metadata

base_model: colorfulscoop/sbert-base-ja
library_name: sentence-transformers
metrics:
  - cosine_accuracy
  - cosine_accuracy_threshold
  - cosine_f1
  - cosine_f1_threshold
  - cosine_precision
  - cosine_recall
  - cosine_ap
  - dot_accuracy
  - dot_accuracy_threshold
  - dot_f1
  - dot_f1_threshold
  - dot_precision
  - dot_recall
  - dot_ap
  - manhattan_accuracy
  - manhattan_accuracy_threshold
  - manhattan_f1
  - manhattan_f1_threshold
  - manhattan_precision
  - manhattan_recall
  - manhattan_ap
  - euclidean_accuracy
  - euclidean_accuracy_threshold
  - euclidean_f1
  - euclidean_f1_threshold
  - euclidean_precision
  - euclidean_recall
  - euclidean_ap
  - max_accuracy
  - max_accuracy_threshold
  - max_f1
  - max_f1_threshold
  - max_precision
  - max_recall
  - max_ap
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:401
  - loss:CosineSimilarityLoss
widget:
  - source_sentence: 昨夜何を食べたの？
    sentences:
      - 昨日作ったのはチキンヌードル？
      - 村長がどうしたの？
      - 雲より高くってどこ？
  - source_sentence: ナイトスタンドにある？
    sentences:
      - わかんない
      - スリッパ履いた？
      - だめじゃん
  - source_sentence: センパイ
    sentences:
      - 昨日の夜は暑かった
      - 調子はどう？
      - トーチ
  - source_sentence: 村人はどんな呪文を使うの？
    sentences:
      - スパイク
      - キミはどんな魔法を使うの？
      - うさんくさい
  - source_sentence: 祭壇の些細な違和感ってどういう意味？
    sentences:
      - 青いオーブがどこにあるか知ってる？
      - 赤い染みが皿にあった
      - これが花
model-index:
  - name: SentenceTransformer based on colorfulscoop/sbert-base-ja
    results:
      - task:
          type: binary-classification
          name: Binary Classification
        dataset:
          name: custom arc semantics data jp
          type: custom-arc-semantics-data-jp
        metrics:
          - type: cosine_accuracy
            value: 0.8855721393034826
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.6970740556716919
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.7766990291262137
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.6637545228004456
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 0.8163265306122449
            name: Cosine Precision
          - type: cosine_recall
            value: 0.7407407407407407
            name: Cosine Recall
          - type: cosine_ap
            value: 0.6606381605892593
            name: Cosine Ap
          - type: dot_accuracy
            value: 0.8805970149253731
            name: Dot Accuracy
          - type: dot_accuracy_threshold
            value: 378.6933898925781
            name: Dot Accuracy Threshold
          - type: dot_f1
            value: 0.7647058823529411
            name: Dot F1
          - type: dot_f1_threshold
            value: 378.6933898925781
            name: Dot F1 Threshold
          - type: dot_precision
            value: 0.8125
            name: Dot Precision
          - type: dot_recall
            value: 0.7222222222222222
            name: Dot Recall
          - type: dot_ap
            value: 0.6865123266544332
            name: Dot Ap
          - type: manhattan_accuracy
            value: 0.8855721393034826
            name: Manhattan Accuracy
          - type: manhattan_accuracy_threshold
            value: 407.9349365234375
            name: Manhattan Accuracy Threshold
          - type: manhattan_f1
            value: 0.7766990291262137
            name: Manhattan F1
          - type: manhattan_f1_threshold
            value: 426.941650390625
            name: Manhattan F1 Threshold
          - type: manhattan_precision
            value: 0.8163265306122449
            name: Manhattan Precision
          - type: manhattan_recall
            value: 0.7407407407407407
            name: Manhattan Recall
          - type: manhattan_ap
            value: 0.6609390536301427
            name: Manhattan Ap
          - type: euclidean_accuracy
            value: 0.8855721393034826
            name: Euclidean Accuracy
          - type: euclidean_accuracy_threshold
            value: 18.663713455200195
            name: Euclidean Accuracy Threshold
          - type: euclidean_f1
            value: 0.7766990291262137
            name: Euclidean F1
          - type: euclidean_f1_threshold
            value: 19.35655975341797
            name: Euclidean F1 Threshold
          - type: euclidean_precision
            value: 0.8163265306122449
            name: Euclidean Precision
          - type: euclidean_recall
            value: 0.7407407407407407
            name: Euclidean Recall
          - type: euclidean_ap
            value: 0.6602743223356511
            name: Euclidean Ap
          - type: max_accuracy
            value: 0.8855721393034826
            name: Max Accuracy
          - type: max_accuracy_threshold
            value: 407.9349365234375
            name: Max Accuracy Threshold
          - type: max_f1
            value: 0.7766990291262137
            name: Max F1
          - type: max_f1_threshold
            value: 426.941650390625
            name: Max F1 Threshold
          - type: max_precision
            value: 0.8163265306122449
            name: Max Precision
          - type: max_recall
            value: 0.7407407407407407
            name: Max Recall
          - type: max_ap
            value: 0.6865123266544332
            name: Max Ap

