cvmil/resnet-50_augmented-v2_tl

resnet-50_rice-leaf-disease-augmented-v2_tl

This model is a fine-tuned version of microsoft/resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3083
• Accuracy: 0.5952

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: 0.0003
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0633	1.0	63	2.0143	0.3452
1.9625	2.0	126	1.8719	0.5060
1.8119	3.0	189	1.7332	0.5
1.6826	4.0	252	1.6271	0.5268
1.5879	5.0	315	1.5436	0.5595
1.516	6.0	378	1.4871	0.5536
1.4572	7.0	441	1.4566	0.5655
1.4104	8.0	504	1.4224	0.5685
1.3734	9.0	567	1.4033	0.5685
1.3414	10.0	630	1.3735	0.5952
1.3186	11.0	693	1.3579	0.5714
1.2972	12.0	756	1.3402	0.5923
1.2862	13.0	819	1.3342	0.5893
1.2716	14.0	882	1.3271	0.5863
1.2632	15.0	945	1.3210	0.6042
1.2546	16.0	1008	1.3146	0.5923
1.2485	17.0	1071	1.3061	0.6012
1.25	18.0	1134	1.3090	0.5923
1.2457	19.0	1197	1.3106	0.6042
1.2466	20.0	1260	1.3083	0.5952

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0