zinc10M_gpt2-medium_SMILES_step1

This model is a fine-tuned version of gpt2-medium on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5598
• Accuracy: 0.8151

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 64
• total_eval_batch_size: 64
• 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
0.635	1.0	10635	0.6169	0.8007
0.6073	2.0	21270	0.5937	0.8066
0.5932	3.0	31905	0.5828	0.8093
0.5843	4.0	42540	0.5754	0.8112
0.5782	5.0	53175	0.5704	0.8124
0.5729	6.0	63810	0.5666	0.8134
0.5691	7.0	74445	0.5638	0.8141
0.5666	8.0	85080	0.5620	0.8145
0.5644	9.0	95715	0.5606	0.8149
0.5629	10.0	106350	0.5598	0.8151

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.1+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0