LeoChiuu's picture
Add new SentenceTransformer model.
bcd2e55 verified
|
raw
history blame
20.1 kB
metadata
base_model: sentence-transformers/all-MiniLM-L6-v2
datasets: []
language: []
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:216
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: Sophie why are you pressured?
    sentences:
      - Sophie Are you pressured?
      - Did you place the scarf in the fireplace?
      - A marked Globe.
  - source_sentence: Because of the red stain from the dish
    sentences:
      - Are you using my slippers?
      - Do you know this book?
      - There was a red stain on the dish
  - source_sentence: Outside
    sentences:
      - To grant the wish of having adventure
      - Let's look inside
      - Let's go outside
  - source_sentence: Actually I want a candle
    sentences:
      - Is that a cloth on the tree?
      - Did you have a beef stew for dinner?
      - Give me a candle
  - source_sentence: I found a flower pot.
    sentences:
      - Last night?
      - I found flowers.
      - Do you know this picture?
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
          type: custom-arc-semantics-data
        metrics:
          - type: cosine_accuracy
            value: 0.9818181818181818
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.26917901635169983
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.9908256880733944
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.26917901635169983
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 1
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9818181818181818
            name: Cosine Recall
          - type: cosine_ap
            value: 1
            name: Cosine Ap
          - type: dot_accuracy
            value: 0.9818181818181818
            name: Dot Accuracy
          - type: dot_accuracy_threshold
            value: 0.2691790461540222
            name: Dot Accuracy Threshold
          - type: dot_f1
            value: 0.9908256880733944
            name: Dot F1
          - type: dot_f1_threshold
            value: 0.2691790461540222
            name: Dot F1 Threshold
          - type: dot_precision
            value: 1
            name: Dot Precision
          - type: dot_recall
            value: 0.9818181818181818
            name: Dot Recall
          - type: dot_ap
            value: 1
            name: Dot Ap
          - type: manhattan_accuracy
            value: 0.9818181818181818
            name: Manhattan Accuracy
          - type: manhattan_accuracy_threshold
            value: 18.48493194580078
            name: Manhattan Accuracy Threshold
          - type: manhattan_f1
            value: 0.9908256880733944
            name: Manhattan F1
          - type: manhattan_f1_threshold
            value: 18.48493194580078
            name: Manhattan F1 Threshold
          - type: manhattan_precision
            value: 1
            name: Manhattan Precision
          - type: manhattan_recall
            value: 0.9818181818181818
            name: Manhattan Recall
          - type: manhattan_ap
            value: 1
            name: Manhattan Ap
          - type: euclidean_accuracy
            value: 0.9818181818181818
            name: Euclidean Accuracy
          - type: euclidean_accuracy_threshold
            value: 1.2088721990585327
            name: Euclidean Accuracy Threshold
          - type: euclidean_f1
            value: 0.9908256880733944
            name: Euclidean F1
          - type: euclidean_f1_threshold
            value: 1.2088721990585327
            name: Euclidean F1 Threshold
          - type: euclidean_precision
            value: 1
            name: Euclidean Precision
          - type: euclidean_recall
            value: 0.9818181818181818
            name: Euclidean Recall
          - type: euclidean_ap
            value: 1
            name: Euclidean Ap
          - type: max_accuracy
            value: 0.9818181818181818
            name: Max Accuracy
          - type: max_accuracy_threshold
            value: 18.48493194580078
            name: Max Accuracy Threshold
          - type: max_f1
            value: 0.9908256880733944
            name: Max F1
          - type: max_f1_threshold
            value: 18.48493194580078
            name: Max F1 Threshold
          - type: max_precision
            value: 1
            name: Max Precision
          - type: max_recall
            value: 0.9818181818181818
            name: Max Recall
          - type: max_ap
            value: 1
            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. 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

Model Sources

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 found a flower pot.',
    'I found flowers.',
    'Do you know this picture?',
]
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

Metric Value
cosine_accuracy 0.9818
cosine_accuracy_threshold 0.2692
cosine_f1 0.9908
cosine_f1_threshold 0.2692
cosine_precision 1.0
cosine_recall 0.9818
cosine_ap 1.0
dot_accuracy 0.9818
dot_accuracy_threshold 0.2692
dot_f1 0.9908
dot_f1_threshold 0.2692
dot_precision 1.0
dot_recall 0.9818
dot_ap 1.0
manhattan_accuracy 0.9818
manhattan_accuracy_threshold 18.4849
manhattan_f1 0.9908
manhattan_f1_threshold 18.4849
manhattan_precision 1.0
manhattan_recall 0.9818
manhattan_ap 1.0
euclidean_accuracy 0.9818
euclidean_accuracy_threshold 1.2089
euclidean_f1 0.9908
euclidean_f1_threshold 1.2089
euclidean_precision 1.0
euclidean_recall 0.9818
euclidean_ap 1.0
max_accuracy 0.9818
max_accuracy_threshold 18.4849
max_f1 0.9908
max_f1_threshold 18.4849
max_precision 1.0
max_recall 0.9818
max_ap 1.0

Training Details

Training Dataset

Unnamed Dataset

  • Size: 216 training samples
  • Columns: text1, text2, and label
  • Approximate statistics based on the first 1000 samples:
    text1 text2 label
    type string string int
    details
    • min: 3 tokens
    • mean: 7.19 tokens
    • max: 13 tokens
    • min: 3 tokens
    • mean: 7.49 tokens
    • max: 13 tokens
    • 1: 100.00%
  • Samples:
    text1 text2 label
    Let's search inside Let's look inside 1
    Do you see your scarf in the wagon? Is your scarf in the wagon? 1
    Scarf on the tree. Is that a scarf, the one on the tree? 1
  • Loss: MultipleNegativesRankingLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "cos_sim"
    }
    

Evaluation Dataset

Unnamed Dataset

  • Size: 55 evaluation samples
  • Columns: text1, text2, and label
  • Approximate statistics based on the first 1000 samples:
    text1 text2 label
    type string string int
    details
    • min: 3 tokens
    • mean: 7.04 tokens
    • max: 12 tokens
    • min: 3 tokens
    • mean: 7.55 tokens
    • max: 13 tokens
    • 1: 100.00%
  • Samples:
    text1 text2 label
    A candle I want a candle 1
    I did I did it 1
    When you had dinner Before cooking dinner 1
  • Loss: MultipleNegativesRankingLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "cos_sim"
    }
    

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_max_ap
None 0 - - 1.0
1.0 27 0.2251 0.1920 1.0
2.0 54 0.1218 0.1768 1.0
3.0 81 0.0466 0.1644 1.0
4.0 108 0.0231 0.1514 1.0
5.0 135 0.0161 0.1374 1.0
6.0 162 0.0119 0.1339 1.0
7.0 189 0.0091 0.1331 1.0
8.0 216 0.0074 0.1292 1.0
9.0 243 0.0054 0.1265 1.0
10.0 270 0.0059 0.1244 1.0
11.0 297 0.0055 0.1254 1.0
12.0 324 0.0068 0.1236 1.0
13.0 351 0.0035 0.1234 1.0

Framework Versions

  • Python: 3.10.14
  • Sentence Transformers: 3.0.1
  • Transformers: 4.44.0
  • PyTorch: 2.4.0+cu121
  • Accelerate: 0.33.0
  • 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",
}

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}
}