This is a Cross Encoder model finetuned from nreimers/MiniLM-L3-H384-uncased 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: nreimers/MiniLM-L3-H384-uncased
Maximum Sequence Length: 64 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("lucasflins/CE-MiniLM-L3-H384-uncased-BCE-20260207-042437")
# Get scores for pairs of texts
pairs = [
    ['aparador off white', 'aparador para sala ambiente classic off white nature - imcal'],
    ['itatiaia renova', 'balcão itatiaia branco renova com 3 portas e 2 gavetas'],
    ['cozinhas moduladas completas 100 mdf', 'cozinha completa 7 peças 100% mdf com portas de vidro americana'],
    ['caixa de som', 'caixa de som torre double 12 2300w bluetooth pulse - ps736'],
    ['escrivaninha 90cm', 'escrivaninha mesa escritório dobrável 90cm industrial steel quadra'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'aparador off white',
    [
        'aparador para sala ambiente classic off white nature - imcal',
        'balcão itatiaia branco renova com 3 portas e 2 gavetas',
        'cozinha completa 7 peças 100% mdf com portas de vidro americana',
        'caixa de som torre double 12 2300w bluetooth pulse - ps736',
        'escrivaninha mesa escritório dobrável 90cm industrial steel quadra',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Training Details

Training Dataset

Unnamed Dataset

Size: 3,891,914 training samples
Columns: query, text, and label
Approximate statistics based on the first 1000 samples:
query text label
type string string float
details
min: 2 characters
mean: 22.67 characters
max: 80 characters
min: 20 characters
mean: 67.39 characters
max: 188 characters
min: 0.0
mean: 0.47
max: 1.0

Samples:

query	text	label
`tenis under armour masculino`	`tênis under armour charged slight 3 se`	`0.9281755975185405`
`jogos de cama casal 100 algodao 4 pecas`	`lençol casal 400 fios 3 peças toque 100% macio com fronhas`	`0.5433727088037827`
`geladeira para caminhao`	`geladeira portátil 40l 12v/24v 110v/220v caminhão ônibus van`	`0.7394712159210983`

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: 1,323,175 evaluation samples
Columns: query, text, and label
Approximate statistics based on the first 1000 samples:
query text label
type string string float
details
min: 3 characters
mean: 22.38 characters
max: 76 characters
min: 15 characters
mean: 66.75 characters
max: 150 characters
min: 0.0
mean: 0.46
max: 1.0

Samples:

query	text	label
`aparador off white`	`aparador para sala ambiente classic off white nature - imcal`	`0.3710301443240389`
`itatiaia renova`	`balcão itatiaia branco renova com 3 portas e 2 gavetas`	`0.35256627704291904`
`cozinhas moduladas completas 100 mdf`	`cozinha completa 7 peças 100% mdf com portas de vidro americana`	`0.2038212525937755`

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: epoch
per_device_train_batch_size: 1024
per_device_eval_batch_size: 32
num_train_epochs: 10
warmup_ratio: 0.1
log_level: info
tf32: True
load_best_model_at_end: True
hub_strategy: end
hub_private_repo: True
hub_always_push: True
eval_on_start: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 1024
per_device_eval_batch_size: 32
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: 5e-05
weight_decay: 0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 10
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_ratio: 0.1
warmup_steps: 0
log_level: info
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: 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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
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: end
hub_private_repo: True
hub_always_push: True
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: True
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
0	0	-	0.6925
1.0	3801	0.6842	0.6915
2.0	7602	0.666	0.7007
3.0	11403	0.655	0.7055
4.0	15204	0.6486	0.7072

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.12.11
Sentence Transformers: 5.2.0
Transformers: 4.57.6
PyTorch: 2.7.1+cu128
Accelerate: 1.9.0
Datasets: 4.1.1
Tokenizers: 0.22.1

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",
}

	query	text	label
type	string	string	float
details	min: 2 characters mean: 22.67 characters max: 80 characters	min: 20 characters mean: 67.39 characters max: 188 characters	min: 0.0 mean: 0.47 max: 1.0