linker_v6

This model is a fine-tuned version of jarod0411/zinc10M_gpt2_SMILES_bpe_combined_step1 on the jarod0411/linker_v6 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3103
• Accuracy: 0.8995

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 1
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 128
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3747	1.0	43879	0.3539	0.8870
0.3495	2.0	87758	0.3322	0.8936
0.3387	3.0	131637	0.3240	0.8959
0.3322	4.0	175516	0.3194	0.8971
0.3279	5.0	219395	0.3164	0.8978
0.325	6.0	263274	0.3140	0.8985
0.3231	7.0	307153	0.3125	0.8989
0.3213	8.0	351032	0.3114	0.8992
0.3201	9.0	394911	0.3107	0.8994
0.3191	10.0	438790	0.3103	0.8995

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2