|
--- |
|
datasets: |
|
- cats_vs_dogs |
|
language: |
|
- en |
|
metrics: |
|
- accuracy |
|
library_name: transformers |
|
--- |
|
|
|
# Image Classification |
|
|
|
Classifies cat and dog images using a subset of the cats_vs_dogs dataset.<br> |
|
Uses PyTorch for the preprocessing, training, and inference. |
|
|
|
``` |
|
output_dir="cats_vs_dogs_model" |
|
remove_unused_columns=False |
|
evaluation_strategy="epoch" |
|
save_strategy="epoch" |
|
learning_rate=5e-5 |
|
per_device_train_batch_size=16 |
|
gradient_accumulation_steps=4 |
|
per_device_eval_batch_size=16 |
|
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 |
|
``` |
|
|
|
Note: during the training, I tried adjusting some of the above hyperparameters (like making the learning rate 0.1 as we have seen in class). But then the model could only classify cats and not dogs. |
|
|
