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.

python

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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "pairwise_cos_sim" 4}

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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "pairwise_cos_sim" 4}

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

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",
}

CoSENTLoss

bibtex

@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},
}

	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