WaheedLone/bge-base-financial-matryoshka

BGE base Financial Matryoshka

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("WaheedLone/bge-base-financial-matryoshka")
# Run inference
sentences = [
    'The consolidated balance sheets of Visa Inc. as of September 30, 2023, list the total current assets at $33,532 million.',
    "What was the total of Visa Inc.'s current assets as of September 30, 2023?",
    "What was Garmin Ltd.'s net income for the fiscal year ended December 30, 2023?",
]
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	0.6886
cosine_accuracy@3	0.8286
cosine_accuracy@5	0.8671
cosine_accuracy@10	0.9129
cosine_precision@1	0.6886
cosine_precision@3	0.2762
cosine_precision@5	0.1734
cosine_precision@10	0.0913
cosine_recall@1	0.6886
cosine_recall@3	0.8286
cosine_recall@5	0.8671
cosine_recall@10	0.9129
cosine_ndcg@10	0.8023
cosine_mrr@10	0.7666
cosine_map@100	0.7697

Information Retrieval

• Dataset: dim_512
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6929
cosine_accuracy@3	0.8229
cosine_accuracy@5	0.8643
cosine_accuracy@10	0.91
cosine_precision@1	0.6929
cosine_precision@3	0.2743
cosine_precision@5	0.1729
cosine_precision@10	0.091
cosine_recall@1	0.6929
cosine_recall@3	0.8229
cosine_recall@5	0.8643
cosine_recall@10	0.91
cosine_ndcg@10	0.8017
cosine_mrr@10	0.7668
cosine_map@100	0.7701

Information Retrieval

• Dataset: dim_256
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6871
cosine_accuracy@3	0.8186
cosine_accuracy@5	0.8629
cosine_accuracy@10	0.9014
cosine_precision@1	0.6871
cosine_precision@3	0.2729
cosine_precision@5	0.1726
cosine_precision@10	0.0901
cosine_recall@1	0.6871
cosine_recall@3	0.8186
cosine_recall@5	0.8629
cosine_recall@10	0.9014
cosine_ndcg@10	0.7963
cosine_mrr@10	0.7623
cosine_map@100	0.7657

Information Retrieval

• Dataset: dim_128
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6743
cosine_accuracy@3	0.8057
cosine_accuracy@5	0.8529
cosine_accuracy@10	0.8943
cosine_precision@1	0.6743
cosine_precision@3	0.2686
cosine_precision@5	0.1706
cosine_precision@10	0.0894
cosine_recall@1	0.6743
cosine_recall@3	0.8057
cosine_recall@5	0.8529
cosine_recall@10	0.8943
cosine_ndcg@10	0.7862
cosine_mrr@10	0.7513
cosine_map@100	0.7549

Information Retrieval

• Dataset: dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6429
cosine_accuracy@3	0.7971
cosine_accuracy@5	0.8186
cosine_accuracy@10	0.8686
cosine_precision@1	0.6429
cosine_precision@3	0.2657
cosine_precision@5	0.1637
cosine_precision@10	0.0869
cosine_recall@1	0.6429
cosine_recall@3	0.7971
cosine_recall@5	0.8186
cosine_recall@10	0.8686
cosine_ndcg@10	0.7591
cosine_mrr@10	0.7237
cosine_map@100	0.7283

Training Details

Training Dataset

Unnamed Dataset

• Size: 6,300 training samples
• Columns: positive and anchor
• Approximate statistics based on the first 1000 samples:

  	positive	anchor
  type	string	string
  details	min: 6 tokens mean: 45.17 tokens max: 260 tokens	min: 7 tokens mean: 20.38 tokens max: 40 tokens

• Samples:

  positive	anchor
  Net revenue for fiscal year 2023 increased by $435 million compared to fiscal year 2022.	How did the net revenue for fiscal year 2023 compare to fiscal year 2022?
  Adjusted Free Cash Flow is defined as operating cash flow less capital spending and excluding payments for the transitional tax resulting from the U.S. Tax Act.	How is Adjusted Free Cash Flow defined in the text?
  During 2023, the Company’s net sales through its direct and indirect distribution channels accounted for 37% and 63%, respectively, of total net sales.	During 2023, what percentage of the Company’s net sales came from direct sales channels?

• 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
• 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: False
• 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	Training Loss	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
0.8122	10	1.6399	-	-	-	-	-
0.9746	12	-	0.7441	0.7580	0.7543	0.7068	0.7632
1.6244	20	0.6475	-	-	-	-	-
1.9492	24	-	0.7530	0.7653	0.7672	0.7244	0.7708
2.4365	30	0.4494	-	-	-	-	-
2.9239	36	-	0.7548	0.7653	0.7683	0.7297	0.7679
3.2487	40	0.4089	-	-	-	-	-
3.8985	48	-	0.7549	0.7657	0.7701	0.7283	0.7697

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• 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}
}