_accuracy@1
- dot_accuracy@3
- dot_accuracy@5
- dot_accuracy@10
- dot_precision@1
- dot_precision@3
- dot_precision@5
- dot_precision@10
- dot_recall@1
- dot_recall@3
- dot_recall@5
- dot_recall@10
- dot_ndcg@10
- dot_mrr@10
- dot_map@100
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:88018
- loss:TripletLoss
widget:
- source_sentence: How should SpinRecords.com notify NETTAXI of a potential indemnifiable
claim?
sentences:
- '4. The Company shall have no obligations to Verenium with respect to the use
of such information, or disclosure to others not party to this Agreement, of such
information which: (d) is rightfully and in good faith developed by Company independently
of any disclosures made under this Agreement, as evidenced by Company’s competent
written records; or '
- 13. Changes To This Privacy Policy We may update this Privacy Policy to reflect
changes to our information practices. If we make any material changes we will
notify you by email (sent to the e-mail address specified in your account) or
by means of a notice on the Services prior to the change becoming effective. We
encourage you to periodically review this page for the latest information on our
privacy practices.
- 7.2 Indemnification by NETTAXI. NETTAXI shall defend, indemnify and ----------------------------
hold SpinRecords.com harmless from any and all damages, liabilities, costs and
expenses (including, but not limited to reasonable attorneys' fees) incurred
by SpinRecords.com as a result of (1) any breach of this Agreement; (ii) any
claim that the NETTAXI Brand Features or any part thereof, infringes or
misappropriates any Intellectual Property Right of a third party; or (iii) any
claim arising out of Spinrecords.com's display of the NETTAXI Brand Features
SpinRecords.com shall provide NETTAXI with written notice of the claim and
permit NETTAXI to control the defense, settlement, adjustment or compromise
of any such claim. SpinRecords.com may employ counsel at its own expense to assist
it with respect to any such claim; provided, however, that if such counsel
is necessary because of a conflict of interest of either NETTAXI or its counsel
or because NETTAXI does not assume control, NETTAXI will bear the expense of
such counsel.
- source_sentence: What types of advertisements does Crazy Labs accept within their
apps?
sentences:
- Each party each agrees that it will not knowingly do anything inconsistent with
the other party's ownership of such party's intellectual property, including without
limitation, questioning the validity of that party's Trademarks or registering
or attempting to register the other party's Trademarks in its own name or that
of any other firm, person or corporation.
- '11. Advertisements We accept advertisements, in various formats (such as banners,
interstitials, rewarded videos, etc.) from third parties ad networks which may
be displayed in our Crazy Labs Apps. These third parties ad networks may collect
and use, inter alia (i) information about your visits to Crazy Labs Apps in connection
with such marketing, sales and advertising activities; and (ii) geographic tracking
and carrier network preferences. (iii) information, such as age, gender and logged
from device to ensure that appropriate advertising is presented within the App
and calculate or control the number of views of an ad, and/or deliver advertisements
relating to User''s interests, and measure the effectiveness of advertisements
campaigns. The delivery of advertisements to you may be based on IP address, device
identifiers and other Personal Information gathered during your use of the Crazy
Labs Apps. Note that third parties ad networks which are referred to in relation
to the Crazy Labs Apps may include third parties service providers, such as Facebook
and other ad networks, in addition to those which are listed in the following
link: https://www.tabtale.com/3rdparties/. Note that if you click on any of these
advertisements, the advertisers may use cookies and other web-tracking technologies
(such as tracking pixel agent or visitor identification technology, etc.) on your
device to collect data regarding advertisement performance, your interaction with
such advertisements and our Crazy Labs Apps and your interests (which may include,
non-personal and/or personal information (such as, device and network information,
unique identifiers, gender, age and geo-location) about you) in order to serve
you advertisements, including targeted advertisements, and for the legitimate
business interests of such Third Parties ad networks. We recommend that you review
the terms of use and privacy policy of any third party advertisers with whom you
are interacting before doing so. Their privacy policy, not ours, will apply to
any of those interactions.'
