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.

python

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("pujithapsx/finetune_bge_reranker_v2_m3_4114_3103")
# Get scores for pairs of texts
pairs = [
    ['flat 9 silver arc towers lucknow', 'flat no 9 silver arc tower lucknow'],
    ['house no 3 village mandapeta east godavari', 'h no 3 mandapeta village east godavari'],
    ['flat 6 next to hospital indore', 'flat no 6 nxt to hospital indore'],
    ['house no 36 village muzaffarpur bihar', 'h no 36 muzaffarpur village bihar'],
    ['flat 10 shanti colony cross 2 kharagpur', 'flat 10 shanti clny cr 2 kharagpur'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'flat 9 silver arc towers lucknow',
    [
        'flat no 9 silver arc tower lucknow',
        'h no 3 mandapeta village east godavari',
        'flat no 6 nxt to hospital indore',
        'h no 36 muzaffarpur village bihar',
        'flat 10 shanti clny cr 2 kharagpur',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Classification

Dataset: entity-matching-eval
Evaluated with CrossEncoderClassificationEvaluator

Metric	Value
accuracy	0.9951
accuracy_threshold	0.9945
f1	0.9958
f1_threshold	0.9945
precision	0.9958
recall	0.9958
average_precision	0.9999

Training Details

Training Dataset

Unnamed Dataset

Size: 2,878 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 label
type string string int
details
min: 4 characters
mean: 30.34 characters
max: 86 characters
min: 4 characters
mean: 31.7 characters
max: 105 characters
0: ~42.10%
1: ~57.90%

Samples:

sentence1	sentence2	label
`pamulaparti venkata subba rao`	`v s rao p`	`1`
`c/o danish khan 18-7-335/190 aman nag kulsum msjd yktpura nr somnath tmpl`	`c/o danish khan 18-7-335/190 aman nagar kulsum masjid yakutpura near somnath temple`	`1`
`balakrishna`	`bala krishna`	`1`

Loss: BinaryCrossEntropyLoss with these parameters:
json
1{ 2 "activation_fn": "torch.nn.modules.linear.Identity", 3 "pos_weight": null 4}

Evaluation Dataset

Unnamed Dataset

Size: 412 evaluation samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 412 samples:
sentence1 sentence2 label
type string string int
details
min: 4 characters
mean: 30.2 characters
max: 97 characters
min: 3 characters
mean: 31.23 characters
max: 101 characters
0: ~42.48%
1: ~57.52%

Samples:

sentence1	sentence2	label
`flat 9 silver arc towers lucknow`	`flat no 9 silver arc tower lucknow`	`1`
`house no 3 village mandapeta east godavari`	`h no 3 mandapeta village east godavari`	`1`
`flat 6 next to hospital indore`	`flat no 6 nxt to hospital indore`	`1`

Loss: BinaryCrossEntropyLoss with these parameters:
json
1{ 2 "activation_fn": "torch.nn.modules.linear.Identity", 3 "pos_weight": null 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_eval_batch_size: 16
learning_rate: 2e-05
weight_decay: 0.01
warmup_steps: 107
remove_unused_columns: False
load_best_model_at_end: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 8
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
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: 0.0
warmup_steps: 107
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
bf16: False
fp16: False
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: False
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}
parallelism_config: 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
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
use_legacy_prediction_loop: False
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_inputs_for_metrics: False
include_for_metrics: []
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
include_tokens_per_second: False
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
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
0.1472	53	0.4704	-	-
0.2944	106	0.2525	-	-
0.2972	107	-	0.1369	0.9971
0.4417	159	0.2747	-	-
0.5889	212	0.2187	-	-
0.5944	214	-	0.1651	0.9957
0.7361	265	0.1826	-	-
0.8833	318	0.1177	-	-
0.8917	321	-	0.0572	0.9995
1.0306	371	0.0885	-	-
1.1778	424	0.128	-	-
1.1889	428	-	0.0160	0.9999
1.325	477	0.0376	-	-
1.4722	530	0.0528	-	-
1.4861	535	-	0.0526	0.9998
1.6194	583	0.0202	-	-
1.7667	636	0.0664	-	-
1.7833	642	-	0.0276	0.9999
1.9139	689	0.0186	-	-
2.0611	742	0.0363	-	-
2.0806	749	-	0.0299	0.9999
2.2083	795	0.0002	-	-
2.3556	848	0.0405	-	-
2.3778	856	-	0.0508	0.9999
2.5028	901	0.0246	-	-
2.65	954	0.0177	-	-
2.675	963	-	0.0424	0.9999
2.7972	1007	0.0064	-	-
2.9444	1060	0.0246	-	-
2.9722	1070	-	0.0501	0.9999

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 5.3.0
Transformers: 4.57.6
PyTorch: 2.10.0+cu128
Accelerate: 1.13.0
Datasets: 4.8.4
Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

bibtex

@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",
}

	sentence1	sentence2	label
type	string	string	int
details	min: 4 characters mean: 30.34 characters max: 86 characters	min: 4 characters mean: 31.7 characters max: 105 characters	0: ~42.10% 1: ~57.90%