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:410745
  - loss:ContrastiveLoss
widget:
  - source_sentence: وینچ
    sentences:
      - >-
        ترقه شکلاتی ( هفت ترقه ) ناریه پارس درجه 1 بسته 15 عددی ترقه شکلاتی (
        هفت ترقه ) ناریه پارس درجه 1 بسته 15 عددی 10عدد ناریه ترقه شکلاتی هفت
        ترقه بار تازه بدون رطوبت وخرابی مارک معتبر نورافشانی
      - پارچه میکرو کجراه
      - >-
        Car winch-1500LBS-KARA وینچ خودرو آفرود ۶۸۰ کیلوگرم کارا ۱۵۰۰lbs وینچ
        خودرویی (جلو ماشینی) 1500LBS کارا (KARA)
  - source_sentence: ' وسپا '
    sentences:
      - پولوشرت زرد وسپا
      - دوچرخه بند سقفی  لیفان X70 ایکس 70 آلومینیومی طرح منابو
      - >-
        دوچرخه ویوا Oxygen سایز 26 دوچرخه 26 ويوا OXYGEN دوچرخه کوهستان ویوا مدل
        OXYGEN سایز 26
  - source_sentence: دوچرخه المپیا سایز 27 5
    sentences:
      - "دوچرخه شهری المپیا کد 16220 سایز 16 دوچرخه شهری المپیا\_کد 16220 سایز 16 دوچرخه المپیا کد 16220 سایز 16 - OLYMPIA"
      - لامپ اس ام دی خودرو مدل 8B بسته 2 عددی
      - قیمت کمپرس سنج موتور
  - source_sentence: دچرخه ی
    sentences:
      - هیدروفیشیال ۷ کاره نیوفیس پلاس متور سنگین ۲۰۲۲
      - جامدادی کیوت
      - جعبه ی کادو ی رنگی
  - source_sentence: هایومکس
    sentences:
      - انگشتر حدید صینی کد2439
      - ژل هایومکس ولومایزر 2 سی سی
      - >-
        دزدگیر پاناتک مدل P-CA501 دزدگیر پاناتک P-CA501-2 دزدگیر پاناتک مدل
        P-CA501-2
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: 0.8531738206358597
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.763870358467102
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.9032999224561303
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.7447167634963989
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 0.8649689236015621
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9451857194374323
            name: Cosine Recall
          - type: cosine_ap
            value: 0.9354580013152192
            name: Cosine Ap
          - type: dot_accuracy
            value: 0.8179627073336401
            name: Dot Accuracy
          - type: dot_accuracy_threshold
            value: 17.24372100830078
            name: Dot Accuracy Threshold
          - type: dot_f1
            value: 0.8831898479427548
            name: Dot F1
          - type: dot_f1_threshold
            value: 16.905807495117188
            name: Dot F1 Threshold
          - type: dot_precision
            value: 0.8255042324171805
            name: Dot Precision
          - type: dot_recall
            value: 0.9495432143286453
            name: Dot Recall
          - type: dot_ap
            value: 0.9192801272426158
            name: Dot Ap
          - type: manhattan_accuracy
            value: 0.8484629374000306
            name: Manhattan Accuracy
          - type: manhattan_accuracy_threshold
            value: 56.168235778808594
            name: Manhattan Accuracy Threshold
          - type: manhattan_f1
            value: 0.9006901291486498
            name: Manhattan F1
          - type: manhattan_f1_threshold
            value: 57.448089599609375
            name: Manhattan F1 Threshold
          - type: manhattan_precision
            value: 0.8601706503309084
            name: Manhattan Precision
          - type: manhattan_recall
            value: 0.9452157711263373
            name: Manhattan Recall
          - type: manhattan_ap
            value: 0.9331690796886208
            name: Manhattan Ap
          - type: euclidean_accuracy
            value: 0.8485944039089375
            name: Euclidean Accuracy
          - type: euclidean_accuracy_threshold
            value: 3.5569825172424316
            name: Euclidean Accuracy Threshold
          - type: euclidean_f1
            value: 0.9009756516265629
            name: Euclidean F1
          - type: euclidean_f1_threshold
            value: 3.694398880004883
            name: Euclidean F1 Threshold
          - type: euclidean_precision
            value: 0.8597717468465025
            name: Euclidean Precision
          - type: euclidean_recall
            value: 0.9463276836158192
            name: Euclidean Recall
          - type: euclidean_ap
            value: 0.9332275611001725
            name: Euclidean Ap
          - type: max_accuracy
            value: 0.8531738206358597
            name: Max Accuracy
          - type: max_accuracy_threshold
            value: 56.168235778808594
            name: Max Accuracy Threshold
          - type: max_f1
            value: 0.9032999224561303
            name: Max F1
          - type: max_f1_threshold
            value: 57.448089599609375
            name: Max F1 Threshold
          - type: max_precision
            value: 0.8649689236015621
            name: Max Precision
          - type: max_recall
            value: 0.9495432143286453
            name: Max Recall
          - type: max_ap
            value: 0.9354580013152192
            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_v5")
# Run inference
sentences = [
    'هایومکس',
    'ژل هایومکس ولومایزر 2 سی سی',
    'دزدگیر پاناتک مدل P-CA501 دزدگیر پاناتک P-CA501-2 دزدگیر پاناتک مدل P-CA501-2',
]
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	0.8532
cosine_accuracy_threshold	0.7639
cosine_f1	0.9033
cosine_f1_threshold	0.7447
cosine_precision	0.865
cosine_recall	0.9452
cosine_ap	0.9355
dot_accuracy	0.818
dot_accuracy_threshold	17.2437
dot_f1	0.8832
dot_f1_threshold	16.9058
dot_precision	0.8255
dot_recall	0.9495
dot_ap	0.9193
manhattan_accuracy	0.8485
manhattan_accuracy_threshold	56.1682
manhattan_f1	0.9007
manhattan_f1_threshold	57.4481
manhattan_precision	0.8602
manhattan_recall	0.9452
manhattan_ap	0.9332
euclidean_accuracy	0.8486
euclidean_accuracy_threshold	3.557
euclidean_f1	0.901
euclidean_f1_threshold	3.6944
euclidean_precision	0.8598
euclidean_recall	0.9463
euclidean_ap	0.9332
max_accuracy	0.8532
max_accuracy_threshold	56.1682
max_f1	0.9033
max_f1_threshold	57.4481
max_precision	0.865
max_recall	0.9495
max_ap	0.9355

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	-	0.8131
0.3115	500	0.0256	-
0.6231	1000	0.0179	-
0.9346	1500	0.0165	-
1.2461	2000	0.0152	-
1.5576	2500	0.0148	-
1.8692	3000	0.0144	-
2.0	3210	-	0.9355

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}
}