interpro_bert3

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4142

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: 0.0001
• train_batch_size: 200
• eval_batch_size: 128
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 1600
• total_eval_batch_size: 1024
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.2045	1.0	18425	1.1198
0.8936	2.0	36850	0.8563
0.7736	3.0	55275	0.7475
0.6946	4.0	73700	0.6877
0.8083	5.0	92125	0.7509
0.677	6.0	110550	0.6578
0.778	7.0	128975	0.7306
0.6017	8.0	147400	0.5994
0.5646	9.0	165825	0.5704
0.5352	10.0	184250	0.5479
0.532	11.0	202675	0.5496
0.495	12.0	221100	0.5198
0.4714	13.0	239525	0.4971
0.4497	14.0	257950	0.4797
0.4312	15.0	276375	0.4670
0.4131	16.0	294800	0.4494
0.4001	17.0	313225	0.4411
0.3828	18.0	331650	0.4316
0.3665	19.0	350075	0.4201
0.3592	20.0	368500	0.4142

Framework versions

• Transformers 4.39.2
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2