- We want our advertising to be as relevant and interesting as the other information
you find on our Services. With this in mind, we use all of the information we
have about you to show you relevant ads. We do not share information that personally
identifies you (personally identifiable information is information like name or
email address that can by itself be used to contact you or identifies who you
are) with advertising, measurement or analytics partners unless you give us permission.
We may provide these partners with information about the reach and effectiveness
of their advertising without providing information that personally identifies
you, or if we have aggregated the information so that it does not personally identify
you. For example, we may tell an advertiser how its ads performed, or how many
people viewed their ads or installed an app after seeing an ad, or provide non-personally
identifying demographic information (such as 25 year old female, in Madrid, who
likes software engineering) to these partners to help them understand their audience
or customers, but only after the advertiser has agreed to abide by our advertiser
guidelines.
- source_sentence: 9.2 Nature of the Association. The parties herein are engaged
as independent entities in accordance with this Agreement, and there exists no
intention to forge any alternate form of association, such as a partnership, franchise,
joint venture, agency, employer/employee relationship, fiduciary connection, or
any other specific relationship. Each party is precluded from conducting themselves
in any way that might suggest or insinuate any association different from that
of an independent entity, nor shall either party possess the authority to obligate
or commit the other party in any manner.
sentences:
- 'The parties hereby grant to each other non-exclusive, fully-paid, royalty-free
licenses to utilize the other party''s trademarks, as follows: (a) Biocept Trademarks.
To facilitate the promotion and performance of Tests, during the Term Biocept
hereby grants Life Technologies a non-exclusive, royalty-free, non-transferable
license to use the Biocept Trademarks solely for<omitted>use in connection with
the promotion and performance of the Tests in the Territory.'
- This Agreement may not be assigned or otherwise transferred, nor may any right
or obligations hereunder be assigned or transferred, by either Party without the
prior written consent of the other Party; provided, however, that Licensor may,
without such consent, assign this Agreement and its rights and obligations hereunder,
in whole or in part, to an Affiliate or in connection with the transfer or sale
of all or substantially all of its assets related to the Licensed Product or the
business relating thereto, or in the event of its merger or consolidation or change
in control or similar transaction.
- 9.2 Relationship of Parties. The parties are independent contractors -------------------------
under this Agreement and no other relationship is intended, including
a partnership, franchise, joint venture, agency, employer/employee, fiduciary,
master/servant relationship, or other special relationship. Neither party shall
act in a manner which expresses or implies a relationship other than that
of independent contractor, nor bind the other party.
- source_sentence: In which section can I find the specifics of the 'Initial Term'?
sentences:
- 1.22 "Initial Term" has the meaning set forth in Section 8.1.
- If the Reseller sells less than 50% of any year's Annual Milestone, Todos, in
its sole discretion, may either (a) cancel the Reseller's exclusivity, and market,
distribute, and sell the Products in the Territory directly or indirectly through
other distributors and resellers, while leaving the Reseller with a non-exclusive
right to distribute and sell the Products for the remainder of the term, or (b)
terminate the Agreement upon one hundred eighty (180) days prior written notice,
provided that the Reseller does not cure its failure to achieve 50% of the applicable
year's Annual Milestone within the 180-day notice period.
- 6 Term and Termination.
- source_sentence: In what circumstances can FCE assume responsibility for a Program
Patent?
sentences:
- We may also collect anonymous, statistical data from users of the Services, such
as a user's browser version, operating system version, country, page loading time,
type of device, number of visits, time using the Services, network, demographic
estimates, flow-through the website and/or Services or referral source, which
may then be aggregated. We may use non-personal data that we collect from you
to improve the Services or to support advertising services. For registered users,
this anonymous, statistical data may include that relating to their activities,
such as high scores, game rankings, league rankings, game challenges, avatars
etc.
- Notwithstanding the foregoing, in the event ExxonMobil decides not to prosecute,
defend, enforce, maintain or decides to abandon any Program Patent, then ExxonMobil
will provide notice thereof to FCE, and FCE will then have the right, but not
the obligation, to prosecute or maintain the Program Patent and sole responsibility
for the continuing costs, taxes, legal fees, maintenance fees and other fees associated
with that Program Patent.
