convnext-large-224-22k-1k-bottomCleanedData

This model is a fine-tuned version of facebook/convnext-large-224-22k-1k on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.0067
Accuracy: 0.9977

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 7
total_train_batch_size: 56
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.01
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2003	1.0	141	0.0628	0.9807
0.1568	2.0	283	0.0173	0.9943
0.1499	2.99	424	0.0211	0.9898
0.1189	4.0	566	0.0140	0.9955
0.084	4.99	707	0.0105	0.9955
0.0797	6.0	849	0.0093	0.9966
0.0781	7.0	991	0.0157	0.9921
0.1075	8.0	1132	0.0079	0.9943
0.0718	9.0	1274	0.0075	0.9966
0.0592	9.96	1410	0.0067	0.9977

Framework versions

Transformers 4.28.1
Pytorch 2.0.0+cu118
Datasets 2.12.0
Tokenizers 0.13.3

作者:

hammami mohamed soulaime,

数据集大小:

748.79 MB