This is a sentence-transformers model finetuned from intfloat/multilingual-e5-small. It maps sentences & paragraphs to a 384-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: intfloat/multilingual-e5-small
Maximum Sequence Length: 256 tokens
Output Dimensionality: 384 dimensions
Similarity Function: Cosine Similarity

Language: multilingual
License: mit

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': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, '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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("Antix5/product-embed-multi-e5-small")
# Run inference
sentences = [
    'Fun World Angelic Maiden Child Costume',
    "Rubie's Costume Co - Girls Gypsy Costume",
    'Melissa & Doug Personalized Pattern Blocks & Boards Classic Toy',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7135, 0.6875],
#         [0.7135, 1.0000, 0.6791],
#         [0.6875, 0.6791, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: ir_eval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.9102
cosine_accuracy@3	0.957
cosine_accuracy@5	0.9727
cosine_accuracy@10	1.0
cosine_precision@1	0.9102
cosine_precision@3	0.5104
cosine_precision@5	0.4008
cosine_precision@10	0.2969
cosine_recall@1	0.1348
cosine_recall@3	0.174
cosine_recall@5	0.1983
cosine_recall@10	0.2487
cosine_ndcg@10	0.465
cosine_mrr@10	0.9379
cosine_map@1	0.9102
cosine_map@3	0.5282
cosine_map@5	0.421
cosine_map@10	0.3311

Training Details

Training Dataset

Unnamed Dataset

Size: 74,864 training samples
Columns: text1, text2, and label
Approximate statistics based on the first 1000 samples:
text1 text2 label
type string string float
details
min: 4 tokens
mean: 19.67 tokens
max: 54 tokens
min: 3 tokens
mean: 15.59 tokens
max: 72 tokens
min: 0.0
mean: 0.53
max: 1.0

Samples:

text1	text2	label
`Premier 26764 Car Spinner, Santa, 25 by 19-1/2-Inch`	`Premier 26764 Tourbillon pour voiture, Santa, 25 x 19-1/2 pouces`	`1.0`
`Premier 26764 Car Spinner, Santa, 25 by 19-1/2-Inch`	`BNTS, ЧИПСЫ ИЗ ФАСОЛИ NV И МОРСКАЯ СОЛЬ`	`0.0`
`Premier 26764 Car Spinner, Santa, 25 by 19-1/2-Inch`	`Beanitos, Чипс из фасоли navy, Сыр на чо`	`0.0`

Loss: CoSENTLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "pairwise_cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 256
learning_rate: 2e-05
num_train_epochs: 2
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: 32
per_device_eval_batch_size: 256
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: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 2
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}
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
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: False
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: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	ir_eval_cosine_ndcg@10
0.0004	1	5.9178	-
0.0427	100	5.7854	-
0.0855	200	5.7118	-
0.1282	300	5.6765	-
0.1709	400	5.647	-
0.2137	500	5.6046	-
0.2564	600	5.5859	-
0.2991	700	5.5586	-
0.3419	800	5.5319	-
0.3846	900	5.564	-
0.4274	1000	5.577	0.4854
0.4701	1100	5.5229	-
0.5128	1200	5.5294	-
0.5556	1300	5.4836	-
0.5983	1400	5.4851	-
0.6410	1500	5.4646	-
0.6838	1600	5.4784	-
0.7265	1700	5.481	-
0.7692	1800	5.4923	-
0.8120	1900	5.4696	-
0.8547	2000	5.4932	0.4749
0.8974	2100	5.4752	-
0.9402	2200	5.459	-
0.9829	2300	5.4371	-
1.0256	2400	5.3701	-
1.0684	2500	5.3562	-
1.1111	2600	5.4101	-
1.1538	2700	5.3829	-
1.1966	2800	5.3687	-
1.2393	2900	5.36	-
1.2821	3000	5.3446	0.4725
1.3248	3100	5.3757	-
1.3675	3200	5.3821	-
1.4103	3300	5.3918	-
1.4530	3400	5.3083	-
1.4957	3500	5.3389	-
1.5385	3600	5.3037	-
1.5812	3700	5.3424	-
1.6239	3800	5.3383	-
1.6667	3900	5.3252	-
1.7094	4000	5.3358	0.4676
1.7521	4100	5.2704	-
1.7949	4200	5.3415	-
1.8376	4300	5.361	-
1.8803	4400	5.3654	-
1.9231	4500	5.3386	-
1.9658	4600	5.3392	-
-1	-1	-	0.4650

Framework Versions

Python: 3.12.11
Sentence Transformers: 5.1.1
Transformers: 4.56.2
PyTorch: 2.8.0+cu126
Accelerate: 1.10.1
Datasets: 2.20.0
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",
}

CoSENTLoss

bibtex

@article{10531646,
    author={Huang, Xiang and Peng, Hao and Zou, Dongcheng and Liu, Zhiwei and Li, Jianxin and Liu, Kay and Wu, Jia and Su, Jianlin and Yu, Philip S.},
    journal={IEEE/ACM Transactions on Audio, Speech, and Language Processing},
    title={CoSENT: Consistent Sentence Embedding via Similarity Ranking},
    year={2024},
    doi={10.1109/TASLP.2024.3402087}
}

	text1	text2	label
type	string	string	float
details	min: 4 tokens mean: 19.67 tokens max: 54 tokens	min: 3 tokens mean: 15.59 tokens max: 72 tokens	min: 0.0 mean: 0.53 max: 1.0