matunderstars/ufes-qa-embedding-finetuned-v2

SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L12-v2 on the train and test datasets. It maps sentences & paragraphs to a 384-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/all-MiniLM-L12-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Datasets:
  • train
  • test

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': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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("matunderstars/ufes-qa-embedding-finetuned-v2")
# Run inference
sentences = [
    'Como realizar o cadastro no Proaes?',
    'Acesse o site da Proaeci em https://proaeci.ufes.br/editais para verificar se há algum edital vigente para o semestre.',
    'Envie um ofício via documento avulso para a DRMN, conforme manual disponível em https://drm.saomateus.ufes.br/manuais-0.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

Training Details

Training Datasets

train

• Dataset: train at 02bfedf
• Size: 100 training samples
• Columns: question and answer
• Approximate statistics based on the first 100 samples:

  	question	answer
  type	string	string
  details	min: 7 tokens mean: 18.35 tokens max: 45 tokens	min: 14 tokens mean: 47.62 tokens max: 128 tokens

• Samples:

  question	answer
  Formatação de computador	A formatação de computadores deve ser solicitada diretamente ao suporte de TI, que avaliará a necessidade de backup e reinstalação dos sistemas operacionais.
  Como solicitar o recolhimento de um bem patrimonial?	Envie um ofício via documento avulso para a DRMN, conforme manual disponível em https://drm.saomateus.ufes.br/manuais-0.
  Como solicitar pagamento de ajuda de custos à estudante?	Cabe à Secretaria Única de Graduação – SUGRAD/CEUNES instruir devidamente o processo digital, encaminhar para análise e aprovação da Direção do Ceunes, que se estiver de acordo, remeterá o mesmo à DCFN (Divisão de Contabilidade e Finanças) para efetivação do pagamento. Informações sobre pagamento de ajuda de custos à estudantes entrar em contato com a DCFN (Divisão de Contabilidade e Finanças). E-mail institucional: [email protected]. Telefones: 3312-1517 e 3312-1518. Demais informações acesse o site: https://www.gestaoadministrativa.saomateus.ufes.br/apresentacao

• Loss: MultipleNegativesRankingLoss with these parameters:

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

test

• Dataset: test at 02bfedf
• Size: 100 training samples
• Columns: question and answer
• Approximate statistics based on the first 100 samples:

  	question	answer
  type	string	string
  details	min: 8 tokens mean: 16.95 tokens max: 33 tokens	min: 19 tokens mean: 58.8 tokens max: 128 tokens

• Samples:

  question	answer
  O que fazer em casos de urgência e emergência?	Siga as orientações disponíveis em : https://www.canva.com/design/DAGSiS46fII/0lsNqyuJYj4EYXkYnQNLyQ/edit?utm_content=DAGSiS46fII&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton
  Como receber os resultados dos exames periódicos?	Aguarde o e-mail de convocação para o recebimento dos resultados. Em caso de dúvidas, favor entrar em contato com o Setor de Enfermagem da DASAS pelo email [email protected] ou compareça presencialmente no Setor para maiores esclarecimentos.
  Como realizar o cadastro no Proaes?	Acesse o site da Proaeci em https://proaeci.ufes.br/editais para verificar se há algum edital vigente para o semestre.

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 300
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• 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
• 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: 300
• 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
• 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss
71.4286	500	0.1318
142.8571	1000	0.0001
214.2857	1500	0.0
285.7143	2000	0.0

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.3.1
• Transformers: 4.46.3
• PyTorch: 2.5.1+cu121
• Accelerate: 1.1.1
• Datasets: 3.2.0
• Tokenizers: 0.20.3

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{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}
}