--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:5302 - loss:MultipleNegativesRankingLoss base_model: intfloat/multilingual-e5-base datasets: - Lettria/GRAG-GO-IDF-Only-Pos pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - pearson_cosine - spearman_cosine - cosine_accuracy - cosine_accuracy_threshold - cosine_f1 - cosine_f1_threshold - cosine_precision - cosine_recall - cosine_ap - cosine_mcc model-index: - name: SentenceTransformer based on intfloat/multilingual-e5-base results: - task: type: semantic-similarity name: Semantic Similarity dataset: name: EmbeddingSimEval type: EmbeddingSimEval metrics: - type: pearson_cosine value: .nan name: Pearson Cosine - type: spearman_cosine value: .nan name: Spearman Cosine - task: type: binary-classification name: Binary Classification dataset: name: BinaryClassifEval type: BinaryClassifEval metrics: - type: cosine_accuracy value: 0.8 name: Cosine Accuracy - type: cosine_accuracy_threshold value: 0.8184928297996521 name: Cosine Accuracy Threshold - type: cosine_f1 value: 0.888888888888889 name: Cosine F1 - type: cosine_f1_threshold value: 0.8184928297996521 name: Cosine F1 Threshold - type: cosine_precision value: 1.0 name: Cosine Precision - type: cosine_recall value: 0.8 name: Cosine Recall - type: cosine_ap value: 1.0 name: Cosine Ap - type: cosine_mcc value: 0.0 name: Cosine Mcc --- # SentenceTransformer based on intfloat/multilingual-e5-base This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) on the [grag-go-idf-only-pos](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos) dataset. 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:** [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 dimensions - **Similarity Function:** Cosine Similarity - **Training Dataset:** - [grag-go-idf-only-pos](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos) ### 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': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel (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}) (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("debug") # Run inference sentences = [ 'The weather is lovely today.', "It's so sunny outside!", 'He drove to the stadium.', ] 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] ``` ## Evaluation ### Metrics #### Semantic Similarity * Dataset: `EmbeddingSimEval` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:--------| | pearson_cosine | nan | | **spearman_cosine** | **nan** | #### Binary Classification * Dataset: `BinaryClassifEval` * Evaluated with [BinaryClassificationEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator) | Metric | Value | |:--------------------------|:--------| | cosine_accuracy | 0.8 | | cosine_accuracy_threshold | 0.8185 | | cosine_f1 | 0.8889 | | cosine_f1_threshold | 0.8185 | | cosine_precision | 1.0 | | cosine_recall | 0.8 | | **cosine_ap** | **1.0** | | cosine_mcc | 0.0 | ## Training Details ### Training Dataset #### grag-go-idf-only-pos * Dataset: [grag-go-idf-only-pos](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos) at [9743952](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos/tree/9743952a5d02847c83f30a59009b3231c56871a3) * Size: 5,302 training samples * Columns: sentence1, sentence2, and label * Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | label | |:--------|:-------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------| | type | string | string | int | | details | | | | * Samples: | sentence1 | sentence2 | label | |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------|:---------------| | Procédures et démarches: Le dépôt des demandes de subvention se fait en ligne sur la plateforme régionale mesdemarches.iledefrance.fr : Session de dépôt unique pour les nouvelles demandes : du 30 septembre au 4 novembre 2024 (11 heures) pour des festivals qui se déroulent entre le 1er mars 2025 et le 28 février 2026 (vote à la CP de mars 2025). Pour les demandes de renouvellement, un mail est envoyé aux structures concernées par le service du Spectacle vivant en amont de chaque session de dépôt.
