GaniduA/bge-finetuned-olscience

SentenceTransformer based on BAAI/bge-base-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-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: BAAI/bge-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: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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})
)

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("GaniduA/bge-finetuned-olscience")
# Run inference
sentences = [
    'Discuss the principles and process of electrolysis, including the conventions adopted in electrolysis.',
    'The development of artificial intelligence has significantly impacted the tech industry, leading to advancements in machine learning and natural language processing.',
    "In the movie 'Inception', directed by Christopher Nolan, the plot revolves around a skilled thief who is given a chance at redemption if he can successfully perform inception by planting an idea into someone's subconscious.",
]
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

Binary Classification

• Dataset: eval
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	1.0
cosine_accuracy_threshold	0.0571
cosine_f1	1.0
cosine_f1_threshold	0.0571
cosine_precision	1.0
cosine_recall	1.0
cosine_ap	1.0
cosine_mcc	1.0

Training Details

Training Dataset

Unnamed Dataset

• Size: 34,969 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 6 tokens mean: 17.43 tokens max: 209 tokens	min: 3 tokens mean: 25.94 tokens max: 335 tokens	min: 0.0 mean: 0.25 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  How does the reaction of zinc with copper sulfate demonstrate a single displacement reaction?	Julius Caesar crossed the Rubicon River in 49 BC, which led to a chain of events culminating in the Roman Civil War.	0.0
  How do you investigate the effect of tightening a screw on the moment of force required to rotate a stick?	Explore the depths of the ocean with a team of deep-sea divers searching for mythical sea creatures and undiscovered shipwrecks.	0.0
  Describe the operation of a photodiode in optical sensing.	A photodiode converts light into an electrical current by generating electron-hole pairs when exposed to light, used in optical sensing and communication applications.	1.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• num_train_epochs: 2
• fp16: True
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• 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
• num_train_epochs: 2
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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}
• tp_size: 0
• 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: None
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• 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
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	eval_cosine_ap
0.0366	20	-	0.9892
0.0731	40	-	0.9978
0.1097	60	-	0.9989
0.1463	80	-	0.9997
0.1828	100	-	0.9999
0.2194	120	-	0.9998
0.2559	140	-	0.9998
0.2925	160	-	0.9998
0.3291	180	-	0.9998
0.3656	200	-	0.9999
0.4022	220	-	0.9998
0.4388	240	-	0.9999
0.4753	260	-	1.0000
0.5119	280	-	1.0000
0.5484	300	-	1.0000
0.5850	320	-	1.0000
0.6216	340	-	1.0000
0.6581	360	-	1.0000
0.6947	380	-	1.0
0.7313	400	-	1.0000
0.7678	420	-	1.0
0.8044	440	-	1.0
0.8410	460	-	1.0000
0.8775	480	-	1.0
0.9141	500	0.0199	1.0000
0.9506	520	-	1.0
0.9872	540	-	1.0000
1.0	547	-	1.0000
1.0238	560	-	1.0000
1.0603	580	-	1.0000
1.0969	600	-	1.0000
1.1335	620	-	1.0000
1.1700	640	-	1.0
1.2066	660	-	1.0000
1.2431	680	-	1.0000
1.2797	700	-	1.0000
1.3163	720	-	1.0000
1.3528	740	-	1.0000
1.3894	760	-	1.0
1.4260	780	-	1.0
1.4625	800	-	1.0000
1.4991	820	-	1.0
1.5356	840	-	1.0000
1.5722	860	-	1.0000
1.6088	880	-	1.0
1.6453	900	-	1.0
1.6819	920	-	1.0
1.7185	940	-	1.0000
1.7550	960	-	1.0000
1.7916	980	-	1.0000
1.8282	1000	0.0012	1.0000
1.8647	1020	-	1.0
1.9013	1040	-	1.0
1.9378	1060	-	1.0
1.9744	1080	-	1.0
2.0	1094	-	1.0

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.4.1
• Transformers: 4.50.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.2
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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",
}