HassanCS/chemBERTa-tuned-on-ClinTox-3

SentenceTransformer based on DeepChem/ChemBERTa-77M-MLM

This is a sentence-transformers model finetuned from DeepChem/ChemBERTa-77M-MLM. 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: DeepChem/ChemBERTa-77M-MLM
• 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: RobertaModel 
  (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})
)

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("HassanCS/chemBERTa-tuned-on-ClinTox-3")
# Run inference
sentences = [
    'CC#CCn1c(N2CCCC([NH3+])C2)nc2c1c(=O)n(Cc1nc(C)c3ccccc3n1)c(=O)n2C',
    'CC12CCC(=O)C=C1CCC1C2C(O)CC2(C)C1CCC2(O)C(=O)COC(=O)CCC1CCCC1',
    'CC(Cc1ccc(O)c(O)c1)C(C)Cc1ccc(O)c(O)c1',
]
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]

Evaluation

Metrics

Binary Classification

• Dataset: all-dev
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.9066
cosine_accuracy_threshold	0.5665
cosine_f1	0.951
cosine_f1_threshold	0.5665
cosine_precision	0.9068
cosine_recall	0.9998
cosine_ap	0.9523

Training Details

Training Dataset

Unnamed Dataset

• Size: 10,000 training samples
• Columns: smiles1, smiles2, and label
• Approximate statistics based on the first 1000 samples:

  	smiles1	smiles2	label
  type	string	string	int
  details	min: 3 tokens mean: 40.69 tokens max: 221 tokens	min: 4 tokens mean: 51.43 tokens max: 221 tokens	0: ~14.90% 1: ~85.10%

• Samples:

  smiles1	smiles2	label
  Cn1c(=O)c2c(ncn2C)n(C)c1=O	Cc1cc2c(s1)=Nc1ccccc1NC=2N1CCNH+CC1	1
  Oc1ccc(OCc2ccccc2)cc1	Oc1ccc(CCCC[NH2+]CC(O)c2ccc(O)c(O)c2)cc1	1
  OCC(S)CS	CC12CCC(=O)C=C1CCC1C2C(O)CC2(C)C1CCC2(O)C(=O)CO	0

• Loss: ContrastiveLoss with these parameters:

  {
      "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
      "margin": 0.5,
      "size_average": true
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 5,000 evaluation samples
• Columns: smiles1, smiles2, and label
• Approximate statistics based on the first 1000 samples:

  	smiles1	smiles2	label
  type	string	string	int
  details	min: 18 tokens mean: 56.96 tokens max: 209 tokens	min: 18 tokens mean: 61.21 tokens max: 244 tokens	0: ~10.00% 1: ~90.00%

• Samples:

  smiles1	smiles2	label
  CC(=CC(=O)OCCCCCCCCC(=O)[O-])CC1OCC(CC2OC2C(C)C(C)O)C(O)C1O	CC(C=CC(C)C(C)(C)O)C1CCC2C(=CC=C3CC(O)CC(O)C3)CCCC21C	1
  C=C1c2cccc([O-])c2C(=O)C2=C([O-])C3(O)C(=O)C(C(N)=O)=C([O-])C(NH+C)C3C(O)C12	CC(c1ncncc1F)C(O)(Cn1cncn1)c1ccc(F)cc1F	1
  CC(C)CC1C(=O)N2CCCC2C2(O)OC(NC(=O)C3C=C4c5cccc6[nH]c(Br)c(c56)CC4NH+C3)(C(C)C)C(=O)N12	CNH+CCC=C1c2ccccc2Sc2ccc(Cl)cc21	1

• Loss: ContrastiveLoss with these parameters:

  {
      "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
      "margin": 0.5,
      "size_average": true
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• num_train_epochs: 5
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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: 2e-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: 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}
• 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	all-dev_cosine_ap
0.8	500	0.0264	0.0112	0.9213
1.6	1000	0.0152	0.0122	0.9362
2.4	1500	0.0134	0.0128	0.9463
3.2	2000	0.0112	0.0134	0.9502
4.0	2500	0.01	0.0125	0.9513
4.8	3000	0.0097	0.0132	0.9523

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.3.1
• Transformers: 4.47.1
• PyTorch: 2.5.1+cu124
• 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",
}

ContrastiveLoss

@inproceedings{hadsell2006dimensionality,
    author={Hadsell, R. and Chopra, S. and LeCun, Y.},
    booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
    title={Dimensionality Reduction by Learning an Invariant Mapping},
    year={2006},
    volume={2},
    number={},
    pages={1735-1742},
    doi={10.1109/CVPR.2006.100}
}