Bénéficiaires: Professionnel - Culture, Association - Fondation, Association - Régie par la loi de 1901, Association - ONG, Collectivité ou institution - Communes de 10 000 à 20 000 hab, Collectivité ou institution - Autre (GIP, copropriété, EPA...), Collectivité ou institution - Communes de 2000 à 10 000 hab, Collectivité ou institution - Communes de < 2000 hab, Collectivité ou institution - Communes de > 20 000 hab, Collectivité ou institution - Département, Collectivité ou institution - EPC... | [Collectivité ou institution - EPCI](bénéficiaire) --- PEUT_BÉNÉFICIER ---> [demandes de subvention](procédure) | 1 | | Type de project: Dans le cadre de sa stratégie « Impact 2028 », la Région s’engage dans la défense de la souveraineté industrielle en renforçant son soutien à une industrie circulaire et décarbonée, porteuse d’innovations et créatrice d’emplois. PM'up Jeunes pousses industrielles soutient les projets d’implantation d’une première usine tournée vers la décarbonation, l’efficacité énergétique et la circularité des processus de production. Ces projets peuvent prendre l'une de ces formes : Une première unité de production industrielle, après une phase de prototypage,Une ligne pilote de production industrielle, en interne ou chez un tiers situé en Île-de-France, à condition que sa production soit destinée à de premières commercialisations,La transformation d’une unité de production pilote à une unité de production industrielle | [Région Île-de-France](organisation) --- soutient ---> [industrie décarbonée](concept) | 1 | | Type de project: L’excès de précipitations tout au long de l’année a conduit à une chute spectaculaire des rendements des céréales d’été et des protéagineux (blé, orge, pois, féverole, etc.) que produisent 90% des agriculteurs d’Île-de-France, historique grenier à blé du pays. Tributaires naturels du fleurissement des cultures, les apiculteurs professionnels de la région ont également souffert de ces dérèglements climatiques.La Région accompagne les exploitations concernées en leur apportant une aide exceptionnelle. | [excès de précipitations](phénomène) --- DIMINUE ---> [rendements des protéagineux](concept) | 1 | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Evaluation Dataset #### grag-go-idf-only-pos * Dataset: [grag-go-idf-only-pos](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos) at [9743952](https://huggingface.co/datasets/Lettria/GRAG-GO-IDF-Only-Pos/tree/9743952a5d02847c83f30a59009b3231c56871a3) * Size: 1,325 evaluation samples * Columns: sentence1, sentence2, and label * Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | label | |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------| | type | string | string | int | | details | | | | * Samples: | sentence1 | sentence2 | label | |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------|:---------------| | Date de début: non précisée
Date de fin (clôture): non précisée
Date de début de la future campagne: non précisée | [Date de fin](concept) --- EST ---> [non précisée](__inferred__) | 1 | | Précision sure les bénéficiaires: Communes,Établissements publics de coopération intercommunale (avec ou sans fiscalité propre),Établissements publics territoriaux franciliens,Départements,Aménageurs publics et privés (lorsque ces derniers interviennent à la demande ou pour le compte d'une collectivité précitée). | [Aménageurs privés](entité) --- INTERVIENT_POUR ---> [Départements](entité) | 1 | | Type de project: Le programme propose des rencontres le samedi après-midi dans une université ou une grande école réputée, entre les professionnels bénévoles et les lycéens et collégiens sous la forme d'atelier thématiques. Ces moments de rencontre touchent à une grande multitude de domaines d’activités. L'objectif est de donner l’opportunité aux jeunes les plus enclavés d’échanger avec des intervenants professionnels aux parcours atypiques et inspirants. Les intervenants suscitent les ambitions et élargissent les perspectives des élèves. | [rencontres](événement) --- impliquent ---> [professionnels bénévoles](groupe) | 1 | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: epoch - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `use_cpu`: True - `dataloader_pin_memory`: False #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: epoch - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `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`: 1.0 - `num_train_epochs`: 1 - `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`: True - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `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`: False - `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`: None - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `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 - `use_liger_kernel`: False - `eval_use_gather_object`: False - `average_tokens_across_devices`: False - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | Validation Loss | EmbeddingSimEval_spearman_cosine | BinaryClassifEval_cosine_ap | |:------:|:----:|:-------------:|:---------------:|:--------------------------------:|:---------------------------:| | 0.6667 | 2 | 0.6092 | - | - | - | | 1.0 | 3 | - | 0.2053 | nan | 1.0 | ### Framework Versions - Python: 3.11.11 - Sentence Transformers: 3.4.1 - Transformers: 4.48.3 - PyTorch: 2.6.0+cpu - Accelerate: 1.4.0 - Datasets: 3.3.1 - Tokenizers: 0.21.0 ## 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", } ``` #### MultipleNegativesRankingLoss ```bibtex @misc{henderson2017efficient, title={Efficient Natural Language Response Suggestion for Smart Reply}, author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil}, year={2017}, eprint={1705.00652}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```