elsayovita/bge-small-en-v1.5-esg-v2

BAAI BGE small en v1.5 ESG

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. 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: BAAI/bge-small-en-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 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': 384, '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("elsayovita/bge-small-en-v1.5-esg-v2")
# Run inference
sentences = [
    'Employee health and well-being has never been more topical than it was in the past year. We understand that people around the world, including our employees, have been increasingly exposed to factors affecting their physical and mental wellbeing. We are committed to creating an environment that supports our employees and ensures they feel valued and have a sense of belonging. We utilised',
    "Question: What is the company's commitment towards its employees' health and well-being based on the provided context information?",
    'What types of skills does NetLink focus on developing through their training and development opportunities for employees?',
]
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]

Evaluation

Metrics

Information Retrieval

• Dataset: dim_384
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.787
cosine_accuracy@3	0.9269
cosine_accuracy@5	0.9446
cosine_accuracy@10	0.9597
cosine_precision@1	0.787
cosine_precision@3	0.309
cosine_precision@5	0.1889
cosine_precision@10	0.096
cosine_recall@1	0.0219
cosine_recall@3	0.0257
cosine_recall@5	0.0262
cosine_recall@10	0.0267
cosine_ndcg@10	0.1946
cosine_mrr@10	0.8588
cosine_map@100	0.0239

Information Retrieval

• Dataset: dim_256
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7815
cosine_accuracy@3	0.9236
cosine_accuracy@5	0.9422
cosine_accuracy@10	0.9596
cosine_precision@1	0.7815
cosine_precision@3	0.3079
cosine_precision@5	0.1884
cosine_precision@10	0.096
cosine_recall@1	0.0217
cosine_recall@3	0.0257
cosine_recall@5	0.0262
cosine_recall@10	0.0267
cosine_ndcg@10	0.194
cosine_mrr@10	0.8551
cosine_map@100	0.0238

Information Retrieval

• Dataset: dim_128
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7713
cosine_accuracy@3	0.9142
cosine_accuracy@5	0.9347
cosine_accuracy@10	0.9532
cosine_precision@1	0.7713
cosine_precision@3	0.3047
cosine_precision@5	0.1869
cosine_precision@10	0.0953
cosine_recall@1	0.0214
cosine_recall@3	0.0254
cosine_recall@5	0.026
cosine_recall@10	0.0265
cosine_ndcg@10	0.192
cosine_mrr@10	0.8457
cosine_map@100	0.0235

Information Retrieval

• Dataset: dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7428
cosine_accuracy@3	0.8924
cosine_accuracy@5	0.9185
cosine_accuracy@10	0.9412
cosine_precision@1	0.7428
cosine_precision@3	0.2975
cosine_precision@5	0.1837
cosine_precision@10	0.0941
cosine_recall@1	0.0206
cosine_recall@3	0.0248
cosine_recall@5	0.0255
cosine_recall@10	0.0261
cosine_ndcg@10	0.1875
cosine_mrr@10	0.822
cosine_map@100	0.0229

Information Retrieval

• Dataset: dim_32
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6669
cosine_accuracy@3	0.8242
cosine_accuracy@5	0.864
cosine_accuracy@10	0.8988
cosine_precision@1	0.6669
cosine_precision@3	0.2747
cosine_precision@5	0.1728
cosine_precision@10	0.0899
cosine_recall@1	0.0185
cosine_recall@3	0.0229
cosine_recall@5	0.024
cosine_recall@10	0.025
cosine_ndcg@10	0.1737
cosine_mrr@10	0.7533
cosine_map@100	0.021

Training Details

Training Dataset

Unnamed Dataset

• Size: 11,863 training samples
• Columns: context and question
• Approximate statistics based on the first 1000 samples:

  	context	question
  type	string	string
  details	min: 13 tokens mean: 40.74 tokens max: 277 tokens	min: 11 tokens mean: 24.4 tokens max: 62 tokens

• Samples:

  context	question
  The engagement with key stakeholders involves various topics and methods throughout the year	Question: What does the engagement with key stakeholders involve throughout the year?
  For unitholders and analysts, the focus is on business and operations, the release of financial results, and the overall performance and announcements	Question: What is the focus for unitholders and analysts in terms of business and operations, financial results, performance, and announcements?
  These are communicated through press releases and other required disclosures via SGXNet and NetLink's website	What platform is used to communicate press releases and required disclosures for NetLink?

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          384,
          256,
          128,
          64,
          32
      ],
      "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
• eval_on_start: False
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_32_cosine_map@100	dim_384_cosine_map@100	dim_64_cosine_map@100
0.4313	10	4.3426	-	-	-	-	-
0.8625	20	2.7083	-	-	-	-	-
1.0350	24	-	0.0229	0.0233	0.0195	0.0234	0.0220
1.2264	30	2.6835	-	-	-	-	-
1.6577	40	2.1702	-	-	-	-	-
1.9164	46	-	0.0230	0.0234	0.0197	0.0235	0.0221
0.4313	10	2.2406	-	-	-	-	-
0.8625	20	1.8606	-	-	-	-	-
1.0350	24	-	0.0233	0.0236	0.0204	0.0237	0.0225
1.2264	30	2.0645	-	-	-	-	-
1.6577	40	1.6752	-	-	-	-	-
2.0458	49	-	0.0235	0.0237	0.0208	0.0238	0.0228
2.0216	50	1.7855	-	-	-	-	-
2.4528	60	1.7333	-	-	-	-	-
2.8841	70	1.5116	-	-	-	-	-
3.0566	74	-	0.0235	0.0238	0.0210	0.0239	0.0229
3.2480	80	1.7812	-	-	-	-	-
3.6792	90	1.4886	-	-	-	-	-
3.7655	92	-	0.0235	0.0238	0.021	0.0239	0.0229

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.1
• Transformers: 4.42.4
• PyTorch: 2.4.0+cu121
• Accelerate: 0.32.1
• Datasets: 2.21.0
• 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}
}