WesleySAlves/e5-hazmat-classifier

SentenceTransformer based on intfloat/multilingual-e5-base

This is a sentence-transformers model finetuned from intfloat/multilingual-e5-base. 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
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• 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': 512, 'do_lower_case': False, 'architecture': '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:

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("WesleySAlves/e5-hazmat-classifier")
# Run inference
sentences = [
    'query: Tinta Esmalte Sintético Brilhante Glasu! 3,6l Cores Cor Verde Colonial ESMALTE SINTÉTICO STANDARD GLASU! (ANTERIORMENTE CHAMADO DE GLASURIT).\n\nIndicado para pintura de superfícies de madeira, metal, alumínio e galvanizados, para ambientes internos e externos. É um produto de fácil aplicação, secagem rápida, bom alastramento e boa aderência.\n\nCARACTERÍSTICAS\n- Secagem m',
    'query: Extintor de incêndio industrial móvel, Mxkfi-003, 68kg, Classe A, b, c',
    'query: Travesseiro Nativa Serena',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9624, 0.8195],
#         [0.9624, 1.0000, 0.7870],
#         [0.8195, 0.7870, 1.0000]])

Evaluation

Metrics

Binary Classification

• Dataset: hazmat-eval
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.8567
cosine_accuracy_threshold	0.8464
cosine_f1	0.8491
cosine_f1_threshold	0.8464
cosine_precision	0.8963
cosine_recall	0.8067
cosine_ap	0.9343
cosine_mcc	0.7169

Training Details

Training Dataset

Unnamed Dataset

• Size: 8,498 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 10 tokens mean: 50.14 tokens max: 140 tokens	min: 9 tokens mean: 52.57 tokens max: 161 tokens

• Samples:

  anchor	positive
  query: Gerador de cloro para piscina ATClor Gerador 3em1	query: Adesivo Chevrolet Adesivo
  query: Cinto Chacal Cinto de couro	query: Taca De Cristal Soda P/ Agua Elisabeth 350ml Azul 6 Peças
  query: Boneca de pelúcia Capivara com Chef Grande Boneca macia e macia marrom claro Boneca Capivara Soft Chef Capivara Material de enchimento: algodão PP Faixa etária aplicável: 0+ Cor: marrom Tamanho: 30CM Embalagem: 1 x pelúcia *Especificidades: * - Primeira qualidade: nossos bichos de pelúcia capycho são feitos de materiais cuidadosamente selecionados com excelente desempenho, preenchidos com algodã	query: Webcam Webcam Microfone USB PC Windows Mac Zoom Você não precisa mais se preocupar se o seu PC não tiver uma câmera. Este dispositivo Zoomy fornece a qualidade de imagem e os recursos de que você precisa para se comunicar de forma fácil e eficaz em realidade virtual.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 500 evaluation samples
• Columns: anchor and positive
• Approximate statistics based on the first 500 samples:

  	anchor	positive
  type	string	string
  details	min: 11 tokens mean: 48.39 tokens max: 131 tokens	min: 9 tokens mean: 52.09 tokens max: 134 tokens

• Samples:

  anchor	positive
  query: Gerador de cloro para piscina ATClor Gerador 3em1	query: Adesivo Chevrolet Adesivo
  query: Cinto Chacal Cinto de couro	query: Taca De Cristal Soda P/ Agua Elisabeth 350ml Azul 6 Peças
  query: Boneca de pelúcia Capivara com Chef Grande Boneca macia e macia marrom claro Boneca Capivara Soft Chef Capivara Material de enchimento: algodão PP Faixa etária aplicável: 0+ Cor: marrom Tamanho: 30CM Embalagem: 1 x pelúcia *Especificidades: * - Primeira qualidade: nossos bichos de pelúcia capycho são feitos de materiais cuidadosamente selecionados com excelente desempenho, preenchidos com algodã	query: Webcam Webcam Microfone USB PC Windows Mac Zoom Você não precisa mais se preocupar se o seu PC não tiver uma câmera. Este dispositivo Zoomy fornece a qualidade de imagem e os recursos de que você precisa para se comunicar de forma fácil e eficaz em realidade virtual.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 64
• learning_rate: 2e-05
• num_train_epochs: 5
• warmup_steps: 0.1
• fp16: True
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 8
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_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: 5
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_ratio: None
• warmup_steps: 0.1
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 42
• data_seed: None
• bf16: False
• fp16: True
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• 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
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• hub_revision: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• use_cache: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	hazmat-eval_cosine_ap
0.3759	50	4.1700	-	-
0.7519	100	4.0818	-	-
1.0	133	-	1.7933	0.9210
1.1278	150	3.8759	-	-
1.5038	200	3.8106	-	-
1.8797	250	3.7647	-	-
2.0	266	-	1.6138	0.9411
2.2556	300	3.6884	-	-
2.6316	350	3.6794	-	-
3.0	399	-	1.5537	0.9284
3.0075	400	3.6409	-	-
3.3835	450	3.6014	-	-
3.7594	500	3.5970	-	-
4.0	532	-	1.5148	0.9298
4.1353	550	3.5763	-	-
4.5113	600	3.5623	-	-
4.8872	650	3.5422	-	-
5.0	665	-	1.4926	0.9343

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.12.13
• Sentence Transformers: 5.3.0
• Transformers: 5.0.0
• PyTorch: 2.10.0+cu128
• Accelerate: 1.13.0
• Datasets: 4.0.0
• Tokenizers: 0.22.2

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",
}

MultipleNegativesRankingLoss

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}