README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:7134
  - loss:TripletLoss
base_model: Alibaba-NLP/gte-base-en-v1.5
widget:
  - source_sentence: NIKON - PHOTO
    sentences:
      - from
      - manufacturer
      - price
      - Pentax Imaging
  - source_sentence: $108.99
    sentences:
      - model
      - $87.99
      - price
      - Coolpix S80 Compact Camera
  - source_sentence: ': Casio'
    sentences:
      - Casio Exlim EX-Z1200 12MP Digtial Camera with 3x Anti Shake Optical Zoom
      - model
      - ': Fuji'
      - manufacturer
  - source_sentence: >-
      Panasonic Dmc-fx37s 10mp Digital Camera 5x Optical Zoom 2.5" Lcd 25mm
      Leica Lens (dmcfx37s)
    sentences:
      - model
      - $84.62
      - price
      - >-
        Ge C1033 Point & Shoot Digital Camera - 10.1 Megapixel - 2.40" Active
        Matrix Tft Color Lcd - Black 3x Optical Zoom - 5.7x - Ge C1033-bk
        (c1033bk)
  - source_sentence: $108.99
    sentences:
      - price
      - Panasonic
      - manufacturer
      - $123.99
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: 0.9974779486656189
            name: Cosine Accuracy
          - type: cosine_accuracy
            value: 0.9988649487495422
            name: Cosine Accuracy
      - task:
          type: silhouette
          name: Silhouette
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: silhouette_cosine
            value: 0.9446604251861572
            name: Silhouette Cosine
          - type: silhouette_euclidean
            value: 0.838313102722168
            name: Silhouette Euclidean
          - type: silhouette_cosine
            value: 0.9412728548049927
            name: Silhouette Cosine
          - type: silhouette_euclidean
            value: 0.832588791847229
            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: 64 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': 64, '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-wo-ref-deepseek-chat-0324")
# Run inference
sentences = [
    '$108.99',
    '$123.99',
    'Panasonic',
]
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	0.9975

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.9447
silhouette_euclidean	0.8383

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9989

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.9413
silhouette_euclidean	0.8326

Training Details

Training Dataset

Unnamed Dataset

• Size: 7,134 training samples
• Columns: anchor, positive, negative, pos_attr_name, neg_attr_name, and website_id
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  type	string	string	string	string	string	int
  details	min: 3 tokens mean: 12.12 tokens max: 64 tokens	min: 3 tokens mean: 12.28 tokens max: 64 tokens	min: 3 tokens mean: 11.66 tokens max: 64 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	0: ~8.80% 1: ~9.40% 2: ~12.30% 3: ~9.60% 4: ~11.00% 5: ~6.50% 6: ~10.00% 7: ~11.00% 8: ~10.70% 9: ~10.70%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  Sakar International, Inc	Fuji Photo Film Co. Ltd	Coolpix S1100pj Compact Camera	manufacturer	model	9
  Olympus Stylus Tough 3000 Point & Shoot Digital Camera - 12 Megapixel - 2.70" Lcd - Pink 3.6x Optical Zoom - 5x (227625)	Canon Powershot A470 Digital Camera With Selphy Cp760 Compact Photo Printer - Blue - 7.1 Megapixel - 16:9 - 3.4x Optical Zoom - 4x Digital Zoom - 2.5" Active Matrix Tft Color Lcd - 32mb Secure Digital	Olympus Corporation	model	manufacturer	1
  $204.95	$89.00	Fujifilm Z800EXR 12 MP Digital Point and Shoot Camera (Red) BigVALUEInc 8PC Saver Bundle!	price	model	0

• Loss: TripletLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

• Size: 793 evaluation samples
• Columns: anchor, positive, negative, pos_attr_name, neg_attr_name, and website_id
• Approximate statistics based on the first 793 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  type	string	string	string	string	string	int
  details	min: 3 tokens mean: 12.85 tokens max: 64 tokens	min: 3 tokens mean: 12.51 tokens max: 64 tokens	min: 3 tokens mean: 11.06 tokens max: 64 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	0: ~9.33% 1: ~10.72% 2: ~12.86% 3: ~10.84% 4: ~8.95% 5: ~5.93% 6: ~13.11% 7: ~10.34% 8: ~8.70% 9: ~9.21%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  VistaQuest Corporation	General Electric Company	EasyShare C142 Compact Camera	manufacturer	model	3
  Kodak EasyShare Z1485 IS Point & Shoot Digital Camera - Pink	Nikon Coolpix S1100pj 14.1 Megapixel Compact Camera - 5 mm-25 mm - Violet	Kodak	model	manufacturer	2
  Panasonic Lumix DMC-ZS7 Point & Shoot Digital Camera - 12.1 Megapixel - 3" Active Matrix TFT Color LCD - Black	Memorex Flash Micro Point & Shoot Digital Camera	Pentax Imaging	model	manufacturer	2

• 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	-	-	0.9180	0.3826
1.0	56	0.2179	0.0492	0.9975	0.9327
2.0	112	0.0169	0.0601	0.9975	0.9443
3.0	168	0.0191	0.0394	0.9962	0.9398
4.0	224	0.0126	0.0457	0.9975	0.9419
5.0	280	0.0135	0.0444	0.9975	0.9447
-1	-1	-	-	0.9989	0.9413

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 4.1.0
• Transformers: 4.45.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.6.0
• 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}
}