aminlouhichi/CDGSmilarity

SentenceTransformer based on sentence-transformers/LaBSE

This is a sentence-transformers model finetuned from sentence-transformers/LaBSE. 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/LaBSE
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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): Dense({'in_features': 768, 'out_features': 768, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
  (3): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("aminlouhichi/CDGSmilarity")
# Run inference
sentences = [
    'Temps partiel surcotisé',
    'Temps partiel surcotisé de droit',
    'Départ définitif - Radiation des cadres',
]
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]

Training Details

Training Dataset

Unnamed Dataset

• Size: 295 training samples
• Columns: premise, hypothesis, and label
• Approximate statistics based on the first 1000 samples:

  	premise	hypothesis	label
  type	string	string	float
  details	min: 4 tokens mean: 9.31 tokens max: 20 tokens	min: 4 tokens mean: 10.41 tokens max: 20 tokens	min: 0.9 mean: 0.95 max: 1.0

• Samples:

  premise	hypothesis	label
  Compte rendu d'entretien professionnel	Synthèse des discussions professionnelles	0.9820208462484844
  Congé Accident de trajet	Arrêt de travail pour accident de trajet	0.9755981363214147
  Retrait ou suppression du CTI (complément de traitement indiciaire)	Retrait du Complément de Traitement Indiciaire (CTI)	0.9524167934189104

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 74 evaluation samples
• Columns: premise, hypothesis, and label
• Approximate statistics based on the first 1000 samples:

  	premise	hypothesis	label
  type	string	string	float
  details	min: 4 tokens mean: 10.26 tokens max: 25 tokens	min: 5 tokens mean: 10.5 tokens max: 20 tokens	min: 0.9 mean: 0.95 max: 1.0

• Samples:

  premise	hypothesis	label
  Sanction disciplinaire	Mesure punitive suite à une violation du règlement	0.958828679924412
  Départ définitif / Radiation - Décès	Départ définitif suite au décès d'un agent	0.9003635138326387
  Nomination par intégration directe	Intégration immédiate avec nomination	0.9993378836623817

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 30
• warmup_ratio: 0.1
• fp16: True

All Hyperparameters

Click to expand

• 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
• learning_rate: 5e-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: 30
• 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: 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
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss
0.5263	10	12.4933	-
1.0526	20	10.5909	-
1.5789	30	7.0607	-
2.1053	40	4.7061	-
2.6316	50	4.7957	-
3.1579	60	4.624	-
3.6842	70	4.7854	-
4.2105	80	4.5902	-
4.7368	90	4.7051	-
5.2632	100	4.5562	4.6756
5.7895	110	4.6376	-
6.3158	120	4.4501	-
6.8421	130	4.5993	-
7.3684	140	4.4878	-
7.8947	150	4.5443	-
8.4211	160	4.3091	-
8.9474	170	4.6699	-
9.4737	180	4.3727	-
10.0	190	4.3888	-
10.5263	200	4.5099	5.3597
11.0526	210	4.3427	-
11.5789	220	4.4409	-
12.1053	230	4.3151	-
12.6316	240	4.3522	-
13.1579	250	4.3133	-
13.6842	260	4.3842	-
14.2105	270	4.2708	-
14.7368	280	4.387	-
15.2632	290	4.1131	-
15.7895	300	4.3394	5.5109
16.3158	310	4.2948	-
16.8421	320	4.3413	-
17.3684	330	4.1427	-
17.8947	340	4.5521	-
18.4211	350	4.2146	-
18.9474	360	4.2039	-
19.4737	370	4.1412	-
20.0	380	4.0869	-
20.5263	390	4.4763	-
21.0526	400	3.9572	5.7054
21.5789	410	4.2114	-
22.1053	420	4.2651	-
22.6316	430	4.2231	-
23.1579	440	4.0521	-
23.6842	450	4.3246	-
24.2105	460	3.9145	-
24.7368	470	4.1701	-
25.2632	480	4.0958	-
25.7895	490	4.1177	-
26.3158	500	4.2388	6.3162
26.8421	510	4.3043	-
27.3684	520	3.9634	-
27.8947	530	4.117	-
28.4211	540	4.1732	-
28.9474	550	4.1243	-
29.4737	560	3.7898	-
30.0	570	4.0227	-

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.0
• Transformers: 4.41.1
• PyTorch: 2.3.0+cu121
• Accelerate: 0.30.1
• Datasets: 2.19.1
• Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}