Saideepthi55/sentencetransformer_ftmodel_on_chemical_dataset

MPNet base trained on sxc_med_llm_chemical_gen

This is a sentence-transformers model finetuned from microsoft/mpnet-base on the sxc_med_llm_chemical_gen dataset. 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: microsoft/mpnet-base
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sxc_med_llm_chemical_gen
• Language: en
• License: apache-2.0

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: MPNetModel 
  (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("Saideepthi55/sentencetransformer_ftmodel_on_chemical_dataset")
# Run inference
sentences = [
    'With a molecule represented by the SMILES string CNNNCC(=O)N[C@H](C)C[C@@H](C)NCc1ccc2c(c1)CCC2, propose adjustments that can increase its logP value while keeping the output molecule structurally related to the input molecule.',
    'Given a molecule expressed in SMILES string, help me optimize it according to my requirements.',
    'In line with your criteria, I\'ve optimized the molecule and present it as "C[C@H](C[C@@H](C)NC(=O)COC(C)(C)C)NCc1ccc2c(c1)CCC2".',
]
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

• Dataset: sxc_med_llm_chemical_gen
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9838
dot_accuracy	0.0162
manhattan_accuracy	0.9827
euclidean_accuracy	0.9836
max_accuracy	0.9838

Training Details

Training Dataset

sxc_med_llm_chemical_gen

• Dataset: sxc_med_llm_chemical_gen at ae28874
• Size: 117,502 training samples
• Columns: input, instruction, and output
• Approximate statistics based on the first 1000 samples:

  	input	instruction	output
  type	string	string	string
  details	min: 2 tokens mean: 50.03 tokens max: 156 tokens	min: 6 tokens mean: 16.19 tokens max: 69 tokens	min: 3 tokens mean: 67.94 tokens max: 321 tokens

• Samples:

  input	instruction	output
  	List the countries of the Middle East	Egypt, Israel, Iran, Iraq, Jordan, Kuwait, Lebanon, Qatar, Saudi Arabia, Syria, Turkey, The United Arab Emirates, Oman, and Yemen.
  Support me in refining the molecule CCC(C)=C(CC)c1ccc(Nc2nc3nonc3nc2N2CCC(Oc3ccc(Cl)cc3)CC2)cc1 to have a more drug-like nature, ensuring it remains close to the input molecule. Please inform me of the QED value of the optimized molecule.	Help me adapt a molecular structure denoted in SMILES string based on my preferences.	I've optimized the molecule according to your requirements, yielding the structure "COC(=O)c1ccc(Nc2nc3nonc3nc2N2CCC(Oc3ccc(Cl)cc3)CC2)cc1" and an estimated QED value of 0.4.
  	Create a social media strategy to attract followers on Twitter.	The social media strategy for Twitter should involve Tweeting on a regular basis, with content that is engaging and relevant to the target audience. This can include retweetable quotes, links to videos and articles, sharing photos/images, polls, and so on. It is also beneficial for the information to include hashtags relevant to the content and any relevant topics of the industry. It is also important to leverage Twitter influencers and accounts, as well as engaging with potential followers regularly and DM them when necessary. Additionally, running campaigns, contests and giveaways to gain more followers should also be part of the strategy. Finally, hashtag research should be conducted to increase the reach of the tweets.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

sxc_med_llm_chemical_gen

• Dataset: sxc_med_llm_chemical_gen at ae28874
• Size: 4,500 evaluation samples
• Columns: input, instruction, and output
• Approximate statistics based on the first 1000 samples:

  	input	instruction	output
  type	string	string	string
  details	min: 2 tokens mean: 68.49 tokens max: 165 tokens	min: 6 tokens mean: 16.89 tokens max: 36 tokens	min: 3 tokens mean: 69.16 tokens max: 205 tokens

• Samples:

  input	instruction	output
  With a molecule represented by the SMILES string COc1ccc(CN(C(=O)Nc2ccc(C@Hc3ccccc3)c(Cl)c2)C2CC2)cc1, propose adjustments that can decrease its blood-brain barrier permeability (BBBP) value by at least 0.1 compared to the pre-optimized value while keeping the output molecule structurally related to the input molecule. Please inform me of the BBBP value of the optimized molecule.	Help me adapt a molecular structure denoted in SMILES string based on my preferences.	I've optimized the molecule according to your requirements, yielding the structure "CC(C)c1cccc(Oc2ccc(CN(C(=O)Nc3ccc(C@Hc4ccccc4)c(Cl)c3)C3CC3)cc2)c1" and an estimated BBBP value of 0.34.
  Which modifications can be applied to the molecule Cc1nn(-c2cc3cc(F)ccc3s2)c2ncc(C(=O)N3CCCC3c3nnc4n3CCCCC4)cc12 to increase its water solubility value while ensuring it remains similar to the original molecule?	Aid me in refining a molecular structure written in SMILES notation based on my criteria.	I've optimized the molecule according to your requirements, resulting in the following structure: "Cc1nn(C)c2ncc(C(=O)N3CCCC3c3nnc4n3CCCCC4)cc12".
  	Describe the benefits of having a mentor.	Having a mentor can provide tremendous benefits, including guidance, advice, support and connection to a network of contacts. Mentors can offer valuable feedback, help to create goals and strategies, and provide the accountability necessary to carry out the desired goals. They can also provide a fresh perspective which can help to create new ideas and solutions.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

