README.md

metadata

base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
datasets: []
language: []
library_name: sentence-transformers
metrics:
  - cosine_accuracy
  - cosine_accuracy_threshold
  - cosine_f1
  - cosine_f1_threshold
  - cosine_precision
  - cosine_recall
  - cosine_ap
  - dot_accuracy
  - dot_accuracy_threshold
  - dot_f1
  - dot_f1_threshold
  - dot_precision
  - dot_recall
  - dot_ap
  - manhattan_accuracy
  - manhattan_accuracy_threshold
  - manhattan_f1
  - manhattan_f1_threshold
  - manhattan_precision
  - manhattan_recall
  - manhattan_ap
  - euclidean_accuracy
  - euclidean_accuracy_threshold
  - euclidean_f1
  - euclidean_f1_threshold
  - euclidean_precision
  - euclidean_recall
  - euclidean_ap
  - max_accuracy
  - max_accuracy_threshold
  - max_f1
  - max_f1_threshold
  - max_precision
  - max_recall
  - max_ap
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:64116
  - loss:ContrastiveLoss
widget:
  - source_sentence: مبل سلتان
    sentences:
      - روسری جین شش عددی عمده نخی
      - مبل راحتی چستر سالوادور مبل راحتی چستر مبل راحتی چستر مکانیزم
      - پاور سانروف فابریک برلیانس
  - source_sentence: لباس پلیسی
    sentences:
      - جا عودی
      - لباس خواب کاستوم فانتزی پلیسی زنانه
      - >-
        روغن حنا (پرپشت کننده مو  ریزش مو  تقویت مو  تقویت ابرو  جلوگیری از
        سفیدی مو  شوره مو  خشکی پوست سر  خارش پوست سر)
  - source_sentence: قابلمه سنگی
    sentences:
      - قابلمه سنگی آقای سنگی 10 نفره
      - گاز مبرد R134a پوکا (POKKA R134)
      - کفش فوتبال بچه گانه آدیداس طرح اصلی مشکی سفید Adidas
  - source_sentence: لوازم آرایشی
    sentences:
      - >-
        جعبه لوازم آرایشی قابل حمل سازمان‌دهنده لوازم آرایش مسافرتی با روکش آینه
        چراغ‌دار LED لوازم آرایشی
      - کفش پاشنه بلند مجلسی دخترانه
      - وکتور بنر فارسی جشن تولد با کیک و جعبه کادو
  - source_sentence: پوست مصنوعی
    sentences:
      - دستگیره حیاطی تک پیچ سرباز دستگیره تک پیچ درب حیاطی سرباز
      - مبل سلطنتی
      - کیف پوست ماری مستطیل جنس چرم مصنوعی کیف پوست ماری مستطیل
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
    results:
      - task:
          type: binary-classification
          name: Binary Classification
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: cosine_accuracy
            value: 0.7607017543859649
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.7412481904029846
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.834358186010761
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.7125277519226074
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 0.7491373360938578
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9414570685169124
            name: Cosine Recall
          - type: cosine_ap
            value: 0.8461870777524143
            name: Cosine Ap
          - type: dot_accuracy
            value: 0.7104561403508772
            name: Dot Accuracy
          - type: dot_accuracy_threshold
            value: 14.821020126342773
            name: Dot Accuracy Threshold
          - type: dot_f1
            value: 0.8054054054054054
            name: Dot F1
          - type: dot_f1_threshold
            value: 14.108308792114258
            name: Dot F1 Threshold
          - type: dot_precision
            value: 0.7062765609676365
            name: Dot Precision
          - type: dot_recall
            value: 0.9369037294015612
            name: Dot Recall
          - type: dot_ap
            value: 0.8122928586516915
            name: Dot Ap
          - type: manhattan_accuracy
            value: 0.7528421052631579
            name: Manhattan Accuracy
          - type: manhattan_accuracy_threshold
            value: 53.40993118286133
            name: Manhattan Accuracy Threshold
          - type: manhattan_f1
            value: 0.828743211792087
            name: Manhattan F1
          - type: manhattan_f1_threshold
            value: 55.60980987548828
            name: Manhattan F1 Threshold
          - type: manhattan_precision
            value: 0.7496491228070176
            name: Manhattan Precision
          - type: manhattan_recall
            value: 0.9264960971379012
            name: Manhattan Recall
          - type: manhattan_ap
            value: 0.8423084093127031
            name: Manhattan Ap
          - type: euclidean_accuracy
            value: 0.7536842105263157
            name: Euclidean Accuracy
          - type: euclidean_accuracy_threshold
            value: 3.543578863143921
            name: Euclidean Accuracy Threshold
          - type: euclidean_f1
            value: 0.829423689545323
            name: Euclidean F1
          - type: euclidean_f1_threshold
            value: 3.609351396560669
            name: Euclidean F1 Threshold
          - type: euclidean_precision
            value: 0.7475204454497999
            name: Euclidean Precision
          - type: euclidean_recall
            value: 0.9314830875975716
            name: Euclidean Recall
          - type: euclidean_ap
            value: 0.8422044822515327
            name: Euclidean Ap
          - type: max_accuracy
            value: 0.7607017543859649
            name: Max Accuracy
          - type: max_accuracy_threshold
            value: 53.40993118286133
            name: Max Accuracy Threshold
          - type: max_f1
            value: 0.834358186010761
            name: Max F1
          - type: max_f1_threshold
            value: 55.60980987548828
            name: Max F1 Threshold
          - type: max_precision
            value: 0.7496491228070176
            name: Max Precision
          - type: max_recall
            value: 0.9414570685169124
            name: Max Recall
          - type: max_ap
            value: 0.8461870777524143
            name: Max Ap

