kallilikhitha123/finetuned-bge-reranker-2403

CrossEncoder based on BAAI/bge-reranker-v2-m3

This is a Cross Encoder model finetuned from BAAI/bge-reranker-v2-m3 using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: BAAI/bge-reranker-v2-m3
• Maximum Sequence Length: 512 tokens
• Number of Output Labels: 1 label

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("kallilikhitha123/finetuned-bge-reranker-2403")
# Get scores for pairs of texts
pairs = [
    ['bhargavi madhavi', 'bhargavi srinidhi'],
    ['navya kiara', 'navya tanvi'],
    ['manish', 'manisha'],
    ['anantha padmanabha', 'ananthpadmanabha'],
    ['nitin singh', 'n singh'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'bhargavi madhavi',
    [
        'bhargavi srinidhi',
        'navya tanvi',
        'manisha',
        'ananthpadmanabha',
        'n singh',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Classification

• Dataset: entity-matching-eval
• Evaluated with CrossEncoderClassificationEvaluator

Metric	Value
accuracy	0.9716
accuracy_threshold	0.9984
f1	0.9762
f1_threshold	0.9965
precision	0.988
recall	0.9647
average_precision	0.9915

Training Details

Training Dataset

Unnamed Dataset

• Size: 1,967 training samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 3 characters mean: 20.08 characters max: 74 characters	min: 3 characters mean: 19.86 characters max: 62 characters	0: ~39.60% 1: ~60.40%

• Samples:

  sentence1	sentence2	label
  whitefield bangalore	whitefield chennai	0
  shastri nagar narangpura ahmedabad	l-5-135 block shastri nagar ahmedabad gujarat	0
  plot 45 kukatpally hyderabad	plot 45 kukatpally bangalore	0

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": null
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 282 evaluation samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 282 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 4 characters mean: 19.49 characters max: 51 characters	min: 3 characters mean: 19.16 characters max: 51 characters	0: ~39.72% 1: ~60.28%

• Samples:

  sentence1	sentence2	label
  bhargavi madhavi	bhargavi srinidhi	0
  navya kiara	navya tanvi	0
  manish	manisha	0

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": null
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• weight_decay: 0.01
• warmup_steps: 73
• remove_unused_columns: False
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 16
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.01
• 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: linear
• lr_scheduler_kwargs: None
• warmup_ratio: None
• warmup_steps: 73
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 42
• data_seed: None
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: False
• label_names: None
• load_best_model_at_end: True
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• 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
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• hub_revision: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• use_cache: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	entity-matching-eval_average_precision
0.1463	36	0.3950	-	-
0.2927	72	0.3475	-	-
0.2967	73	-	0.3461	0.9648
0.1463	36	0.0637	-	-
0.2927	72	0.1365	-	-
0.2967	73	-	0.4392	0.9630
0.4390	108	0.4193	-	-
0.5854	144	0.2634	-	-
0.5935	146	-	0.3738	0.9602
0.7317	180	0.3739	-	-
0.8780	216	0.1583	-	-
0.8902	219	-	0.1850	0.9893
1.0244	252	0.1806	-	-
1.1707	288	0.1287	-	-
1.1870	292	-	0.4037	0.9836
1.3171	324	0.1883	-	-
1.4634	360	0.1562	-	-
1.4837	365	-	0.3427	0.9868
1.6098	396	0.0619	-	-
1.7561	432	0.0854	-	-
1.7805	438	-	0.2923	0.9887
1.9024	468	0.1240	-	-
2.0488	504	0.1076	-	-
2.0772	511	-	0.2441	0.9915
2.1951	540	0.1289	-	-
2.3415	576	0.0311	-	-
2.3740	584	-	0.2151	0.9923
2.4878	612	0.0181	-	-
2.6341	648	0.0982	-	-
2.6707	657	-	0.2268	0.9915
2.7805	684	0.0010	-	-
2.9268	720	0.0321	-	-
2.9675	730	-	0.2393	0.9915

Framework Versions

• Python: 3.12.12
• Sentence Transformers: 5.2.3
• Transformers: 5.0.0
• PyTorch: 2.10.0+cpu
• Accelerate: 1.12.0
• Datasets: 4.8.3
• Tokenizers: 0.22.2

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",
}