esm2_t33_650M_UR50D-uniprot-7class

This model is a fine-tuned version of facebook/esm2_t33_650M_UR50D on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4652
• Accuracy: 0.8605

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.9846	4	1.0980	0.6919
No log	1.9692	8	0.9456	0.6919
No log	2.9538	12	0.8277	0.6919
No log	3.9385	16	0.7322	0.7267
No log	4.9231	20	0.6554	0.7616
No log	5.9077	24	0.5913	0.8198
No log	6.8923	28	0.5352	0.8314
No log	7.8769	32	0.4964	0.8314
No log	8.8615	36	0.4736	0.8547
No log	9.8462	40	0.4652	0.8605

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1