albertus-sussex/veriscrape-sbert-camera-wo-ref-deepseek-chat

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: 32 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': 32, '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")
# Run inference
sentences = [
    ': Bell + Howell',
    ': Olympus',
    'Bell & Howell Z10T ZoomTouch 10MP Touchscreen Digital Camera with Movie Mode, 3x Optical Zoom Lens, 3.0" LCD Screen, USB 2.0 - Silver',
]
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	0.9742

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.8823
silhouette_euclidean	0.7565

Triplet

• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9733

Silhouette

• Evaluated with veriscrape.training.SilhouetteEvaluator

Metric	Value
silhouette_cosine	0.8747
silhouette_euclidean	0.7523

Training Details

Training Dataset

Unnamed Dataset

• Size: 6,964 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: 12.74 tokens max: 32 tokens	min: 3 tokens mean: 12.66 tokens max: 32 tokens	min: 3 tokens mean: 12.28 tokens max: 32 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	0: ~9.80% 1: ~9.00% 2: ~6.80% 3: ~11.00% 4: ~10.20% 5: ~9.60% 6: ~10.80% 7: ~10.70% 8: ~10.30% 9: ~11.80%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  Casio Computer Co., Ltd	Eastman Kodak Company	$324.99	manufacturer	price	9
  $188.99	$96.99	Panasonic	price	manufacturer	9
  GE J1250 Point & Shoot Digital Camera - Black	Pentax K-r 12.4 Megapixel Digital SLR Camera (Body with Lens Kit) - 18 mm-55 mm - Red	$299.00	model	price	2

• Loss: TripletLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

• Size: 774 evaluation samples
• Columns: anchor, positive, negative, pos_attr_name, neg_attr_name, and website_id
• Approximate statistics based on the first 774 samples:

  	anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  type	string	string	string	string	string	int
  details	min: 3 tokens mean: 13.09 tokens max: 32 tokens	min: 3 tokens mean: 13.14 tokens max: 32 tokens	min: 3 tokens mean: 12.73 tokens max: 32 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	min: 3 tokens mean: 3.0 tokens max: 3 tokens	0: ~10.98% 1: ~11.11% 2: ~5.30% 3: ~10.59% 4: ~10.34% 5: ~9.04% 6: ~10.59% 7: ~10.85% 8: ~12.02% 9: ~9.17%

• Samples:

  anchor	positive	negative	pos_attr_name	neg_attr_name	website_id
  Eastman Kodak Company	Panasonic	Kodak EasyShare C143 12 Megapixel Compact Camera - Silver	manufacturer	model	9
  SL605 - Digital camera - compact - 12.2 Mpix - optical zoom: 5 x - supported memory: SD, SDHC - black (EC-SL605ZBPBUS)	EASYSHARE C195 - Digital camera - compact - 14.0 Mpix - optical zoom: 5 x - supported memory: SD, SDHC - silver (8770414)	$184.99	model	price	7
  $186.99	$96.99	10.2MP Cyber-shot TX7 Digital Camera - Red (DSCTX7/R)	price	model	7

• 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	-	-	0.9057	0.3675
1.0	55	0.3791	0.2219	0.9703	0.8847
2.0	110	0.1751	0.2184	0.9703	0.8844
3.0	165	0.1592	0.2365	0.9703	0.8832
4.0	220	0.1416	0.2636	0.9742	0.8812
5.0	275	0.1219	0.2560	0.9742	0.8823
-1	-1	-	-	0.9733	0.8747

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 4.0.1
• 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}
}