bau0221 / ptz_embedding

SentenceTransformer based on sentence-transformers/multi-qa-MiniLM-L6-cos-v1

This is a sentence-transformers model finetuned from sentence-transformers/multi-qa-MiniLM-L6-cos-v1 . 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/multi-qa-MiniLM-L6-cos-v1
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity

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': 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("bau0221/ptz_embedding")
# Run inference
sentences = [
    'group1 put Abigail on the right side',
    'Move Charlotte to the right on Camera 4',
    'Set Evelyn at the right side on Camera 1',
]
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]

Training Details

Training Dataset

Unnamed Dataset

• Size: 710 training samples
• Columns: query , pos , and neg
• Approximate statistics based on the first 710 samples:

  	query	pos	neg
  type	string	list	list
  details	min: 5 tokens mean: 11.46 tokens max: 28 tokens	size: 3 elements	size: 3 elements

• Samples:

  query	pos	neg
  Set camera 1 to track target A at bottom_right with fast speed.	['Set camera 1 to track target A at bottom_right with fast speed.', 'Set camera 1 to track target A at bottom_right with fast speed.', 'Set camera 1 to track target A at bottom_right with fast speed.']	['Camera 2 tracking Kyle', 'Set camera 3 to track target B at the top with slow speed.', 'Turn camera 2 to the right for 5 seconds.']
  Camera 2 tracking Kyle	['Camera 2 tracking Kyle', 'Camera 4 tracking Kyle', 'Cam 2 tracking Kyle']	['Set camera 1 to track target A at bottom_right with fast speed.', 'Set camera 3 to track target B at the top with slow speed.', 'Turn camera 2 to the right for 5 seconds.']
  Set camera 3 to track target B at the top with slow speed.	['Set camera 3 to track target B at the top with slow speed.', 'Set camera 3 to track target B at the top with slow speed.', 'Set camera 3 to track target B at the top with slow speed.']	['Set camera 1 to track target A at bottom_right with fast speed.', 'Camera 2 tracking Kyle', 'Turn camera 2 to the right for 5 seconds.']

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

• Size: 71 evaluation samples
• Columns: query , pos , and neg
• Approximate statistics based on the first 71 samples:

  	query	pos	neg
  type	string	list	list
  details	min: 6 tokens mean: 10.35 tokens max: 12 tokens	size: 3 elements	size: 3 elements

• Samples:

  query	pos	neg
  Camera 3 put Harper at the right side	['Camera 3 put Harper at the right side', 'Cam 3 put Harper at the right side', 'Camera 3 put Harper at the right side']	['Set camera 1 to track target A at bottom_right with fast speed.', 'Camera 2 tracking Kyle', 'Set camera 3 to track target B at the top with slow speed.']
  Camera 4 put Amelia on the left side	['Camera 4 put Amelia on the left side', 'Cam 4 put Amelia on the left side', 'Camera 4 put Amelia on the left side']	['Set camera 1 to track target A at bottom_right with fast speed.', 'Camera 2 tracking Kyle', 'Set camera 3 to track target B at the top with slow speed.']
  Group2 put Logan at the right side	['Group2 put Logan at the right side', 'Group2 put Logan at the right side', 'Group2 put Logan at the right side']	['Set camera 1 to track target A at bottom_right with fast speed.', 'Camera 2 tracking Kyle', 'Set camera 3 to track target B at the top with slow speed.']

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

All Hyperparameters

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.3.1
• Transformers: 4.47.1
• PyTorch: 2.5.1+cu121
• Accelerate: 1.2.1
• Datasets: 3.2.0
• Tokenizers: 0.21.0

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}
}