Chanisorn/thai-food-mpnet-tuned

Thai Food Ingredients → Dish Prediction

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-mpnet-base-v2. 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: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Language: th
• License: apache-2.0

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: XLMRobertaModel 
  (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("thai-food-mpnet-tuned")
# Run inference
sentences = [
    'นมเปรี้ยว, เยลลี่รวมรสผลไม้',
    'ไอศกรีมโยเกิร์ตเยลลี่ปีโป้',
    'ทองหยอด',
]
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

• Dataset: thai-food-eval
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.551
cosine_accuracy@3	0.7959
cosine_accuracy@5	0.898
cosine_accuracy@10	0.9592
cosine_precision@1	0.551
cosine_precision@3	0.2653
cosine_precision@5	0.1796
cosine_recall@1	0.551
cosine_recall@3	0.7959
cosine_recall@5	0.898
cosine_ndcg@10	0.7617
cosine_mrr@10	0.698
cosine_map@100	0.6998

Training Details

Training Dataset

Unnamed Dataset

• Size: 400 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 400 samples:

  	anchor	positive
  type	string	string
  details	min: 4 tokens mean: 44.22 tokens max: 105 tokens	min: 4 tokens mean: 9.92 tokens max: 24 tokens

• Samples:

  anchor	positive
  ผักกระเฉด, กุ้งผ่า, ปลากระป๋อง, พริกสด, หอมแดง, กระชาย, กระปิ, เกลือป่น, น้ำตาล, น้ำปลา, น้ำมะขามเปียก, มะนาว	แกงส้มผักกระเฉด
  หมูสามชั้น, น้ำส้มสายชู, เกลือ, น้ำมันงา, กระเทียม, หอมแดง, รากผักชี, เต้าเจี้ยว, ซอสหอยนางรม, น้ำตาล, ซีอิ๊วดำ, น้ำซุป, นมสด, แป้งมัน, งาขาว	ข้าวหมูกรอบ
  ซอสถั่วเหลืองจิ้ม, ไข่ไก่, เบคอน, ไส้กรอก, แฮม, ชาร้อน, น้ำสะอาด	ชุดอาหารเช้า

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 49 evaluation samples
• Columns: anchor and positive
• Approximate statistics based on the first 49 samples:

  	anchor	positive
  type	string	string
  details	min: 8 tokens mean: 43.39 tokens max: 105 tokens	min: 5 tokens mean: 9.39 tokens max: 18 tokens

• Samples:

  anchor	positive
  เห็ด, กะปิ, น้ำปลา, ตะไคร้, หอมแดง, พริกขี้หนู, พริกแดง, ผักหวาน, ชะอม, ใบแมงลัก, ใบย่านาง, น้ำเปล่า, หน่อไม้	แกงเห็ดผักหวานใส่กะปิ
  หมูสามชั้น, พริกไทย, กระเทียม, รากผักชี, อบเชย, ดอกจันทร์, ซีอิ้วขาว, ซีอิ้วดำ, น้ำตาลทราย, น้ำตาลปี๊บ, เกลือ, คนอร์	หมูฮ้อง
  สะโพกไก่, เกลือ, พริกไทย, มันฝรั่ง, แครอท, แป้งสาลี, น้ำมัน, เนย, กระเทียม, หอม, ผงปรุงรส, ซอสมะเขือเทศ	สตูไก่ สูตรไม่ใส่เครื่องเทศ

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 24
• per_device_eval_batch_size: 24
• learning_rate: 5e-06
• num_train_epochs: 16
• warmup_ratio: 0.1
• load_best_model_at_end: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 24
• per_device_eval_batch_size: 24
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 5e-06
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 16
• 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: 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: None
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• 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
• 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
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	thai-food-eval_cosine_ndcg@10
0.5882	10	3.2903	-	-
1.0	17	-	2.3505	0.4150
1.1765	20	2.467	-	-
1.7647	30	1.9808	-	-
2.0	34	-	1.5080	0.6269
2.3529	40	1.6699	-	-
2.9412	50	1.3974	-	-
3.0	51	-	1.2515	0.6827
3.5294	60	1.1846	-	-
4.0	68	-	1.1762	0.6735
4.1176	70	1.1183	-	-
4.7059	80	1.001	-	-
5.0	85	-	1.1104	0.6995
5.2941	90	0.9919	-	-
5.8824	100	0.8285	-	-
6.0	102	-	1.0806	0.7152
6.4706	110	0.7873	-	-
7.0	119	-	1.0573	0.7333
7.0588	120	0.7359	-	-
7.6471	130	0.6526	-	-
8.0	136	-	1.0051	0.7566
8.2353	140	0.6004	-	-
8.8235	150	0.571	-	-
9.0	153	-	0.9988	0.7700
9.4118	160	0.5254	-	-
10.0	170	0.5119	1.0087	0.7590
10.5882	180	0.492	-	-
11.0	187	-	0.9756	0.77
11.1765	190	0.5334	-	-
11.7647	200	0.4395	-	-
12.0	204	-	0.9851	0.7707
12.3529	210	0.4362	-	-
12.9412	220	0.3905	-	-
13.0	221	-	1.0020	0.7578
13.5294	230	0.4388	-	-
14.0	238	-	0.9994	0.7563
14.1176	240	0.4594	-	-
14.7059	250	0.4502	-	-
15.0	255	-	1.0004	0.7631
15.2941	260	0.3539	-	-
15.8824	270	0.4144	-	-
16.0	272	-	0.9997	0.7617

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.12
• Sentence Transformers: 4.1.0
• Transformers: 4.52.2
• PyTorch: 2.6.0+cu124
• Accelerate: 1.7.0
• Datasets: 2.14.4
• Tokenizers: 0.21.1

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",
}

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}
}