internal.swin-base-food101-int8-structured38.63 – AI Model by yujiepan | AlphaNeural AI

You can deploy this model and start earning money today!

yujiepan/
internal.swin-base-food101-int8-structured38.63

like

0

image-classification

generated_from_trainer

endpoints_compatible

Views

No views yet

swin-food101-jpqd-1to2r1.5-epo7-finetuned-student

This model is a fine-tuned version of skylord/swin-finetuned-food101 on the food101 dataset. It achieves the following results on the evaluation set:

Loss: 0.2658
Accuracy: 0.9124

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: 5e-05
train_batch_size: 16
eval_batch_size: 128
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 7.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2977	0.42	500	0.1949	0.9112
0.3183	0.84	1000	0.1867	0.9144
99.9552	1.27	1500	99.4882	0.7577
162.4195	1.69	2000	162.7763	0.3373
1.2272	2.11	2500	0.7333	0.8564
1.0236	2.54	3000	0.5016	0.8823
0.6472	2.96	3500	0.4337	0.8908
0.52	3.38	4000	0.3927	0.8974
0.6075	3.8	4500	0.3506	0.9011
0.5348	4.23	5000	0.3425	0.9006
0.444	4.65	5500	0.3268	0.9044
0.5787	5.07	6000	0.3020	0.9078
0.3995	5.49	6500	0.2932	0.9095
0.414	5.92	7000	0.2806	0.9104
0.4386	6.34	7500	0.2738	0.9112
0.452	6.76	8000	0.2673	0.9127

Framework versions

Transformers 4.26.0
Pytorch 1.13.1+cu116
Datasets 2.8.0
Tokenizers 0.13.2