Hugging Face's logo

dariolopez
/
bge-m3-es-legal-tmp-6

Sentence Similarity
sentence-transformers
Safetensors
Spanish
xlm-roberta
feature-extraction
Generated from Trainer
dataset_size:2947
loss:MatryoshkaLoss
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
Edit model card

BGE large Legal Spanish

This is a sentence-transformers model finetuned from BAAI/bge-m3. It maps sentences & paragraphs to a 1024-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: BAAI/bge-m3
  • Maximum Sequence Length: 8192 tokens
  • Output Dimensionality: 1024 tokens
  • Similarity Function: Cosine Similarity
  • Language: es
  • License: apache-2.0

Model Sources

Full Model Architecture 

SentenceTransformer(
  (0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 1024, '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): 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("dariolopez/bge-m3-es-legal-tmp-6")
# Run inference
sentences = [
    'Artículo 6. Definiciones. 1. Discriminación directa e indirecta. b) La discriminación indirecta se produce cuando una disposición, criterio o práctica aparentemente neutros ocasiona o puede ocasionar a una o varias personas una desventaja particular con respecto a otras por razón de las causas previstas en el apartado 1 del artículo 2.',
    '¿Qué se considera discriminación indirecta?',
    '¿Qué tipo de información se considera veraz?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

Metric Value
cosine_accuracy@1 0.5518
cosine_accuracy@3 0.8049
cosine_accuracy@5 0.8445
cosine_accuracy@10 0.9024
cosine_precision@1 0.5518
cosine_precision@3 0.2683
cosine_precision@5 0.1689
cosine_precision@10 0.0902
cosine_recall@1 0.5518
cosine_recall@3 0.8049
cosine_recall@5 0.8445
cosine_recall@10 0.9024
cosine_ndcg@10 0.738
cosine_mrr@10 0.6842
cosine_map@100 0.6881

Information Retrieval

Metric Value
cosine_accuracy@1 0.5488
cosine_accuracy@3 0.8049
cosine_accuracy@5 0.8506
cosine_accuracy@10 0.9024
cosine_precision@1 0.5488
cosine_precision@3 0.2683
cosine_precision@5 0.1701
cosine_precision@10 0.0902
cosine_recall@1 0.5488
cosine_recall@3 0.8049
cosine_recall@5 0.8506
cosine_recall@10 0.9024
cosine_ndcg@10 0.7361
cosine_mrr@10 0.6816
cosine_map@100 0.6855

Information Retrieval

Metric Value
cosine_accuracy@1 0.5579
cosine_accuracy@3 0.811
cosine_accuracy@5 0.8506
cosine_accuracy@10 0.8933
cosine_precision@1 0.5579
cosine_precision@3 0.2703
cosine_precision@5 0.1701
cosine_precision@10 0.0893
cosine_recall@1 0.5579
cosine_recall@3 0.811
cosine_recall@5 0.8506
cosine_recall@10 0.8933
cosine_ndcg@10 0.7363
cosine_mrr@10 0.6845
cosine_map@100 0.6889

Information Retrieval

Metric Value
cosine_accuracy@1 0.5549
cosine_accuracy@3 0.7957
cosine_accuracy@5 0.8323
cosine_accuracy@10 0.8841
cosine_precision@1 0.5549
cosine_precision@3 0.2652
cosine_precision@5 0.1665
cosine_precision@10 0.0884
cosine_recall@1 0.5549
cosine_recall@3 0.7957
cosine_recall@5 0.8323
cosine_recall@10 0.8841
cosine_ndcg@10 0.7307
cosine_mrr@10 0.6804
cosine_map@100 0.6851

Information Retrieval

Metric Value
cosine_accuracy@1 0.5213
cosine_accuracy@3 0.7622
cosine_accuracy@5 0.814
cosine_accuracy@10 0.8659
cosine_precision@1 0.5213
cosine_precision@3 0.2541
cosine_precision@5 0.1628
cosine_precision@10 0.0866
cosine_recall@1 0.5213
cosine_recall@3 0.7622
cosine_recall@5 0.814
cosine_recall@10 0.8659
cosine_ndcg@10 0.7028
cosine_mrr@10 0.6495
cosine_map@100 0.655

Information Retrieval

Metric Value
cosine_accuracy@1 0.4848
cosine_accuracy@3 0.7256
cosine_accuracy@5 0.7805
cosine_accuracy@10 0.8537
cosine_precision@1 0.4848
cosine_precision@3 0.2419
cosine_precision@5 0.1561
cosine_precision@10 0.0854
cosine_recall@1 0.4848
cosine_recall@3 0.7256
cosine_recall@5 0.7805
cosine_recall@10 0.8537
cosine_ndcg@10 0.6729
cosine_mrr@10 0.6147
cosine_map@100 0.6198

Training Details

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: epoch
  • per_device_train_batch_size: 16
  • per_device_eval_batch_size: 16
  • gradient_accumulation_steps: 16
  • learning_rate: 2e-05
  • num_train_epochs: 6
  • lr_scheduler_type: cosine
  • warmup_ratio: 0.1
  • bf16: True
  • tf32: True
  • load_best_model_at_end: True
  • optim: adamw_torch_fused
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: epoch
  • 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: 16
  • 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: 6
  • max_steps: -1
  • lr_scheduler_type: cosine
  • 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: True
  • fp16: False
  • fp16_opt_level: O1
  • half_precision_backend: auto
  • bf16_full_eval: False
  • fp16_full_eval: False
  • tf32: True
  • 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_fused
  • 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
  • batch_sampler: no_duplicates
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss loss dim_1024_cosine_map@100 dim_128_cosine_map@100 dim_256_cosine_map@100 dim_512_cosine_map@100 dim_64_cosine_map@100 dim_768_cosine_map@100
0.4324 5 1.6507 - - - - - - -
0.8649 10 0.9598 - - - - - - -
0.9514 11 - 0.5477 0.6833 0.6616 0.6836 0.6758 0.5994 0.6744
1.2973 15 0.8248 - - - - - - -
1.7297 20 0.3858 - - - - - - -
1.9892 23 - 0.4242 0.6748 0.6544 0.6833 0.6740 0.6233 0.6697
2.1622 25 0.32 - - - - - - -
2.5946 30 0.1703 - - - - - - -
2.9405 34 - 0.3940 0.6755 0.6523 0.6823 0.6797 0.6196 0.6776
3.0270 35 0.1337 - - - - - - -
3.4595 40 0.0949 - - - - - - -
3.8919 45 0.0594 - - - - - - -
3.9784 46 - 0.3735 0.6867 0.6588 0.6865 0.6854 0.6189 0.6826
4.3243 50 0.07 - - - - - - -
4.7568 55 0.0524 - - - - - - -
4.9297 57 - 0.3642 0.6870 0.6577 0.6858 0.6871 0.6228 0.6853
5.1892 60 0.0598 - - - - - - -
5.6216 65 0.0491 - - - - - - -
5.7081 66 - 0.3626 0.6881 0.6550 0.6851 0.6889 0.6198 0.6855
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.12
  • Sentence Transformers: 3.0.1
  • Transformers: 4.42.3
  • PyTorch: 2.2.0+cu121
  • Accelerate: 0.32.1
  • Datasets: 2.20.0
  • 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",
}

MatryoshkaLoss 

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss 

@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}
}
Downloads last month
18
Safetensors
Model size
568M params
Tensor type
F32
·
Inference Examples
Sentence Similarity
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for dariolopez/bge-m3-es-legal-tmp-6

Base model

BAAI/bge-m3
Finetuned
(125)
this model

Evaluation results

View on Papers With Code