Image classification of vegetable diseases

Beginners
1

Hello,
I am trying to create a model for the classification of vegetable diseases. On colab I transferred the dataset with the photos of the leaves of the vegetables from kaggle. Then I copied and pasted from a colab notebook I found in the hugging face tutorial on image classification and everything works. I could compile the model and transfer it to hugging face but it would take 9 hours! Of the parameters, however, I’m not sure whether to leave the ones I found in the tutorial, so I ask you if you could possibly change them yourself in order to reduce the time required to fill in the model. Thank you

These would be the parameters to be changed if necessary:

model_checkpoint = "microsoft/swin-tiny-patch4-window7-224" 
batch_size = 32

args = TrainingArguments(
    f"{model_name}-finetuned-plantdisease",
    remove_unused_columns=False,
    evaluation_strategy = "epoch",
    save_strategy = "epoch",
    learning_rate=5e-5,
    per_device_train_batch_size=batch_size,
    gradient_accumulation_steps=4,
    per_device_eval_batch_size=batch_size,
    num_train_epochs=3,
    warmup_ratio=0.1,
    logging_steps=10,
    load_best_model_at_end=True,
    metric_for_best_model="accuracy",
    push_to_hub=True,
)