- 4. Limitation of Liability of the Sponsor. The Sponsor shall not be liable for
any error of judgment or mistake of law or for any act or omission in the oversight,
administration or management of the Trust or the performance of its duties hereunder,
except for willful misfeasance, bad faith or gross negligence in the performance
of its duties, or by reason of the reckless disregard of its obligations and duties
hereunder. As used in this Section 4, the term "Sponsor" shall include Domini
and/or any of its affiliates and the directors, officers and employees of Domini
and/or any of its affiliates.
model-index:
- name: SentenceTransformer based on sentence-transformers/msmarco-distilbert-base-v2
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: msmarco distilbert base v2
type: msmarco-distilbert-base-v2
metrics:
- type: cosine_accuracy@1
value: 0.6422082459818309
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.8230607966457023
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.872816212438854
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9382250174703005
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.6422082459818309
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.27435359888190075
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.1745632424877708
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09382250174703004
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.6422082459818309
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8230607966457023
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.872816212438854
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9382250174703005
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.790195916846684
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.7427224274289222
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.7454587747656682
name: Cosine Map@100
- type: dot_accuracy@1
value: 0.6317260656883298
name: Dot Accuracy@1
- type: dot_accuracy@3
value: 0.8204053109713487
name: Dot Accuracy@3
- type: dot_accuracy@5
value: 0.8735150244584207
name: Dot Accuracy@5
- type: dot_accuracy@10
value: 0.9375262054507337
name: Dot Accuracy@10
- type: dot_precision@1
value: 0.6317260656883298
name: Dot Precision@1
- type: dot_precision@3
value: 0.27346843699044954
name: Dot Precision@3
- type: dot_precision@5
value: 0.17470300489168414
name: Dot Precision@5
- type: dot_precision@10
value: 0.09375262054507337
name: Dot Precision@10
- type: dot_recall@1
value: 0.6317260656883298
name: Dot Recall@1
- type: dot_recall@3
value: 0.8204053109713487
name: Dot Recall@3
- type: dot_recall@5
value: 0.8735150244584207
name: Dot Recall@5
- type: dot_recall@10
value: 0.9375262054507337
name: Dot Recall@10
- type: dot_ndcg@10
value: 0.7853441093620476
name: Dot Ndcg@10
- type: dot_mrr@10
value: 0.7364890242143864
name: Dot Mrr@10
- type: dot_map@100
value: 0.7392413927907737
name: Dot Map@100
---
# SentenceTransformer based on sentence-transformers/msmarco-distilbert-base-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/msmarco-distilbert-base-v2](https://huggingface.co/sentence-transformers/msmarco-distilbert-base-v2). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/msmarco-distilbert-base-v2](https://huggingface.co/sentence-transformers/msmarco-distilbert-base-v2) <!-- at revision 741fcf2d6eabaf0927bfe49c6d9c577df95d3c40 -->
- **Maximum Sequence Length:** 350 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 350, 'do_lower_case': False}) with Transformer model: DistilBertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("kperkins411/msmarco-distilbert-base-v2_triplet_legal")
# Run inference
sentences = [
'In what circumstances can FCE assume responsibility for a Program Patent?',
'Notwithstanding the foregoing, in the event ExxonMobil decides not to prosecute, defend, enforce, maintain or decides to abandon any Program Patent, then ExxonMobil will provide notice thereof to FCE, and FCE will then have the right, but not the obligation, to prosecute or maintain the Program Patent and sole responsibility for the continuing costs, taxes, legal fees, maintenance fees and other fees associated with that Program Patent.',
