finetuned-vit-flowers

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1365
• Accuracy: 0.9653

Model description

Entrenamiento apoyado de: https://github.com/huggingface/notebooks/blob/main/examples/image_classification.ipynb

Intended uses & limitations

Proyecto final

Training and evaluation data

https://huggingface.co/datasets/DeadPixels/DPhi_Sprint_25_Flowers

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1236	0.99	36	0.1509	0.9730
0.1043	2.0	73	0.1235	0.9730
0.1077	2.96	108	0.1365	0.9653

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0