Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka

SentenceTransformer based on aubmindlab/bert-base-arabertv02

This is a sentence-transformers model finetuned from aubmindlab/bert-base-arabertv02 on the Omartificial-Intelligence-Space/arabic-n_li-triplet dataset. It maps sentences & paragraphs to a 768-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: aubmindlab/bert-base-arabertv02
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 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': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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/Arabic-arabert-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, 768]

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

Evaluation

Metrics

Semantic Similarity

• Dataset: sts-test-768
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.595
spearman_cosine	0.616
pearson_manhattan	0.6296
spearman_manhattan	0.627
pearson_euclidean	0.6327
spearman_euclidean	0.6317
pearson_dot	0.4282
spearman_dot	0.4295
pearson_max	0.6327
spearman_max	0.6317

Semantic Similarity

• Dataset: sts-test-512
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5846
spearman_cosine	0.6064
pearson_manhattan	0.6288
spearman_manhattan	0.6264
pearson_euclidean	0.6313
spearman_euclidean	0.6302
pearson_dot	0.3789
spearman_dot	0.3768
pearson_max	0.6313
spearman_max	0.6302

Semantic Similarity

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

Metric	Value
pearson_cosine	0.5779
spearman_cosine	0.596
pearson_manhattan	0.6243
spearman_manhattan	0.6217
pearson_euclidean	0.6238
spearman_euclidean	0.6215
pearson_dot	0.3597
spearman_dot	0.353
pearson_max	0.6243
spearman_max	0.6217

Semantic Similarity

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

Metric	Value
pearson_cosine	0.5831
spearman_cosine	0.6022
pearson_manhattan	0.6152
spearman_manhattan	0.6122
pearson_euclidean	0.6162
spearman_euclidean	0.6153
pearson_dot	0.4044
spearman_dot	0.4015
pearson_max	0.6162
spearman_max	0.6153

Semantic Similarity

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

Metric	Value
pearson_cosine	0.5725
spearman_cosine	0.5914
pearson_manhattan	0.6024
spearman_manhattan	0.5967
pearson_euclidean	0.6069
spearman_euclidean	0.6041
pearson_dot	0.3632
spearman_dot	0.3585
pearson_max	0.6069
spearman_max	0.6041

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: 4 tokens mean: 8.02 tokens max: 41 tokens	min: 4 tokens mean: 10.03 tokens max: 34 tokens	min: 4 tokens mean: 10.72 tokens max: 38 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": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          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: 4 tokens mean: 14.87 tokens max: 70 tokens	min: 4 tokens mean: 7.54 tokens max: 26 tokens	min: 4 tokens mean: 8.14 tokens max: 23 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": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          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-512_spearman_cosine	sts-test-64_spearman_cosine	sts-test-768_spearman_cosine
0.0229	200	14.4811	-	-	-	-	-
0.0459	400	9.0389	-	-	-	-	-
0.0688	600	8.1478	-	-	-	-	-
0.0918	800	7.168	-	-	-	-	-
0.1147	1000	7.1998	-	-	-	-	-
0.1377	1200	6.7985	-	-	-	-	-
0.1606	1400	6.3754	-	-	-	-	-
0.1835	1600	6.3202	-	-	-	-	-
0.2065	1800	5.9186	-	-	-	-	-
0.2294	2000	5.9594	-	-	-	-	-
0.2524	2200	6.0211	-	-	-	-	-
0.2753	2400	5.9984	-	-	-	-	-
0.2983	2600	5.8321	-	-	-	-	-
0.3212	2800	5.621	-	-	-	-	-
0.3442	3000	5.9004	-	-	-	-	-
0.3671	3200	5.562	-	-	-	-	-
0.3900	3400	5.5125	-	-	-	-	-
0.4130	3600	5.4922	-	-	-	-	-
0.4359	3800	5.3023	-	-	-	-	-
0.4589	4000	5.4376	-	-	-	-	-
0.4818	4200	5.1048	-	-	-	-	-
0.5048	4400	5.0605	-	-	-	-	-
0.5277	4600	4.9985	-	-	-	-	-
0.5506	4800	5.2594	-	-	-	-	-
0.5736	5000	5.2183	-	-	-	-	-
0.5965	5200	5.1621	-	-	-	-	-
0.6195	5400	5.166	-	-	-	-	-
0.6424	5600	5.2241	-	-	-	-	-
0.6654	5800	5.1342	-	-	-	-	-
0.6883	6000	5.2267	-	-	-	-	-
0.7113	6200	5.1083	-	-	-	-	-
0.7342	6400	5.0119	-	-	-	-	-
0.7571	6600	4.6471	-	-	-	-	-
0.7801	6800	3.6699	-	-	-	-	-
0.8030	7000	3.2954	-	-	-	-	-
0.8260	7200	3.1039	-	-	-	-	-
0.8489	7400	3.001	-	-	-	-	-
0.8719	7600	2.8992	-	-	-	-	-
0.8948	7800	2.7504	-	-	-	-	-
0.9177	8000	2.7891	-	-	-	-	-
0.9407	8200	2.7157	-	-	-	-	-
0.9636	8400	2.6795	-	-	-	-	-
0.9866	8600	2.6278	-	-	-	-	-
1.0	8717	-	0.6022	0.5960	0.6064	0.5914	0.6160

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}
}