resnet-50_finetuned

This model is a fine-tuned version of microsoft/resnet-50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7209
• Precision: 0.3702
• Recall: 0.5
• F1: 0.4254
• Accuracy: 0.7404

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	46	0.6599	0.3702	0.5	0.4254	0.7404
No log	2.0	92	0.6725	0.3702	0.5	0.4254	0.7404
No log	3.0	138	nan	0.8714	0.5062	0.4384	0.7436
No log	4.0	184	nan	0.8714	0.5062	0.4384	0.7436
No log	5.0	230	nan	0.8714	0.5062	0.4384	0.7436
No log	6.0	276	nan	0.8714	0.5062	0.4384	0.7436
No log	7.0	322	nan	0.8714	0.5062	0.4384	0.7436
No log	8.0	368	nan	0.8714	0.5062	0.4384	0.7436
No log	9.0	414	nan	0.8714	0.5062	0.4384	0.7436
No log	10.0	460	0.7209	0.3702	0.5	0.4254	0.7404

Framework versions

• Transformers 4.22.1
• Pytorch 1.12.1+cu113
• Datasets 2.5.1
• Tokenizers 0.12.1