elsayovita/bge-micro-v2-esg-v2

BGE micro v2 ESG

This is a sentence-transformers model finetuned from TaylorAI/bge-micro-v2. 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: TaylorAI/bge-micro-v2
• 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': 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})
)

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-micro-v2-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	0.755
cosine_accuracy@3	0.8992
cosine_accuracy@5	0.9237
cosine_accuracy@10	0.9447
cosine_precision@1	0.755
cosine_precision@3	0.2997
cosine_precision@5	0.1847
cosine_precision@10	0.0945
cosine_recall@1	0.021
cosine_recall@3	0.025
cosine_recall@5	0.0257
cosine_recall@10	0.0262
cosine_ndcg@10	0.1891
cosine_mrr@10	0.8309
cosine_map@100	0.0231

Information Retrieval

• Dataset: dim_256
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7496
cosine_accuracy@3	0.8958
cosine_accuracy@5	0.9187
cosine_accuracy@10	0.9418
cosine_precision@1	0.7496
cosine_precision@3	0.2986
cosine_precision@5	0.1837
cosine_precision@10	0.0942
cosine_recall@1	0.0208
cosine_recall@3	0.0249
cosine_recall@5	0.0255
cosine_recall@10	0.0262
cosine_ndcg@10	0.1882
cosine_mrr@10	0.8262
cosine_map@100	0.023

Information Retrieval

• Dataset: dim_128
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7356
cosine_accuracy@3	0.8875
cosine_accuracy@5	0.9106
cosine_accuracy@10	0.9342
cosine_precision@1	0.7356
cosine_precision@3	0.2958
cosine_precision@5	0.1821
cosine_precision@10	0.0934
cosine_recall@1	0.0204
cosine_recall@3	0.0247
cosine_recall@5	0.0253
cosine_recall@10	0.0259
cosine_ndcg@10	0.1858
cosine_mrr@10	0.8144
cosine_map@100	0.0227

Information Retrieval

• Dataset: dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6972
cosine_accuracy@3	0.8494
cosine_accuracy@5	0.8831
cosine_accuracy@10	0.9132
cosine_precision@1	0.6972
cosine_precision@3	0.2831
cosine_precision@5	0.1766
cosine_precision@10	0.0913
cosine_recall@1	0.0194
cosine_recall@3	0.0236
cosine_recall@5	0.0245
cosine_recall@10	0.0254
cosine_ndcg@10	0.1788
cosine_mrr@10	0.7793
cosine_map@100	0.0217

Information Retrieval

• Dataset: dim_32
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.5974
cosine_accuracy@3	0.7523
cosine_accuracy@5	0.797
cosine_accuracy@10	0.8448
cosine_precision@1	0.5974
cosine_precision@3	0.2508
cosine_precision@5	0.1594
cosine_precision@10	0.0845
cosine_recall@1	0.0166
cosine_recall@3	0.0209
cosine_recall@5	0.0221
cosine_recall@10	0.0235
cosine_ndcg@10	0.1593
cosine_mrr@10	0.685
cosine_map@100	0.0191

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: False
• 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: False
• 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	5.2501	-	-	-	-	-
0.8625	20	3.4967	-	-	-	-	-
1.0350	24	-	0.0221	0.0224	0.0185	0.0226	0.0210
1.2264	30	3.1196	-	-	-	-	-
1.6577	40	2.4428	-	-	-	-	-
2.0458	49	-	0.0226	0.0229	0.0189	0.0230	0.0215
2.0216	50	2.2222	-	-	-	-	-
2.4528	60	2.3441	-	-	-	-	-
2.8841	70	2.0096	-	-	-	-	-
3.0566	74	-	0.0227	0.0230	0.0191	0.0231	0.0217
3.2480	80	2.3019	-	-	-	-	-
3.6792	90	1.9538	-	-	-	-	-
3.7655	92	-	0.0227	0.023	0.0191	0.0231	0.0217

• 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}
}