This is a Cross Encoder model finetuned from cross-encoder/stsb-roberta-large 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: cross-encoder/stsb-roberta-large
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/address-crossencoder-stsb-roberta-large-finetuned")
# Get scores for pairs of texts
pairs = [
    ['C/O Rakesh Tower C Sector 137 Gurgaon', 'C Tower Sec-137 Gurugram'],
    ['Tellapur Hyderabad', 'Telapur Hyderabad'],
    ['Flat 703 Electronic City Bangalore', 'Flat 703 Electronic City Mumbai'],
    ['B-12 Malviya Nagar Delhi', 'B-22 Malviya Nagar Delhi'],
    ['Flat 1203 Lower Parel Mumbai', 'Flat 1203 Lower Parel Chennai'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'C/O Rakesh Tower C Sector 137 Gurgaon',
    [
        'C Tower Sec-137 Gurugram',
        'Telapur Hyderabad',
        'Flat 703 Electronic City Mumbai',
        'B-22 Malviya Nagar Delhi',
        'Flat 1203 Lower Parel Chennai',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Classification

Dataset: validation
Evaluated with CrossEncoderClassificationEvaluator

Metric	Value
accuracy	0.95
accuracy_threshold	0.4996
f1	0.9517
f1_threshold	0.3665
precision	0.9452
recall	0.9583
average_precision	0.9753

Training Details

Training Dataset

Unnamed Dataset

Size: 1,122 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 label
type string string float
details
min: 8 characters
mean: 27.83 characters
max: 55 characters
min: 8 characters
mean: 27.83 characters
max: 61 characters
min: 0.0
mean: 0.55
max: 1.0

Samples:

sentence1	sentence2	label
`Eighty Eight 8th Cross HSR Layout Bengaluru`	`52 Fifty Two D Second Lane Marathahalli Bengaluru`	`0.0`
`Flat 301 C/O Sharma Kondapur Near Hitech City Hyderabad`	`Flat 301 C/O Sharma Kondapoor Near Hi Tech City Hyd`	`1.0`
`Anna Nagar 12B Chennai 600040`	`12B Anna Nagar Chennai`	`1.0`

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: 140 evaluation samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 140 samples:
sentence1 sentence2 label
type string string float
details
min: 13 characters
mean: 29.04 characters
max: 74 characters
min: 12 characters
mean: 29.13 characters
max: 62 characters
min: 0.0
mean: 0.51
max: 1.0
Samples:
sentence1 sentence2 label
C/O Rakesh Tower C Sector 137 Gurgaon C Tower Sec-137 Gurugram 1.0
Tellapur Hyderabad Telapur Hyderabad 1.0
Flat 703 Electronic City Bangalore Flat 703 Electronic City Mumbai 0.0
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

num_train_epochs: 6
learning_rate: 1.5e-05
warmup_steps: 0.1
weight_decay: 0.01
gradient_accumulation_steps: 4
disable_tqdm: True
eval_strategy: epoch
per_device_eval_batch_size: 16
load_best_model_at_end: True

All Hyperparameters

Click to expand

per_device_train_batch_size: 8
num_train_epochs: 6
max_steps: -1
learning_rate: 1.5e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0.1
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.01
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 4
average_tokens_across_devices: True
max_grad_norm: 1.0
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: False
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: True
project: huggingface
trackio_space_id: trackio
eval_strategy: epoch
per_device_eval_batch_size: 16
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: True
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: 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
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_backend: None
ddp_timeout: 1800
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	validation_average_precision
0.2837	10	0.4552	-	-
0.5674	20	0.4294	-	-
0.8511	30	0.4078	-	-
1.0	36	-	0.2787	0.9570
1.1135	40	0.3982	-	-
1.3972	50	0.3678	-	-
1.6809	60	0.3367	-	-
1.9645	70	0.4198	-	-
2.0	72	-	0.2252	0.9702
2.2270	80	0.3148	-	-
2.5106	90	0.3862	-	-
2.7943	100	0.3374	-	-
3.0	108	-	0.1974	0.9725
3.0567	110	0.3272	-	-
3.3404	120	0.2932	-	-
3.6241	130	0.3010	-	-
3.9078	140	0.3119	-	-
4.0	144	-	0.1829	0.9736
4.1702	150	0.3005	-	-
4.4539	160	0.3292	-	-
4.7376	170	0.2207	-	-
5.0	180	0.2954	0.1745	0.9750
5.2837	190	0.2853	-	-
5.5674	200	0.2969	-	-
5.8511	210	0.2600	-	-
6.0	216	-	0.1719	0.9753

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.11
Sentence Transformers: 5.3.0
Transformers: 5.3.0
PyTorch: 2.11.0+cpu
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	float
details	min: 8 characters mean: 27.83 characters max: 55 characters	min: 8 characters mean: 27.83 characters max: 61 characters	min: 0.0 mean: 0.55 max: 1.0

sentence1	sentence2	label
`C/O Rakesh Tower C Sector 137 Gurgaon`	`C Tower Sec-137 Gurugram`	`1.0`
`Tellapur Hyderabad`	`Telapur Hyderabad`	`1.0`
`Flat 703 Electronic City Bangalore`	`Flat 703 Electronic City Mumbai`	`0.0`