metadata

language:
  - en
license: apache-2.0
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:30
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-large-en-v1.5
widget:
  - source_sentence: >-
      O NAT foi criado em 13 de setembro de 2012 pelo Ato n.º 25 da
      Procuradoria-Geral de Justiça do MPAC.
    sentences:
      - Quando o NAT foi criado?
      - O que significa NAT?
      - Quem instituiu o NAT?
  - source_sentence: >-
      A Lei Complementar n.º 291 de 2014 regulamentou o NAT como um órgão
      auxiliar do MPAC, fortalecendo seu papel de apoio técnico e científico.
    sentences:
      - O NAT é parte de qual organização?
      - Qual é a função do NAT no MPAC?
      - Como o NAT foi regulamentado pela Lei Complementar 291?
  - source_sentence: >-
      O NAT é o Núcleo de Apoio Técnico do MPAC, criado para prestar apoio em
      inteligência, segurança e operações técnico-científicas aos órgãos de
      execução, especialmente ao GAECO.
    sentences:
      - Quem são os coordenadores do NAT?
      - Qual é a função do NAT no LAB-LD?
      - Me explique o que é o NAT no Ministério Público.
  - source_sentence: >-
      O NAT é responsável por fornecer inteligência, suporte técnico-científico
      e segurança ao MPAC, além de gerenciar o SIMBA e o LAB-LD.
    sentences:
      - O que é o SIMBA, gerenciado pelo NAT?
      - Quais são as responsabilidades do NAT?
      - Para que foi criado o NAT?
  - source_sentence: >-
      NAT é o Núcleo de Apoio Técnico do Ministério Público do Estado do Acre,
      criado para fornecer suporte especializado em inteligência, segurança
      institucional e operações técnico-científicas.
    sentences:
      - Explique o que é o NAT no MPAC.
      - O NAT trabalha com o GAECO?
      - O que significa NAT no Ministério Público?
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
model-index:
  - name: MPAC BGE Large
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7544456402014998
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.675
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.675
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 1
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3333333333333333
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 1
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7827324383928644
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7083333333333333
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.7083333333333333
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7544456402014998
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.675
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.675
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7326691395183482
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.6458333333333333
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.6458333333333333
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.16666666666666666
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.5
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7043823413269836
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.6125
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.6125
            name: Cosine Map@100

MPAC BGE Large

This is a sentence-transformers model finetuned from BAAI/bge-large-en-v1.5 on the json dataset. 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-large-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 1024 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- json
Language: en
License: apache-2.0

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': True}) with Transformer model: BertModel 
  (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("mpac/mpac-bge-large")
# Run inference
sentences = [
    'NAT é o Núcleo de Apoio Técnico do Ministério Público do Estado do Acre, criado para fornecer suporte especializado em inteligência, segurança institucional e operações técnico-científicas.',
    'O que significa NAT no Ministério Público?',
    'O NAT trabalha com o GAECO?',
]
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

Datasets: dim_768, dim_512, dim_256, dim_128 and dim_64
Evaluated with InformationRetrievalEvaluator

Metric	dim_768	dim_512	dim_256	dim_128	dim_64
cosine_accuracy@1	0.5	0.5	0.5	0.5	0.5
cosine_accuracy@3	0.75	1.0	0.75	0.75	0.5
cosine_accuracy@5	1.0	1.0	1.0	1.0	1.0
cosine_accuracy@10	1.0	1.0	1.0	1.0	1.0
cosine_precision@1	0.5	0.5	0.5	0.5	0.5
cosine_precision@3	0.25	0.3333	0.25	0.25	0.1667
cosine_precision@5	0.2	0.2	0.2	0.2	0.2
cosine_precision@10	0.1	0.1	0.1	0.1	0.1
cosine_recall@1	0.5	0.5	0.5	0.5	0.5
cosine_recall@3	0.75	1.0	0.75	0.75	0.5
cosine_recall@5	1.0	1.0	1.0	1.0	1.0
cosine_recall@10	1.0	1.0	1.0	1.0	1.0
cosine_ndcg@10	0.7544	0.7827	0.7544	0.7327	0.7044
cosine_mrr@10	0.675	0.7083	0.675	0.6458	0.6125
cosine_map@100	0.675	0.7083	0.675	0.6458	0.6125

Training Details

Training Dataset

json

Dataset: json
Size: 30 training samples
Columns: positive and anchor
Approximate statistics based on the first 30 samples:
positive anchor
type string string
details
min: 38 tokens
mean: 54.3 tokens
max: 75 tokens

min: 8 tokens
mean: 13.5 tokens
max: 18 tokens

	positive	anchor
type	string	string
details	min: 38 tokens mean: 54.3 tokens max: 75 tokens	min: 8 tokens mean: 13.5 tokens max: 18 tokens

Samples:

positive	anchor
`O NAT foi criado em 13 de setembro de 2012 pelo Ato n.º 25 da Procuradoria-Geral de Justiça do MPAC.`	`Quando o NAT foi criado?`
`O NAT é vinculado à Procuradoria-Geral de Justiça e presta apoio técnico especializado ao MPAC.`	`O NAT é vinculado a qual órgão?`
`Os coordenadores do Núcleo de Apoio Técnico (NAT) são Marcela Cristina Ozório, como Coordenadora Geral e Bernardo Fiterman Albano, como Coordenador Adjunto`	`Quem são os coordenadores do NAT?`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 4
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: 32
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: 4
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
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
1.0	1	0.75	0.75	0.75	0.75	0.75
2.0	2	0.7468	0.7468	0.6967	0.7654	0.6973
3.0	3	0.7544	0.7654	0.7544	0.7327	0.6967
4.0	4	0.7544	0.7827	0.7544	0.7327	0.7044

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.12.7
Sentence Transformers: 3.3.1
Transformers: 4.41.2
PyTorch: 2.5.1+cu124
Accelerate: 1.1.0
Datasets: 2.19.1
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}
}