cl-nagoya-sup-simcse-ja-nss-v1_0_2

SentenceTransformer

This is a sentence-transformers model trained. 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:** [Unknown](https://huggingface.co/unknown) -->

• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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': True, 'pooling_mode_mean_tokens': False, '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("Detomo/cl-nagoya-sup-simcse-ja-nss-v1_0_2")
# Run inference
sentences = [
    '科目：ユニット及びその他。名称：#FHCU#床室ｶｳﾝﾀｰ。',
    '科目：ユニット及びその他。名称：#階数表示（階段室内･踊り場）。',
    '科目：ユニット及びその他。名称：Co-#ﾋﾟｸﾄｻｲﾝ。',
]
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]

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

Unnamed Dataset

• Size: 16,547 training samples
• Columns: <code>sentence</code> and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	sentence	label
  type	string	int
  details	<ul><li>min: 11 tokens</li><li>mean: 18.77 tokens</li><li>max: 72 tokens</li></ul>	<ul><li>0: ~0.30%</li><li>1: ~0.30%</li><li>2: ~0.30%</li><li>3: ~0.30%</li><li>4: ~0.30%</li><li>5: ~3.40%</li><li>6: ~0.30%</li><li>7: ~0.30%</li><li>8: ~0.30%</li><li>9: ~0.30%</li><li>10: ~0.30%</li><li>11: ~0.30%</li><li>12: ~0.40%</li><li>13: ~0.30%</li><li>14: ~0.30%</li><li>15: ~0.40%</li><li>16: ~0.30%</li><li>17: ~0.30%</li><li>18: ~0.30%</li><li>19: ~0.90%</li><li>20: ~0.30%</li><li>21: ~0.30%</li><li>22: ~1.10%</li><li>23: ~0.30%</li><li>24: ~0.30%</li><li>25: ~0.30%</li><li>26: ~0.30%</li><li>27: ~0.30%</li><li>28: ~0.30%</li><li>29: ~0.30%</li><li>30: ~0.30%</li><li>31: ~0.30%</li><li>32: ~0.30%</li><li>33: ~0.30%</li><li>34: ~0.30%</li><li>35: ~0.30%</li><li>36: ~0.30%</li><li>37: ~0.30%</li><li>38: ~0.30%</li><li>39: ~0.30%</li><li>40: ~0.40%</li><li>41: ~0.30%</li><li>42: ~0.30%</li><li>43: ~0.30%</li><li>44: ~0.60%</li><li>45: ~0.70%</li><li>46: ~0.30%</li><li>47: ~0.30%</li><li>48: ~0.30%</li><li>49: ~0.30%</li><li>50: ~0.30%</li><li>51: ~0.30%</li><li>52: ~0.30%</li><li>53: ~0.30%</li><li>54: ~0.30%</li><li>55: ~0.30%</li><li>56: ~0.30%</li><li>57: ~0.80%</li><li>58: ~0.30%</li><li>59: ~0.30%</li><li>60: ~0.60%</li><li>61: ~0.30%</li><li>62: ~0.30%</li><li>63: ~0.30%</li><li>64: ~0.50%</li><li>65: ~0.30%</li><li>66: ~0.30%</li><li>67: ~0.30%</li><li>68: ~0.30%</li><li>69: ~0.50%</li><li>70: ~0.60%</li><li>71: ~0.30%</li><li>72: ~0.30%</li><li>73: ~0.30%</li><li>74: ~0.30%</li><li>75: ~0.30%</li><li>76: ~0.30%</li><li>77: ~0.30%</li><li>78: ~0.50%</li><li>79: ~0.30%</li><li>80: ~0.30%</li><li>81: ~0.30%</li><li>82: ~0.30%</li><li>83: ~0.80%</li><li>84: ~0.60%</li><li>85: ~0.50%</li><li>86: ~0.30%</li><li>87: ~0.30%</li><li>88: ~16.50%</li><li>89: ~0.30%</li><li>90: ~0.30%</li><li>91: ~0.30%</li><li>92: ~0.30%</li><li>93: ~0.30%</li><li>94: ~0.30%</li><li>95: ~0.30%</li><li>96: ~0.30%</li><li>97: ~0.50%</li><li>98: ~0.30%</li><li>99: ~0.30%</li><li>100: ~0.30%</li><li>101: ~0.30%</li><li>102: ~0.30%</li><li>103: ~0.30%</li><li>104: ~0.30%</li><li>105: ~0.70%</li><li>106: ~0.70%</li><li>107: ~0.30%</li><li>108: ~3.20%</li><li>109: ~0.30%</li><li>110: ~0.40%</li><li>111: ~2.30%</li><li>112: ~0.30%</li><li>113: ~0.30%</li><li>114: ~0.50%</li><li>115: ~0.50%</li><li>116: ~0.50%</li><li>117: ~0.40%</li><li>118: ~0.30%</li><li>119: ~0.30%</li><li>120: ~0.30%</li><li>121: ~0.80%</li><li>122: ~0.30%</li><li>123: ~0.30%</li><li>124: ~0.30%</li><li>125: ~0.30%</li><li>126: ~0.30%</li><li>127: ~0.30%</li><li>128: ~0.30%</li><li>129: ~0.30%</li><li>130: ~0.50%</li><li>131: ~0.30%</li><li>132: ~0.40%</li><li>133: ~0.30%</li><li>134: ~0.30%</li><li>135: ~0.30%</li><li>136: ~0.30%</li><li>137: ~0.30%</li><li>138: ~0.30%</li><li>139: ~0.30%</li><li>140: ~0.30%</li><li>141: ~0.30%</li><li>142: ~0.30%</li><li>143: ~0.40%</li><li>144: ~0.30%</li><li>145: ~0.30%</li><li>146: ~0.30%</li><li>147: ~0.30%</li><li>148: ~0.30%</li><li>149: ~0.30%</li><li>150: ~0.70%</li><li>151: ~0.30%</li><li>152: ~0.30%</li><li>153: ~0.30%</li><li>154: ~1.30%</li><li>155: ~0.30%</li><li>156: ~0.40%</li><li>157: ~0.30%</li><li>158: ~0.30%</li><li>159: ~0.30%</li><li>160: ~1.50%</li><li>161: ~0.30%</li><li>162: ~0.30%</li><li>163: ~0.30%</li><li>164: ~0.30%</li><li>165: ~0.30%</li><li>166: ~0.30%</li><li>167: ~0.30%</li><li>168: ~1.50%</li><li>169: ~0.30%</li><li>170: ~0.30%</li><li>171: ~7.20%</li><li>172: ~0.30%</li><li>173: ~1.00%</li><li>174: ~0.30%</li><li>175: ~0.30%</li><li>176: ~0.30%</li><li>177: ~1.80%</li><li>178: ~0.30%</li><li>179: ~0.50%</li><li>180: ~0.70%</li><li>181: ~0.10%</li></ul>

• Samples:

  sentence	label
  <code>科目：コンクリート。名称：免震基礎天端ｸﾞﾗｳﾄ注入。</code>	<code>0</code>
  <code>科目：コンクリート。名称：免震基礎天端ｸﾞﾗｳﾄ注入。</code>	<code>0</code>
  <code>科目：コンクリート。名称：免震基礎天端ｸﾞﾗｳﾄ注入。</code>	<code>0</code>

• Loss: <code>sentence_transformer_lib.custom_batch_all_trip_loss.CustomBatchAllTripletLoss</code>

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• learning_rate: 1e-05
• weight_decay: 0.01
• num_train_epochs: 200
• warmup_ratio: 0.15
• fp16: True
• batch_sampler: group_by_label

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• 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: 1e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 200
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.15
• 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}
• tp_size: 0
• 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: group_by_label
• multi_dataset_batch_sampler: proportional

</details>

Training Logs

Framework Versions

• Python: 3.11.12
• Sentence Transformers: 3.4.1
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.2
• Datasets: 3.5.0
• 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",
}

CustomBatchAllTripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->