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2. 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: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 tokens
• Similarity Function: Cosine Similarity

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': 128, 'do_lower_case': False}) with Transformer model: 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})
)

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 SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("DashReza7/sentence-transformers_paraphrase-multilingual-MiniLM-L12-v2_FINETUNED_on_torob_data_v6")
# Run inference
sentences = [
    'پوست مصنوعی',
    'کیف پوست ماری مستطیل جنس چرم مصنوعی کیف پوست ماری مستطیل',
    'مبل سلطنتی',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Binary Classification

• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.7607
cosine_accuracy_threshold	0.7412
cosine_f1	0.8344
cosine_f1_threshold	0.7125
cosine_precision	0.7491
cosine_recall	0.9415
cosine_ap	0.8462
dot_accuracy	0.7105
dot_accuracy_threshold	14.821
dot_f1	0.8054
dot_f1_threshold	14.1083
dot_precision	0.7063
dot_recall	0.9369
dot_ap	0.8123
manhattan_accuracy	0.7528
manhattan_accuracy_threshold	53.4099
manhattan_f1	0.8287
manhattan_f1_threshold	55.6098
manhattan_precision	0.7496
manhattan_recall	0.9265
manhattan_ap	0.8423
euclidean_accuracy	0.7537
euclidean_accuracy_threshold	3.5436
euclidean_f1	0.8294
euclidean_f1_threshold	3.6094
euclidean_precision	0.7475
euclidean_recall	0.9315
euclidean_ap	0.8422
max_accuracy	0.7607
max_accuracy_threshold	53.4099
max_f1	0.8344
max_f1_threshold	55.6098
max_precision	0.7496
max_recall	0.9415
max_ap	0.8462

Training Details

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 256
• per_device_eval_batch_size: 256
• learning_rate: 2e-05
• num_train_epochs: 2
• warmup_ratio: 0.1
• fp16: 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: 256
• 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
• 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}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch
• 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• dispatch_batches: None
• split_batches: 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
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	max_ap
None	0	-	0.7365
1.9920	500	0.0242	-
2.0	502	-	0.8462

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.1
• Transformers: 4.42.4
• PyTorch: 2.4.0+cu121
• Accelerate: 0.32.1
• Datasets: 2.21.0
• Tokenizers: 0.19.1

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",
}

ContrastiveLoss

@inproceedings{hadsell2006dimensionality,
    author={Hadsell, R. and Chopra, S. and LeCun, Y.},
    booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, 
    title={Dimensionality Reduction by Learning an Invariant Mapping}, 
    year={2006},
    volume={2},
    number={},
    pages={1735-1742},
    doi={10.1109/CVPR.2006.100}
}