SentenceTransformer based on colorfulscoop/sbert-base-ja

This is a sentence-transformers model finetuned from colorfulscoop/sbert-base-ja on the csv 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: colorfulscoop/sbert-base-ja
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Dataset:
- csv

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("sentence_transformers_model_id")
# Run inference
sentences = [
    '祭壇の些細な違和感ってどういう意味？',
    '青いオーブがどこにあるか知ってる？',
    'これが花',
]
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

Binary Classification

Dataset: custom-arc-semantics-data-jp
Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.8856
cosine_accuracy_threshold	0.6971
cosine_f1	0.7767
cosine_f1_threshold	0.6638
cosine_precision	0.8163
cosine_recall	0.7407
cosine_ap	0.6606
dot_accuracy	0.8806
dot_accuracy_threshold	378.6934
dot_f1	0.7647
dot_f1_threshold	378.6934
dot_precision	0.8125
dot_recall	0.7222
dot_ap	0.6865
manhattan_accuracy	0.8856
manhattan_accuracy_threshold	407.9349
manhattan_f1	0.7767
manhattan_f1_threshold	426.9417
manhattan_precision	0.8163
manhattan_recall	0.7407
manhattan_ap	0.6609
euclidean_accuracy	0.8856
euclidean_accuracy_threshold	18.6637
euclidean_f1	0.7767
euclidean_f1_threshold	19.3566
euclidean_precision	0.8163
euclidean_recall	0.7407
euclidean_ap	0.6603
max_accuracy	0.8856
max_accuracy_threshold	407.9349
max_f1	0.7767
max_f1_threshold	426.9417
max_precision	0.8163
max_recall	0.7407
max_ap	0.6865

Training Details

Training Dataset

csv

Dataset: csv
Size: 401 training samples
Columns: text1, text2, and label
Approximate statistics based on the first 401 samples:
text1 text2 label
type string string int
details
min: 4 tokens
mean: 8.34 tokens
max: 14 tokens

min: 4 tokens
mean: 7.9 tokens
max: 14 tokens

0: ~67.00%
1: ~33.00%

	text1	text2	label
type	string	string	int
details	min: 4 tokens mean: 8.34 tokens max: 14 tokens	min: 4 tokens mean: 7.9 tokens max: 14 tokens	0: ~67.00% 1: ~33.00%

Samples:

text1	text2	label
`雲より高くってどういう意味？`	`猫好き`	`0`
`花の囁きってなに？`	`リリアンについて教えて`	`0`
`リリアンってものの姿を変える魔法を使える？`	`どんな魔法なの？`	`0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Evaluation Dataset

csv

Dataset: csv
Size: 401 evaluation samples
Columns: text1, text2, and label
Approximate statistics based on the first 401 samples:
text1 text2 label
type string string int
details
min: 4 tokens
mean: 8.23 tokens
max: 15 tokens

min: 4 tokens
mean: 7.46 tokens
max: 14 tokens

0: ~73.13%
1: ~26.87%

	text1	text2	label
type	string	string	int
details	min: 4 tokens mean: 8.23 tokens max: 15 tokens	min: 4 tokens mean: 7.46 tokens max: 14 tokens	0: ~73.13% 1: ~26.87%

Samples:

text1	text2	label
`棚からトマトがなくなってたから`	`トマトが棚からなくなっていたから`	`1`
`欲しくない`	`家の中へ行こう`	`0`
`昨日は何を作ったの？`	`ビーフシチュー食べた？`	`0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
learning_rate: 2e-05
num_train_epochs: 10
warmup_ratio: 0.4
fp16: 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: 8
per_device_eval_batch_size: 8
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: 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: 10
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.4
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}
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
eval_on_start: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	custom-arc-semantics-data-jp_max_ap
1.0	25	0.2393	0.2390	0.5655
2.0	50	0.1724	0.1689	0.6075
3.0	75	0.1046	0.1326	0.6478
4.0	100	0.062	0.1183	0.6618
5.0	125	0.0349	0.1158	0.6683
6.0	150	0.0258	0.1142	0.6772
7.0	175	0.0211	0.1168	0.6739
8.0	200	0.0204	0.1180	0.6765
9.0	225	0.0194	0.1178	0.6869
10.0	250	0.0185	0.1180	0.6865

Framework Versions

Python: 3.10.14
Sentence Transformers: 3.1.0
Transformers: 4.44.2
PyTorch: 2.4.1+cu121
Accelerate: 0.34.2
Datasets: 2.20.0
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",
}