'4. Limitation of Liability of the Sponsor. The Sponsor shall not be liable for any error of judgment or mistake of law or for any act or omission in the oversight, administration or management of the Trust or the performance of its duties hereunder, except for willful misfeasance, bad faith or gross negligence in the performance of its duties, or by reason of the reckless disregard of its obligations and duties hereunder. As used in this Section 4, the term "Sponsor" shall include Domini and/or any of its affiliates and the directors, officers and employees of Domini and/or any of its affiliates.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Dataset: `msmarco-distilbert-base-v2`
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.6422 |
| cosine_accuracy@3 | 0.8231 |
| cosine_accuracy@5 | 0.8728 |
| cosine_accuracy@10 | 0.9382 |
| cosine_precision@1 | 0.6422 |
| cosine_precision@3 | 0.2744 |
| cosine_precision@5 | 0.1746 |
| cosine_precision@10 | 0.0938 |
| cosine_recall@1 | 0.6422 |
| cosine_recall@3 | 0.8231 |
| cosine_recall@5 | 0.8728 |
| cosine_recall@10 | 0.9382 |
| cosine_ndcg@10 | 0.7902 |
| cosine_mrr@10 | 0.7427 |
| **cosine_map@100** | **0.7455** |
| dot_accuracy@1 | 0.6317 |
| dot_accuracy@3 | 0.8204 |
| dot_accuracy@5 | 0.8735 |
| dot_accuracy@10 | 0.9375 |
| dot_precision@1 | 0.6317 |
| dot_precision@3 | 0.2735 |
| dot_precision@5 | 0.1747 |
| dot_precision@10 | 0.0938 |
| dot_recall@1 | 0.6317 |
| dot_recall@3 | 0.8204 |
| dot_recall@5 | 0.8735 |
| dot_recall@10 | 0.9375 |
| dot_ndcg@10 | 0.7853 |
| dot_mrr@10 | 0.7365 |
| dot_map@100 | 0.7392 |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 88,018 training samples
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 7 tokens</li><li>mean: 17.42 tokens</li><li>max: 104 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 102.85 tokens</li><li>max: 350 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 103.73 tokens</li><li>max: 350 tokens</li></ul> |
* Samples:
| anchor | positive | negative |
|:--------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>What happens if a Party fails to retain records for the required period?</code> | <code>Each Party will retain such records for at least three (3) years following expiration or termination of this Agreement or such longer period as may be required by applicable law or regulation.</code> | <code>Either party hereto may terminate this Agreement after the Initial Period upon at least six (6) months' prior written notice to the other party thereof.</code> |
| <code>What happens if a Party fails to retain records for the required period?</code> | <code>Each Party will retain such records for at least three (3) years following expiration or termination of this Agreement or such longer period as may be required by applicable law or regulation.</code> | <code>The Agreement may be terminated by both Parties with a notification period of *** before the end of the Initial Term of the Agreement.</code> |
| <code>What happens if a Party fails to retain records for the required period?</code> | <code>Each Party will retain such records for at least three (3) years following expiration or termination of this Agreement or such longer period as may be required by applicable law or regulation.</code> | <code>For twelve (12) months after delivery of the Master Copy of each Licensed Product to Licensee, Licensor warrants that the media in which the Licensed Products are stored shall be free from defects in materials and workmanship, assuming normal use. Licensee may return any defective media to Licensor for replacement free of charge during such twelve (12) month period.</code> |
* Loss: [<code>TripletLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters:
```json
{
"distance_metric": "TripletDistanceMetric.EUCLIDEAN",
"triplet_margin": 5
}
```
### Evaluation Dataset
#### Unnamed Dataset
* Size: 1,084 evaluation samples
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 6 tokens</li><li>mean: 20.24 tokens</li><li>max: 124 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 97.01 tokens</li><li>max: 350 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 105.03 tokens</li><li>max: 350 tokens</li></ul> |
* Samples:
| anchor | positive | negative |