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: 1
• warmup_ratio: 0.1
• fp16: True

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: 1
• 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: 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
• eval_use_gather_object: False
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	sxc_med_llm_chemical_gen_max_accuracy
0	0	-	-	0.7076
0.0136	100	4.1705	3.6314	0.7469
0.0272	200	3.0088	2.4771	0.8191
0.0408	300	2.3803	2.3765	0.8716
0.0545	400	2.2281	2.3122	0.9316
0.0681	500	2.1647	2.2997	0.9436
0.0817	600	2.1289	2.2663	0.9511
0.0953	700	2.0663	2.2601	0.9629
0.1089	800	2.065	2.2500	0.9687
0.1225	900	2.0399	2.2595	0.9693
0.1362	1000	1.9939	2.2375	0.9707
0.1498	1100	1.9858	2.2220	0.9684
0.1634	1200	2.0069	2.2265	0.9758
0.1770	1300	1.9591	2.2073	0.9702
0.1906	1400	1.9288	2.2078	0.976
0.2042	1500	1.9466	2.2036	0.9758
0.2179	1600	1.9175	2.2086	0.9764
0.2315	1700	1.8835	2.2329	0.9796
0.2451	1800	1.9134	2.2003	0.9796
0.2587	1900	1.8809	2.2003	0.9811
0.2723	2000	1.9263	2.2039	0.9824
0.2859	2100	1.9101	2.2084	0.9804
0.2996	2200	1.8846	2.2052	0.9831
0.3132	2300	1.8842	2.1903	0.9818
0.3268	2400	1.8945	2.1984	0.9807
0.3404	2500	1.9217	2.1859	0.9829
0.3540	2600	1.8704	2.1995	0.982
0.3676	2700	1.889	2.2038	0.9822
0.3813	2800	1.875	2.2079	0.9829
0.3949	2900	1.8792	2.1975	0.9833
0.4085	3000	1.882	2.1895	0.9796
0.4221	3100	1.8886	2.2115	0.9831
0.4357	3200	1.8629	2.2040	0.9838
0.4493	3300	1.8647	2.1973	0.9836
0.4630	3400	1.8888	2.1961	0.9838
0.4766	3500	1.8692	2.2027	0.9829
0.4902	3600	1.8846	2.1954	0.9838
0.5038	3700	1.8482	2.1888	0.9822
0.5174	3800	1.8527	2.1873	0.9824
0.5310	3900	1.8378	2.1940	0.9811
0.5447	4000	1.8679	2.2008	0.9833
0.5583	4100	1.8421	2.1845	0.9842
0.5719	4200	1.8325	2.1948	0.9847
0.5855	4300	1.8675	2.1750	0.9836
0.5991	4400	1.8483	2.1828	0.9831
0.6127	4500	1.854	2.1886	0.9831
0.6264	4600	1.827	2.1876	0.9824
0.6400	4700	1.8863	2.1849	0.9836
0.6536	4800	1.8919	2.1816	0.984
0.6672	4900	1.8211	2.1830	0.9847
0.6808	5000	1.8345	2.1847	0.9842
0.6944	5100	1.8685	2.1855	0.9853
0.7081	5200	1.85	2.1864	0.9844
0.7217	5300	1.8222	2.1875	0.9842
0.7353	5400	1.8179	2.1923	0.9844
0.7489	5500	1.7992	2.1909	0.9851
0.7625	5600	1.8495	2.1811	0.9847
0.7761	5700	1.808	2.1763	0.9842
0.7898	5800	1.8293	2.1861	0.9849
0.8034	5900	1.8184	2.1845	0.9851
0.8170	6000	1.8256	2.1956	0.9849
0.8306	6100	1.7904	2.1916	0.9842
0.8442	6200	1.8028	2.1918	0.9847
0.8578	6300	1.8316	2.1917	0.9856
0.8715	6400	1.7951	2.1929	0.9851
0.8851	6500	1.8175	2.1866	0.9847
0.8987	6600	1.8071	2.1899	0.9853
0.9123	6700	1.8632	2.1905	0.9844
0.9259	6800	1.8441	2.1885	0.984
0.9395	6900	1.8243	2.1865	0.9836
0.9532	7000	1.8055	2.1852	0.9842
0.9668	7100	1.8227	2.1843	0.984
0.9804	7200	1.8287	2.1831	0.984
0.9940	7300	1.8379	2.1838	0.9838

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.2.0
• Transformers: 4.44.2
• PyTorch: 2.4.1+cu121
• Accelerate: 0.34.2
• Datasets: 3.0.1
• Tokenizers: 0.19.1

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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}