README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1788
  - loss:TripletLoss
base_model: Alibaba-NLP/gte-base-en-v1.5
widget:
  - source_sentence: $194.99
    sentences:
      - Polaroid t1235 Point & Shoot Digital Camera - Blue
      - $199.00
      - price
      - model
  - source_sentence: >-
      Recertified: OLYMPUS Stylus 1040 Black 10.0 MP 2.7" 230K LCD 3X Optical
      Zoom Digital Camera
    sentences:
      - >-
        Nikon COOLPIX S60 10 Megapixel Digital Camera w/ 5x Optical Zoom,
        Optical VR Image Stabilization, 3.5" High Res Touch Panel LCD, HD
        Pictmotion Slide Shows, & Autofocus and Auto Exposure - Red
      - model
      - manufacturer
      - Olympus America Inc.
  - source_sentence: Kodak
    sentences:
      - Fuji
      - model
      - manufacturer
      - FUJIFILM Instax 210 Instant Photo Camera
  - source_sentence: $95.99
    sentences:
      - $299.00
      - model
      - >-
        SONY Cyber-shot S2100 Silver 12.1 MP 3.0" 230K LCD 3X Optical Zoom
        Digital Camera
      - price
  - source_sentence: >-
      Nikon COOLPIX S640 Precious Pink 12.2 MP 2.7" 230K LCD 5X Optical Zoom
      28mm Wide Angle Digital Camera
    sentences:
      - $649.00
      - price
      - model
      - >-
        Pentax K-r 12.4 Megapixel Digital SLR Camera (Body with Lens Kit) - 18
        mm-55 mm (Lens 1), 55 mm-300 mm (Lens 2) - Black
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy
  - silhouette_cosine
  - silhouette_euclidean
model-index:
  - name: SentenceTransformer based on Alibaba-NLP/gte-base-en-v1.5
    results:
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: cosine_accuracy
            value: 1
            name: Cosine Accuracy
          - type: cosine_accuracy
            value: 1
            name: Cosine Accuracy
      - task:
          type: silhouette
          name: Silhouette
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: silhouette_cosine
            value: 0.983250617980957
            name: Silhouette Cosine
          - type: silhouette_euclidean
            value: 0.8853786587715149
            name: Silhouette Euclidean
          - type: silhouette_cosine
            value: 0.9841798543930054
            name: Silhouette Cosine
          - type: silhouette_euclidean
            value: 0.888205349445343
            name: Silhouette Euclidean

SentenceTransformer based on Alibaba-NLP/gte-base-en-v1.5

This is a sentence-transformers model finetuned from Alibaba-NLP/gte-base-en-v1.5. 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: Alibaba-NLP/gte-base-en-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 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: NewModel 
  (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("albertus-sussex/veriscrape-sbert-camera-reference_2_to_verify_8-fold-10")
# Run inference
sentences = [
    'Nikon COOLPIX S640 Precious Pink 12.2 MP 2.7" 230K LCD 5X Optical Zoom 28mm Wide Angle Digital Camera',
    'Pentax K-r 12.4 Megapixel Digital SLR Camera (Body with Lens Kit) - 18 mm-55 mm (Lens 1), 55 mm-300 mm (Lens 2) - Black',
    '$649.00',
]
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

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	1.0

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.9833
silhouette_euclidean	0.8854

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	1.0

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.9842
silhouette_euclidean	0.8882

Training Details

Training Dataset

Unnamed Dataset

• Size: 1,788 training samples
• Columns: anchor, positive, negative, pos_attr_name, and neg_attr_name
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name
  type	string	string	string	string	string
  details	min: 3 tokens mean: 12.37 tokens max: 50 tokens	min: 3 tokens mean: 11.89 tokens max: 55 tokens	min: 3 tokens mean: 10.37 tokens max: 45 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name
  Sony	FUJIFILM	SONY Cyber-shot S2100 Silver 12.1 MP 3.0" 230K LCD 3X Optical Zoom Digital Camera	manufacturer	model
  SAKAR-VIVITAR	Kodak	Vivitar ViviCam T328 Black 12.0 MP 3.0" LCD 3X Optical Zoom Digital Camera	manufacturer	model
  Panasonic	Sony	Panasonic Lumix DMC-FX37 Point & Shoot Digital Camera - White	manufacturer	model

• Loss: TripletLoss with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 5
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 199 evaluation samples
• Columns: anchor, positive, negative, pos_attr_name, and neg_attr_name
• Approximate statistics based on the first 199 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name
  type	string	string	string	string	string
  details	min: 3 tokens mean: 10.52 tokens max: 50 tokens	min: 3 tokens mean: 10.95 tokens max: 55 tokens	min: 3 tokens mean: 9.41 tokens max: 51 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name
  Pentax Imaging	Pentax	FinePix XP10 Compact Camera	manufacturer	model
  Olympus America Inc.	Canon	$143.02	manufacturer	price
  $403.41	$179.95	Nikon Coolpix S210 Point & Shoot Digital Camera - Plum	price	model

• Loss: TripletLoss with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 5
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• num_train_epochs: 5
• warmup_ratio: 0.1

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• 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: 5e-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: 5
• 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: False
• 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
• use_liger_kernel: False
• eval_use_gather_object: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	cosine_accuracy	silhouette_cosine
-1	-1	-	-	0.8492	0.3908
1.0	14	0.5592	0.0	1.0	0.9717
2.0	28	0.0	0.0	1.0	0.9823
3.0	42	0.0	0.0	1.0	0.9831
4.0	56	0.0	0.0	1.0	0.9832
5.0	70	0.0	0.0	1.0	0.9833
-1	-1	-	-	1.0	0.9842

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 4.0.1
• Transformers: 4.45.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.5.2
• Datasets: 3.1.0
• Tokenizers: 0.20.3

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",
}

TripletLoss

@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}
}