rnbokade/custom-bge

SentenceTransformer based on BAAI/bge-large-en

This is a sentence-transformers model finetuned from BAAI/bge-large-en. 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
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 tokens
• Similarity Function: Cosine Similarity

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("rnbokade/custom-bge")
# Run inference
sentences = [
    '3783 UC Davis (Northern Cal - Jon Sanguinetti)->Seq 18-P-3783',
    'dwg-3783s18 : 3783 Seq 18 - Drawings',
    'mat-3783s5 : 3783 Seq 5 - Material Order',
]
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

Triplet

• Dataset: custom-bge-dev
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9838
dot_accuracy	0.0162
manhattan_accuracy	0.9838
euclidean_accuracy	0.9838
max_accuracy	0.9838

Triplet

• Dataset: custom-bge-test
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9838
dot_accuracy	0.0162
manhattan_accuracy	0.9838
euclidean_accuracy	0.9838
max_accuracy	0.9838

Training Details

Training Dataset

Unnamed Dataset

• Size: 22,604 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 22 tokens mean: 25.35 tokens max: 27 tokens	min: 15 tokens mean: 18.84 tokens max: 24 tokens	min: 6 tokens mean: 16.74 tokens max: 38 tokens

• Samples:

  anchor	positive	negative
  MOD 1- Metal Decking - Floor Stud Wall Panels Floor Sheathing (Megaboard) Layout of Dirtt Frame Centerlines	EW1001-125 : Door Slabs / Frames / Hardware	dwg-3783s16 : 3783 Seq 16 - Drawings
  MOD 1- Metal Decking - Floor Stud Wall Panels Floor Sheathing (Megaboard) Layout of Dirtt Frame Centerlines	EW1001-125 : Door Slabs / Frames / Hardware	mat-3783s16 : 3783 Seq 16 - Material Order
  MOD 1- Metal Decking - Floor Stud Wall Panels Floor Sheathing (Megaboard) Layout of Dirtt Frame Centerlines	EW1001-125 : Door Slabs / Frames / Hardware	dwg-3786s292 : 3786 Seq 292 - Drawings

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 618 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 22 tokens mean: 33.18 tokens max: 45 tokens	min: 13 tokens mean: 17.48 tokens max: 22 tokens	min: 13 tokens mean: 17.48 tokens max: 22 tokens

• Samples:

  anchor	positive	negative
  23-0125 - Crispr mRNA Fume Hood Installations->Construction->QC Lab 1218 Fume Hood Install->Electrical - Fume Hood Power/Grounding Terminations - QC Lab	26-0500-1.0 : Breakers (3P 20A)	dwg-3786s17 : 3786 Seq 17 - Drawings
  23-0125 - Crispr mRNA Fume Hood Installations->Construction->QC Lab 1218 Fume Hood Install->Electrical - Fume Hood Power/Grounding Terminations - QC Lab	26-0500-1.0 : Breakers (3P 20A)	mat-3786s17 : 3786 Seq 17 - Material Order
  23-0125 - Crispr mRNA Fume Hood Installations->Construction->QC Lab 1218 Fume Hood Install->Electrical - Fume Hood Power/Grounding Terminations - QC Lab	26-0500-1.0 : Breakers (3P 20A)	09-9000-2.0 : Paint and Coatings

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• learning_rate: 5e-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: 1
• max_steps: -1
• lr_scheduler_type: linear
• 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: True
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• 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: False
• 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
• 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	loss	custom-bge-dev_max_accuracy	custom-bge-test_max_accuracy
0	0	-	-	0.8463	-
0.0708	100	0.5651	0.6065	0.9919	-
0.1415	200	0.168	0.4217	0.9935	-
0.2123	300	0.0499	0.6747	0.9951	-
0.2831	400	0.2205	0.8112	0.9951	-
0.3539	500	0.1167	0.7040	0.9903	-
0.4246	600	0.0968	0.7364	0.9822	-
0.4954	700	0.1704	0.5540	0.9968	-
0.5662	800	0.1104	0.7266	0.9951	-
0.6369	900	0.1698	1.1020	0.9725	-
0.7077	1000	0.1077	0.9028	0.9790	-
0.7785	1100	0.1667	0.8478	0.9757	-
0.8493	1200	0.0707	0.7629	0.9887	-
0.9200	1300	0.0299	0.8024	0.9871	-
0.9908	1400	0.0005	0.8161	0.9838	-
1.0	1413	-	-	-	0.9838

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.1
• Transformers: 4.42.4
• PyTorch: 2.3.1+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",
}

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}
}