README.md

metadata

language: []
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:557850
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
datasets: []
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: ذكر متوازن بعناية يقف على قدم واحدة بالقرب من منطقة شاطئ المحيط النظيفة
    sentences:
      - رجل يقدم عرضاً
      - هناك رجل بالخارج قرب الشاطئ
      - رجل يجلس على أريكه
  - source_sentence: رجل يقفز إلى سريره القذر
    sentences:
      - السرير قذر.
      - رجل يضحك أثناء غسيل الملابس
      - الرجل على القمر
  - source_sentence: الفتيات بالخارج
    sentences:
      - امرأة تلف الخيط إلى كرات بجانب كومة من الكرات
      - فتيان يركبان في جولة متعة
      - >-
        ثلاث فتيات يقفون سوية في غرفة واحدة تستمع وواحدة تكتب على الحائط
        والثالثة تتحدث إليهن
  - source_sentence: الرجل يرتدي قميصاً أزرق.
    sentences:
      - >-
        رجل يرتدي قميصاً أزرق يميل إلى الجدار بجانب الطريق مع شاحنة زرقاء وسيارة
        حمراء مع الماء في الخلفية.
      - كتاب القصص مفتوح
      - رجل يرتدي قميص أسود يعزف على الجيتار.
  - source_sentence: يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة شابة.
    sentences:
      - ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه
      - رجل يستلقي على وجهه على مقعد في الحديقة.
      - الشاب نائم بينما الأم تقود ابنتها إلى الحديقة
pipeline_tag: sentence-similarity
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test 256
          type: sts-test-256
        metrics:
          - type: pearson_cosine
            value: 0.8264447022356382
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8386403752382455
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8219134931449013
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.825509659109493
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8223094468630248
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8260503151751462
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.6375226884845725
            name: Pearson Dot
          - type: spearman_dot
            value: 0.6287228614640888
            name: Spearman Dot
          - type: pearson_max
            value: 0.8264447022356382
            name: Pearson Max
          - type: spearman_max
            value: 0.8386403752382455
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test 128
          type: sts-test-128
        metrics:
          - type: pearson_cosine
            value: 0.8209661910768973
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8347149482673766
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8082811559854036
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.8148314269262763
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8093138512113149
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8156468458613929
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.5795109620454884
            name: Pearson Dot
          - type: spearman_dot
            value: 0.5760223026552876
            name: Spearman Dot
          - type: pearson_max
            value: 0.8209661910768973
            name: Pearson Max
          - type: spearman_max
            value: 0.8347149482673766
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts test 64
          type: sts-test-64
        metrics:
          - type: pearson_cosine
            value: 0.808708530451336
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8217532539767914
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.7876121380998453
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.7969092304137347
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.7902997966909958
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7987635968785215
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.495047136234386
            name: Pearson Dot
          - type: spearman_dot
            value: 0.49287000679901516
            name: Spearman Dot
          - type: pearson_max
            value: 0.808708530451336
            name: Pearson Max
          - type: spearman_max
            value: 0.8217532539767914
            name: Spearman Max

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 on the Omartificial-Intelligence-Space/arabic-n_li-triplet dataset. 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
• Training Dataset:
  • Omartificial-Intelligence-Space/arabic-n_li-triplet

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("Omartificial-Intelligence-Space/MiniLM-L12-v2-all-nli-triplet")
# Run inference
sentences = [
    'يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة شابة.',
    'ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه',
    'الشاب نائم بينما الأم تقود ابنتها إلى الحديقة',
]
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

Semantic Similarity

• Dataset: sts-test-256
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8264
spearman_cosine	0.8386
pearson_manhattan	0.8219
spearman_manhattan	0.8255
pearson_euclidean	0.8223
spearman_euclidean	0.8261
pearson_dot	0.6375
spearman_dot	0.6287
pearson_max	0.8264
spearman_max	0.8386

Semantic Similarity

• Dataset: sts-test-128
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.821
spearman_cosine	0.8347
pearson_manhattan	0.8083
spearman_manhattan	0.8148
pearson_euclidean	0.8093
spearman_euclidean	0.8156
pearson_dot	0.5795
spearman_dot	0.576
pearson_max	0.821
spearman_max	0.8347

Semantic Similarity

• Dataset: sts-test-64
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8087
spearman_cosine	0.8218
pearson_manhattan	0.7876
spearman_manhattan	0.7969
pearson_euclidean	0.7903
spearman_euclidean	0.7988
pearson_dot	0.495
spearman_dot	0.4929
pearson_max	0.8087
spearman_max	0.8218

