README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:33
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: keepitreal/vietnamese-sbert
widget:
  - source_sentence: Áo Polo Lacoste với chất liệu Petit Piqué và thiết kế cổ gập kinh điển
    sentences:
      - Giày cao gót đẳng cấp
      - Xe điều khiển từ xa
      - Áo polo sang trọng
  - source_sentence: Sony Alpha A7 IV với cảm biến CMOS Exmor R 33MP và khả năng quay 4K 60fps
    sentences:
      - Giày cao gót sang trọng
      - Sách văn học tuổi thơ
      - Máy ảnh chuyên nghiệp
  - source_sentence: >-
      Laneige Water Bank Cream với công nghệ Hydro Ionized Mineral Water và kết
      cấu gel mỏng nhẹ
    sentences:
      - Điện thoại flagship cao cấp
      - Kem dưỡng ẩm nổi bật
      - Giày tây nam lịch lãm
  - source_sentence: Adidas Ultraboost với công nghệ Boost™ và đế ngoài Continental™ Rubber
    sentences:
      - Tai nghe chống ồn hàng đầu
      - Quần short kaki trẻ trung
      - Giày chạy bộ hiện đại
  - source_sentence: Áo thun từ cotton mềm mại, kiểu dáng đa dạng phù hợp cho nhiều phong cách
    sentences:
      - Máy ảnh vlog chuyên nghiệp
      - Áo thun thoải mái
      - Laptop hiệu năng mạnh mẽ
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
model-index:
  - name: SentenceTransformer based on keepitreal/vietnamese-sbert
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.25
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.75
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.25
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.15000000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.25
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.75
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.6377310833652008
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.525
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.525
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.25
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.75
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.25
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.15000000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.25
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.75
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.6079899373088598
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.4861111111111111
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.4861111111111111
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.25
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.75
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.75
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.25
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.15000000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.07500000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.25
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.75
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.75
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.41666666666666663
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.43749999999999994
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.5
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.75
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.75
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.5
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.15000000000000002
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.5
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.75
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.75
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7410657717261977
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.6607142857142857
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.6607142857142857
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.5
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.5
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.75
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.16666666666666666
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.1
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.07500000000000001
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.5
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.5
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.75
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4045166735627343
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.29166666666666663
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.30952380952380953
            name: Cosine Map@100

SentenceTransformer based on keepitreal/vietnamese-sbert

This is a sentence-transformers model finetuned from keepitreal/vietnamese-sbert on the json 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: keepitreal/vietnamese-sbert
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • json

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}) with Transformer model: RobertaModel 
  (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("zxcvo/product-search-model")
# Run inference
sentences = [
    'Áo thun từ cotton mềm mại, kiểu dáng đa dạng phù hợp cho nhiều phong cách',
    'Áo thun thoải mái',
    'Laptop hiệu năng mạnh mẽ',
]
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

Information Retrieval

• Datasets: dim_768, dim_512, dim_256, dim_128 and dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	dim_768	dim_512	dim_256	dim_128	dim_64
cosine_accuracy@1	0.25	0.25	0.25	0.5	0.0
cosine_accuracy@3	0.75	0.75	0.75	0.75	0.5
cosine_accuracy@5	0.75	0.75	0.75	0.75	0.5
cosine_accuracy@10	1.0	1.0	0.75	1.0	0.75
cosine_precision@1	0.25	0.25	0.25	0.5	0.0
cosine_precision@3	0.25	0.25	0.25	0.25	0.1667
cosine_precision@5	0.15	0.15	0.15	0.15	0.1
cosine_precision@10	0.1	0.1	0.075	0.1	0.075
cosine_recall@1	0.25	0.25	0.25	0.5	0.0
cosine_recall@3	0.75	0.75	0.75	0.75	0.5
cosine_recall@5	0.75	0.75	0.75	0.75	0.5
cosine_recall@10	1.0	1.0	0.75	1.0	0.75
cosine_ndcg@10	0.6377	0.608	0.5	0.7411	0.4045
cosine_mrr@10	0.525	0.4861	0.4167	0.6607	0.2917
cosine_map@100	0.525	0.4861	0.4375	0.6607	0.3095

Training Details

Training Dataset

json

• Dataset: json
• Size: 33 training samples
• Columns: positive and anchor
• Approximate statistics based on the first 33 samples:

  	positive	anchor
  type	string	string
  details	min: 18 tokens mean: 22.82 tokens max: 34 tokens	min: 6 tokens mean: 6.97 tokens max: 8 tokens

• Samples:

  positive	anchor
  Áo Sơ Mi Nam Trắng Classic với chất liệu cotton cao cấp, kiểu dáng lịch lãm	Áo sơ mi tinh tế
  Đắc Nhân Tâm của Dale Carnegie với những nguyên tắc xây dựng mối quan hệ hiệu quả	Sách kinh điển về giao tiếp
  Nike Air Force 1 với thiết kế logo Swoosh và công nghệ Air-Sole	Giày sneaker cổ điển

• 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

• eval_strategy: epoch
• per_device_train_batch_size: 32
• gradient_accumulation_steps: 16
• learning_rate: 2e-05
• num_train_epochs: 4
• bf16: True
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 8
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 16
• 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: 4
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: True
• fp16: False
• 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: True
• 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
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
1.0	1	0.6050	0.6050	0.5	0.7083	0.5767
2.0	2	0.605	0.608	0.5	0.7411	0.4045
3.0	3	0.6377	0.6080	0.5	0.6488	0.4045
4.0	4	0.6377	0.6080	0.5	0.7411	0.4045

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.0
• Sentence Transformers: 3.3.1
• Transformers: 4.41.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• Datasets: 2.19.1
• 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}
}