Hugging Face's logo

cherifkhalifah
/
finetuned-snli-MiniLM-L12-v2-100k-en-fr

Sentence Similarity
sentence-transformers
Safetensors
bert
feature-extraction
Generated from Trainer
dataset_size:100000
loss:CosineSimilarityLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
finetuned-snli-MiniLM-L12-v2-100k-en-fr / README.md
cherifkhalifah's picture
cherifkhalifah
Finetuned model on SNLI
b8be1b1 verified
preview code
|
raw
history blame
18.9 kB
---
base_model: sentence-transformers/all-MiniLM-L12-v2
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:100000
- loss:CosineSimilarityLoss
widget:
- source_sentence: NIPA personal income includes pension contributions by employers
in the year income is earned , and benefits paid at retirement are not a component
of NIPA income .
sentences:
- While not the only makeup of income , NIPA is one of the more well known income
distinctions .
- Les temples de karnak et de Louxor ont été démolis pour faire place à des projets
de construction en Cisjordanie .
- Les restaurants sont tenus à des règles strictes pour contenir leur odeur .
- source_sentence: right right you know the one that 's one reason we bought a house
here in Plano we were hoping you know well the school district 's gonna be good
you know for resale value and so on and so forth but
sentences:
- We moved to Plano because we thought the school district was good .
- These and those .
- L' obsession a suscité une suggestion que tous étaient des boucs émissaires de
la guerre .
- source_sentence: Dans l' homme invisible , le talentueux dixième narrateur doit
surmonter non seulement les différentes idéologies qui lui sont présentées comme
masques ou subversions d' identité , mais aussi les différents rôles et prescriptions
pour le leadership que sa propre race lui souhaite de réaliser .
sentences:
- '" We ''re too uptight now ! " Said Tommy'
- Le talentueux dixième narrateur doit surmonter les idéologies .
- Saddam is not taking advantage of the current Arab love towards the United States
- source_sentence: Les lacunes trouvées au cours de la surveillance en cours ou au
moyen d' évaluations distinctes devraient être communiquées à l' individu responsable
de la fonction et à au moins un niveau de gestion au-dessus de cet individu .
sentences:
- L' économie diminuera également si les conditions du marché changent .
- The Watergate comparison wasn 't just for Democratic bashing .
- Il n' y a pas lieu de signaler les lacunes .
- source_sentence: it looks fertile and it it um i mean it rains enough they have
the climate and the rain and if not it 's like i 've been to Saint Thomas and
it just starts from the ocean up
sentences:
- Il n' a jamais triché .
- They don 't know how to do it .
- They have the rain and the climate so I imagine the lands would be fertile .
model-index:
- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
results:
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: snli dev
type: snli-dev
metrics:
- type: pearson_cosine
value: 0.3725313255221131
name: Pearson Cosine
- type: spearman_cosine
value: 0.3729470854776107
name: Spearman Cosine
- type: pearson_manhattan
value: 0.3650227128515394
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.37250760289182383
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.36567325497563746
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.37294699995093694
name: Spearman Euclidean
- type: pearson_dot
value: 0.3725313190046259
name: Pearson Dot
- type: spearman_dot
value: 0.3729474276296007
name: Spearman Dot
- type: pearson_max
value: 0.3725313255221131
name: Pearson Max
- type: spearman_max
value: 0.3729474276296007
name: Spearman Max
---
# SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2). It maps sentences & paragraphs to a 384-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/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2) <!-- at revision a05860a77cef7b37e0048a7864658139bc18a854 -->
- **Maximum Sequence Length:** 128 tokens
- **Output Dimensionality:** 384 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': 128, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, '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})
(2): Normalize()
)
```
## 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("cherifkhalifah/finetuned-snli-MiniLM-L12-v2-100k-en-fr")
# Run inference
sentences = [
"it looks fertile and it it um i mean it rains enough they have the climate and the rain and if not it 's like i 've been to Saint Thomas and it just starts from the ocean up",
'They have the rain and the climate so I imagine the lands would be fertile .',