Training Details

Training Dataset

Omartificial-Intelligence-Space/arabic-n_li-triplet

• Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet
• Size: 557,850 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 5 tokens mean: 10.33 tokens max: 52 tokens	min: 5 tokens mean: 13.21 tokens max: 49 tokens	min: 5 tokens mean: 15.32 tokens max: 53 tokens

• Samples:

  anchor	positive	negative
  شخص على حصان يقفز فوق طائرة معطلة	شخص في الهواء الطلق، على حصان.	شخص في مطعم، يطلب عجة.
  أطفال يبتسمون و يلوحون للكاميرا	هناك أطفال حاضرون	الاطفال يتجهمون
  صبي يقفز على لوح التزلج في منتصف الجسر الأحمر.	الفتى يقوم بخدعة التزلج	الصبي يتزلج على الرصيف

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

Omartificial-Intelligence-Space/arabic-n_li-triplet

• Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet
• Size: 6,584 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 5 tokens mean: 21.86 tokens max: 105 tokens	min: 4 tokens mean: 10.22 tokens max: 49 tokens	min: 4 tokens mean: 11.2 tokens max: 33 tokens

• Samples:

  anchor	positive	negative
  امرأتان يتعانقان بينما يحملان حزمة	إمرأتان يحملان حزمة	الرجال يتشاجرون خارج مطعم
  طفلين صغيرين يرتديان قميصاً أزرق، أحدهما يرتدي الرقم 9 والآخر يرتدي الرقم 2 يقفان على خطوات خشبية في الحمام ويغسلان أيديهما في المغسلة.	طفلين يرتديان قميصاً مرقماً يغسلون أيديهم	طفلين يرتديان سترة يذهبان إلى المدرسة
  رجل يبيع الدونات لعميل خلال معرض عالمي أقيم في مدينة أنجليس	رجل يبيع الدونات لعميل	امرأة تشرب قهوتها في مقهى صغير

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• learning_rate: 5e-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: 1
• 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
• 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, '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_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	sts-test-128_spearman_cosine	sts-test-256_spearman_cosine	sts-test-64_spearman_cosine
0.0229	200	6.2204	-	-	-
0.0459	400	4.9559	-	-	-
0.0688	600	4.7835	-	-	-
0.0918	800	4.2725	-	-	-
0.1147	1000	4.291	-	-	-
0.1377	1200	4.0704	-	-	-
0.1606	1400	3.7962	-	-	-
0.1835	1600	3.7447	-	-	-
0.2065	1800	3.569	-	-	-
0.2294	2000	3.5373	-	-	-
0.2524	2200	3.608	-	-	-
0.2753	2400	3.5609	-	-	-
0.2983	2600	3.5231	-	-	-
0.3212	2800	3.3312	-	-	-
0.3442	3000	3.4803	-	-	-
0.3671	3200	3.3552	-	-	-
0.3900	3400	3.3024	-	-	-
0.4130	3600	3.2559	-	-	-
0.4359	3800	3.1882	-	-	-
0.4589	4000	3.227	-	-	-
0.4818	4200	3.0889	-	-	-
0.5048	4400	3.0861	-	-	-
0.5277	4600	3.0178	-	-	-
0.5506	4800	3.231	-	-	-
0.5736	5000	3.1593	-	-	-
0.5965	5200	3.1101	-	-	-
0.6195	5400	3.1307	-	-	-
0.6424	5600	3.1265	-	-	-
0.6654	5800	3.1116	-	-	-
0.6883	6000	3.1417	-	-	-
0.7113	6200	3.0862	-	-	-
0.7342	6400	2.9652	-	-	-
0.7571	6600	2.8466	-	-	-
0.7801	6800	2.271	-	-	-
0.8030	7000	2.046	-	-	-
0.8260	7200	1.9634	-	-	-
0.8489	7400	1.8875	-	-	-
0.8719	7600	1.7655	-	-	-
0.8948	7800	1.6874	-	-	-
0.9177	8000	1.7315	-	-	-
0.9407	8200	1.6674	-	-	-
0.9636	8400	1.6574	-	-	-
0.9866	8600	1.6142	-	-	-
1.0	8717	-	0.8347	0.8386	0.8218

Framework Versions

• Python: 3.9.18
• Sentence Transformers: 3.0.1
• Transformers: 4.40.0
• PyTorch: 2.2.2+cu121
• Accelerate: 0.26.1
• Datasets: 2.19.0
• 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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}