SentenceTransformer based on BAAI/bge-base-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. 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: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
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': 768, '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("MugheesAwan11/bge-base-securiti-dataset-1-v4")
# Run inference
sentences = [
    "Thailand's PDPA applies to any legal entity collecting, using, or disclosing a natural (and alive) person's personal data.",
    "Who does the Thailand's PDPA apply to?",
    "What penalties could an organization face for infringing Kenya's Data Protection Act?",
]
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

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.5556
cosine_accuracy@3	0.8333
cosine_accuracy@5	0.9444
cosine_accuracy@10	1.0
cosine_precision@1	0.5556
cosine_precision@3	0.2778
cosine_precision@5	0.1889
cosine_precision@10	0.1
cosine_recall@1	0.5556
cosine_recall@3	0.8333
cosine_recall@5	0.9444
cosine_recall@10	1.0
cosine_ndcg@10	0.776
cosine_mrr@10	0.7043
cosine_map@100	0.7043

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.5
cosine_accuracy@3	0.7778
cosine_accuracy@5	0.8889
cosine_accuracy@10	1.0
cosine_precision@1	0.5
cosine_precision@3	0.2593
cosine_precision@5	0.1778
cosine_precision@10	0.1
cosine_recall@1	0.5
cosine_recall@3	0.7778
cosine_recall@5	0.8889
cosine_recall@10	1.0
cosine_ndcg@10	0.7499
cosine_mrr@10	0.6701
cosine_map@100	0.6701

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.5
cosine_accuracy@3	0.8889
cosine_accuracy@5	0.9444
cosine_accuracy@10	1.0
cosine_precision@1	0.5
cosine_precision@3	0.2963
cosine_precision@5	0.1889
cosine_precision@10	0.1
cosine_recall@1	0.5
cosine_recall@3	0.8889
cosine_recall@5	0.9444
cosine_recall@10	1.0
cosine_ndcg@10	0.769
cosine_mrr@10	0.6932
cosine_map@100	0.6932

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.5
cosine_accuracy@3	0.8333
cosine_accuracy@5	0.8889
cosine_accuracy@10	0.9444
cosine_precision@1	0.5
cosine_precision@3	0.2778
cosine_precision@5	0.1778
cosine_precision@10	0.0944
cosine_recall@1	0.5
cosine_recall@3	0.8333
cosine_recall@5	0.8889
cosine_recall@10	0.9444
cosine_ndcg@10	0.7364
cosine_mrr@10	0.6682
cosine_map@100	0.6701

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.4444
cosine_accuracy@3	0.6667
cosine_accuracy@5	0.7778
cosine_accuracy@10	1.0
cosine_precision@1	0.4444
cosine_precision@3	0.2222
cosine_precision@5	0.1556
cosine_precision@10	0.1
cosine_recall@1	0.4444
cosine_recall@3	0.6667
cosine_recall@5	0.7778
cosine_recall@10	1.0
cosine_ndcg@10	0.6963
cosine_mrr@10	0.6026
cosine_map@100	0.6026

Training Details

Training Dataset

Unnamed Dataset

Size: 161 training samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 5 tokens
mean: 40.09 tokens
max: 481 tokens

min: 7 tokens
mean: 13.01 tokens
max: 24 tokens

	positive	anchor
type	string	string
details	min: 5 tokens mean: 40.09 tokens max: 481 tokens	min: 7 tokens mean: 13.01 tokens max: 24 tokens

Samples:

positive	anchor
`The DPA may impose administrative fines of up to €10 million, or up to 2% of worldwide turnover. The DPA may also impose heavier fines up to €20 million, or up to 4% of worldwide turnover.`	`What is the penalty for non-compliance with the GDPR in Italy?`
`As per the DPA, the data handler must seek consent in writing from the data subject to collect any sensitive personal data.`	`What are the consent requirements under the DPA?`
`China's cybersecurity laws include the Cybersecurity Law, which governs various aspects of cybersecurity, data protection, and the obligations of organizations to ensure the security of networks and data within China's territory.`	`What are the cybersecurity laws in China?`

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: 2
learning_rate: 2e-05
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: 2
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: 3
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
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	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
1.0	3	0.6510	0.6686	0.6534	0.5641	0.6515
2.0	6	0.67	0.6821	0.6719	0.591	0.6951
3.0	9	0.6701	0.6932	0.6701	0.6026	0.7043

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.14
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.1.2+cu121
Accelerate: 0.31.0
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}
}

MugheesAwan11
/

bge-base-securiti-dataset-1-v4

SentenceTransformer based on BAAI/bge-base-en-v1.5

Model Details

Model Description

Model Sources

Full Model Architecture

Usage

Direct Usage (Sentence Transformers)

Evaluation

Metrics

Information Retrieval

Information Retrieval

Information Retrieval

Information Retrieval

Information Retrieval

Training Details

Training Dataset

Unnamed Dataset

Training Hyperparameters

Non-Default Hyperparameters

All Hyperparameters

Training Logs

Framework Versions

Citation

BibTeX

Sentence Transformers

MatryoshkaLoss

MultipleNegativesRankingLoss

Model tree for MugheesAwan11/bge-base-securiti-dataset-1-v4

Evaluation results