"They don 't know how to do it .",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]
# 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
#### Semantic Similarity
* Dataset: `snli-dev`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:-------------------|:-----------|
| pearson_cosine | 0.3725 |
| spearman_cosine | 0.3729 |
| pearson_manhattan | 0.365 |
| spearman_manhattan | 0.3725 |
| pearson_euclidean | 0.3657 |
| spearman_euclidean | 0.3729 |
| pearson_dot | 0.3725 |
| spearman_dot | 0.3729 |
| pearson_max | 0.3725 |
| **spearman_max** | **0.3729** |
<!--
## 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: 100,000 training samples
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
* Approximate statistics based on the first 1000 samples:
| | sentence_0 | sentence_1 | label |
|:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:--------------------------------------------------------------|
| type | string | string | float |
| details | <ul><li>min: 5 tokens</li><li>mean: 35.27 tokens</li><li>max: 128 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 18.46 tokens</li><li>max: 66 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.5</li><li>max: 1.0</li></ul> |
* Samples:
| sentence_0 | sentence_1 | label |
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------|:-----------------|
| <code>Natalia M' a regardé .</code> | <code>Natalia a regardé et attend que je lui donne l' épée .</code> | <code>0.5</code> |
| <code>And he sounded sincere .</code> | <code>He sounded sincere.He was sounding sincere in his words .</code> | <code>0.0</code> |
| <code>There 's a small zoo area where you can see snakes , lizards , birds of prey , wolves , hyenas , foxes , and various desert cats , including cheetahs and leopards .</code> | <code>The zoo is home to some endangered desert animals .</code> | <code>0.5</code> |
* Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
```json
{
"loss_fct": "torch.nn.modules.loss.MSELoss"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `num_train_epochs`: 4
- `fp16`: True
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 4
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `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`: False
- `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
- `eval_on_start`: False
- `eval_use_gather_object`: False
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | Training Loss | snli-dev_spearman_max |
|:------:|:-----:|:-------------:|:---------------------:|
| 0.08 | 500 | 0.2008 | 0.0433 |
| 0.16 | 1000 | 0.1757 | 0.1024 |
| 0.24 | 1500 | 0.1732 | 0.1503 |
| 0.32 | 2000 | 0.1685 | 0.2168 |
| 0.4 | 2500 | 0.1702 | 0.2206 |
| 0.48 | 3000 | 0.1676 | 0.2117 |
| 0.56 | 3500 | 0.1637 | 0.2624 |
| 0.64 | 4000 | 0.1636 | 0.2169 |
| 0.72 | 4500 | 0.1608 | 0.0051 |
| 0.8 | 5000 | 0.1601 | 0.2236 |
| 0.88 | 5500 | 0.1597 | 0.2471 |
| 0.96 | 6000 | 0.1596 | 0.2934 |
| 1.0 | 6250 | - | 0.2905 |
| 1.04 | 6500 | 0.1602 | 0.3001 |
| 1.12 | 7000 | 0.1571 | 0.3116 |
| 1.2 | 7500 | 0.1588 | 0.3145 |
| 1.28 | 8000 | 0.1562 | 0.3304 |
| 1.3600 | 8500 | 0.1548 | 0.3376 |
| 1.44 | 9000 | 0.156 | 0.3359 |
| 1.52 | 9500 | 0.1552 | 0.3194 |
| 1.6 | 10000 | 0.153 | 0.3474 |
| 1.6800 | 10500 | 0.1529 | 0.3220 |
| 1.76 | 11000 | 0.1518 | 0.3255 |
| 1.8400 | 11500 | 0.1499 | 0.3332 |
| 1.92 | 12000 | 0.1524 | 0.3521 |
| 2.0 | 12500 | 0.1512 | 0.3425 |
| 2.08 | 13000 | 0.1514 | 0.3462 |
| 2.16 | 13500 | 0.1516 | 0.3414 |
| 2.24 | 14000 | 0.1532 | 0.3453 |
| 2.32 | 14500 | 0.1459 | 0.3699 |
| 2.4 | 15000 | 0.1524 | 0.3576 |
| 2.48 | 15500 | 0.1506 | 0.3418 |
| 2.56 | 16000 | 0.1488 | 0.3559 |
| 2.64 | 16500 | 0.1486 | 0.3597 |
| 2.7200 | 17000 | 0.1469 | 0.3552 |
| 2.8 | 17500 | 0.1448 | 0.3459 |
| 2.88 | 18000 | 0.1458 | 0.3503 |
| 2.96 | 18500 | 0.1468 | 0.3647 |
| 3.0 | 18750 | - | 0.3611 |
| 3.04 | 19000 | 0.1472 | 0.3741 |
| 3.12 | 19500 | 0.1457 | 0.3603 |
| 3.2 | 20000 | 0.147 | 0.3576 |
| 3.2800 | 20500 | 0.1451 | 0.3663 |
| 3.36 | 21000 | 0.1438 | 0.3734 |
| 3.44 | 21500 | 0.1471 | 0.3698 |
| 3.52 | 22000 | 0.1462 | 0.3646 |
| 3.6 | 22500 | 0.1436 | 0.3740 |
| 3.68 | 23000 | 0.1441 | 0.3696 |
| 3.76 | 23500 | 0.1423 | 0.3636 |
| 3.84 | 24000 | 0.1411 | 0.3713 |
| 3.92 | 24500 | 0.1438 | 0.3706 |
| 4.0 | 25000 | 0.1421 | 0.3729 |
### Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.1.1
- Transformers: 4.44.2
- PyTorch: 2.4.1+cu121
- Accelerate: 0.34.2
- Datasets: 3.0.1
- 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",
}
```
<!--
## 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.*
-->