|:--------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>Are Capital Contributions categorized as either 'Initial' or 'Additional' in the accounts?</code> | <code>Capital Accounts<br><br>An individual capital account (the "Capital Accounts") will be maintained for each Participant and their Initial Capital Contribution will be credited to this account. Any Additional Capital Contributions made by any Participant will be credited to that Participant's individual Capital Account.</code> | <code>Section 4.3 Deposits and Payments 19</code> |
| <code>Are Capital Contributions categorized as either 'Initial' or 'Additional' in the accounts?</code> | <code>Capital Accounts<br><br>An individual capital account (the "Capital Accounts") will be maintained for each Participant and their Initial Capital Contribution will be credited to this account. Any Additional Capital Contributions made by any Participant will be credited to that Participant's individual Capital Account.</code> | <code>Section 2.1 The Fund agrees at its own expense to execute any and all documents, to furnish any and all information, and to take any other actions that may be reasonably necessary in connection with the qualification of the Shares for sale in those states that Integrity may designate.</code> |
| <code>Are Capital Contributions categorized as either 'Initial' or 'Additional' in the accounts?</code> | <code>Capital Accounts<br><br>An individual capital account (the "Capital Accounts") will be maintained for each Participant and their Initial Capital Contribution will be credited to this account. Any Additional Capital Contributions made by any Participant will be credited to that Participant's individual Capital Account.</code> | <code>Section 1.9 Integrity shall prepare and deliver reports to the Treasurer of the Fund and to the Investment Adviser on a regular, at least quarterly, basis, showing the distribution expenses incurred pursuant to this Agreement and the Plan and the purposes therefore, as well as any supplemental reports as the Trustees, from time to time, may reasonably request.</code> |
* Loss: [<code>TripletLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters:
```json
{
"distance_metric": "TripletDistanceMetric.EUCLIDEAN",
"triplet_margin": 5
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: epoch
- `per_device_train_batch_size`: 128
- `per_device_eval_batch_size`: 128
- `learning_rate`: 2e-05
- `warmup_ratio`: 0.1
- `fp16`: True
- `load_best_model_at_end`: True
- `batch_sampler`: no_duplicates
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 128
- `per_device_eval_batch_size`: 128
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `learning_rate`: 2e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 3
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.1
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: True
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss | loss | msmarco-distilbert-base-v2_cosine_map@100 |
|:----------:|:--------:|:-------------:|:----------:|:-----------------------------------------:|
| 0 | 0 | - | - | 0.6601 |
| 0.1453 | 100 | 1.5696 | - | - |
| 0.2907 | 200 | 0.7941 | - | - |
| 0.4360 | 300 | 0.6151 | - | - |
| 0.5814 | 400 | 0.5458 | - | - |
| 0.7267 | 500 | 0.5085 | - | - |
| 0.8721 | 600 | 0.4601 | - | - |
| 1.0131 | 697 | - | 0.3492 | - |
| 1.0044 | 700 | 0.4055 | - | - |
| 1.1497 | 800 | 0.3538 | - | - |
| 1.2951 | 900 | 0.2245 | - | - |
| 1.4404 | 1000 | 0.1821 | - | - |
| 1.5858 | 1100 | 0.1761 | - | - |
| 1.7311 | 1200 | 0.1872 | - | - |
| 1.8765 | 1300 | 0.169 | - | - |
| 2.0131 | 1394 | - | 0.2674 | - |
| 2.0087 | 1400 | 0.1502 | - | - |
| 2.1541 | 1500 | 0.1416 | - | - |
| 2.2994 | 1600 | 0.0914 | - | - |
| 2.4448 | 1700 | 0.0868 | - | - |
| 2.5901 | 1800 | 0.0854 | - | - |
| 2.7355 | 1900 | 0.0905 | - | - |
| 2.8808 | 2000 | 0.0888 | - | - |
| **2.9738** | **2064** | **-** | **0.2272** | **0.7455** |
* The bold row denotes the saved checkpoint.
### Framework Versions
- Python: 3.11.9
- Sentence Transformers: 3.1.0.dev0
- Transformers: 4.41.2
- PyTorch: 2.1.2+cu121
- Accelerate: 0.31.0
- Datasets: 2.19.1
- Tokenizers: 0.19.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### TripletLoss
```bibtex
@misc{hermans2017defense,
title={In Defense of the Triplet Loss for Person Re-Identification},
author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
year={2017},
eprint={1703.07737},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->