README.md

metadata

base_model: sentence-transformers/all-MiniLM-L6-v2
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:593
  - loss:OnlineContrastiveLoss
widget:
  - source_sentence: What city
    sentences:
      - What magic do other villagers use?
      - What does between the gods mean?
      - what about the city
  - source_sentence: What's your name?
    sentences:
      - what mystery?
      - Is this the flower
      - A globe.
  - source_sentence: I think we'll find dragons.
    sentences:
      - Do you know a mage who changes shape of material?
      - I don't think we'll find dragons.
      - The curtain is moving in the breeze
  - source_sentence: What happened to her?
    sentences:
      - Is this the flower
      - Do you have a second bucket?
      - There was a red stain on the dish
  - source_sentence: I don't see tomato on the shelf
    sentences:
      - What magic do other villagers use?
      - Yes please
      - Because the pot smelled spicy
model-index:
  - name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
    results:
      - task:
          type: binary-classification
          name: Binary Classification
        dataset:
          name: custom arc semantics data en
          type: custom-arc-semantics-data-en
        metrics:
          - type: cosine_accuracy
            value: 0.9495798319327731
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.6676459908485413
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.9
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.6361173391342163
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 0.9
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9
            name: Cosine Recall
          - type: cosine_ap
            value: 0.8400025542161988
            name: Cosine Ap
          - type: dot_accuracy
            value: 0.9495798319327731
            name: Dot Accuracy
          - type: dot_accuracy_threshold
            value: 0.6676459908485413
            name: Dot Accuracy Threshold
          - type: dot_f1
            value: 0.9
            name: Dot F1
          - type: dot_f1_threshold
            value: 0.6361173391342163
            name: Dot F1 Threshold
          - type: dot_precision
            value: 0.9
            name: Dot Precision
          - type: dot_recall
            value: 0.9
            name: Dot Recall
          - type: dot_ap
            value: 0.8400025542161988
            name: Dot Ap
          - type: manhattan_accuracy
            value: 0.9495798319327731
            name: Manhattan Accuracy
          - type: manhattan_accuracy_threshold
            value: 12.677780151367188
            name: Manhattan Accuracy Threshold
          - type: manhattan_f1
            value: 0.896551724137931
            name: Manhattan F1
          - type: manhattan_f1_threshold
            value: 12.677780151367188
            name: Manhattan F1 Threshold
          - type: manhattan_precision
            value: 0.9285714285714286
            name: Manhattan Precision
          - type: manhattan_recall
            value: 0.8666666666666667
            name: Manhattan Recall
          - type: manhattan_ap
            value: 0.8387174899512584
            name: Manhattan Ap
          - type: euclidean_accuracy
            value: 0.9495798319327731
            name: Euclidean Accuracy
          - type: euclidean_accuracy_threshold
            value: 0.8152118921279907
            name: Euclidean Accuracy Threshold
          - type: euclidean_f1
            value: 0.9
            name: Euclidean F1
          - type: euclidean_f1_threshold
            value: 0.8530915379524231
            name: Euclidean F1 Threshold
          - type: euclidean_precision
            value: 0.9
            name: Euclidean Precision
          - type: euclidean_recall
            value: 0.9
            name: Euclidean Recall
          - type: euclidean_ap
            value: 0.8400025542161988
            name: Euclidean Ap
          - type: max_accuracy
            value: 0.9495798319327731
            name: Max Accuracy
          - type: max_accuracy_threshold
            value: 12.677780151367188
            name: Max Accuracy Threshold
          - type: max_f1
            value: 0.9
            name: Max F1
          - type: max_f1_threshold
            value: 12.677780151367188
            name: Max F1 Threshold
          - type: max_precision
            value: 0.9285714285714286
            name: Max Precision
          - type: max_recall
            value: 0.9
            name: Max Recall
          - type: max_ap
            value: 0.8400025542161988
            name: Max Ap

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the csv dataset. It maps sentences & paragraphs to a 384-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/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 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': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (2): Normalize()
)

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("LeoChiuu/all-MiniLM-L6-v2-arc")
# Run inference
sentences = [
    "I don't see tomato on the shelf",
    'Because the pot smelled spicy',
    'What magic do other villagers use?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# 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-en
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.9496
cosine_accuracy_threshold	0.6676
cosine_f1	0.9
cosine_f1_threshold	0.6361
cosine_precision	0.9
cosine_recall	0.9
cosine_ap	0.84
dot_accuracy	0.9496
dot_accuracy_threshold	0.6676
dot_f1	0.9
dot_f1_threshold	0.6361
dot_precision	0.9
dot_recall	0.9
dot_ap	0.84
manhattan_accuracy	0.9496
manhattan_accuracy_threshold	12.6778
manhattan_f1	0.8966
manhattan_f1_threshold	12.6778
manhattan_precision	0.9286
manhattan_recall	0.8667
manhattan_ap	0.8387
euclidean_accuracy	0.9496
euclidean_accuracy_threshold	0.8152
euclidean_f1	0.9
euclidean_f1_threshold	0.8531
euclidean_precision	0.9
euclidean_recall	0.9
euclidean_ap	0.84
max_accuracy	0.9496
max_accuracy_threshold	12.6778
max_f1	0.9
max_f1_threshold	12.6778
max_precision	0.9286
max_recall	0.9
max_ap	0.84

Training Details

Training Dataset

csv

• Dataset: csv
• Size: 593 training samples
• Columns: text1, text2, and label
• Approximate statistics based on the first 593 samples:

  	text1	text2	label
  type	string	string	int
  details	min: 3 tokens mean: 7.2 tokens max: 18 tokens	min: 3 tokens mean: 6.8 tokens max: 18 tokens	0: ~68.14% 1: ~31.86%

• Samples:

  text1	text2	label
  Something is different	What did you say?	0
  what are the properties?	what about Jack?	0
  hint	hints	1

• Loss: OnlineContrastiveLoss

Evaluation Dataset

csv

• Dataset: csv
• Size: 593 evaluation samples
• Columns: text1, text2, and label
• Approximate statistics based on the first 593 samples:

  	text1	text2	label
  type	string	string	int
  details	min: 3 tokens mean: 7.26 tokens max: 13 tokens	min: 3 tokens mean: 7.13 tokens max: 13 tokens	0: ~74.79% 1: ~25.21%

• Samples:

  text1	text2	label
  To have an adventure with us	Its name is Oblivion.	0
  Is the scarf on the nightstand?	Are you using my slippers?	0
  To test Unravel Spell	Tell me about Lily	0

• Loss: OnlineContrastiveLoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• learning_rate: 2e-05
• num_train_epochs: 13
• warmup_ratio: 0.1
• 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: 13
• 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: 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-en_max_ap
None	0	-	-	0.7634
1.0	60	0.3053	0.1297	0.7825
2.0	120	0.1478	0.1071	0.8071
3.0	180	0.0357	0.0904	0.8387
4.0	240	0.0139	0.0829	0.8412
5.0	300	0.017	0.0704	0.8429
6.0	360	0.0132	0.0779	0.8411
7.0	420	0.0	0.0700	0.8433
8.0	480	0.0079	0.0808	0.8403
9.0	540	0.0098	0.0808	0.8404
10.0	600	0.0039	0.0804	0.8387
11.0	660	0.0001	0.0815	0.8398
12.0	720	0.0039	0.0816	0.8397
13.0	780	0.0034	0.0814	0.8400

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",
}