resnet-50-finetuned-galaxy10-decals

This model is a fine-tuned version of microsoft/resnet-50 on the matthieulel/galaxy10_decals dataset. It achieves the following results on the evaluation set:

• Loss: 1.5789
• Accuracy: 0.4138
• Precision: 0.4493
• Recall: 0.4138
• F1: 0.3134

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• 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	Precision	Recall	F1
2.2417	0.9940	124	2.2265	0.2373	0.0697	0.2373	0.1077
2.1616	1.9960	249	2.1268	0.1950	0.1132	0.1950	0.0889
2.0459	2.9980	374	1.9901	0.2401	0.1029	0.2401	0.1290
1.9203	4.0	499	1.8571	0.3303	0.3116	0.3303	0.2052
1.8347	4.9940	623	1.7692	0.3613	0.2694	0.3613	0.2457
1.7628	5.9960	748	1.6926	0.3850	0.4172	0.3850	0.2758
1.723	6.9980	873	1.6342	0.3985	0.4428	0.3985	0.2922
1.71	8.0	998	1.6071	0.4104	0.4369	0.4104	0.3122
1.6948	8.9940	1122	1.5789	0.4138	0.4493	0.4138	0.3134
1.656	9.9399	1240	1.5805	0.4053	0.4176	0.4053	0.3034

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1