An image classification model from data collection, cleaning, model training, deployment and API integration.
The model can classify 25 different types of wildflowers
The types of wildflowers are:
- Alpine Aster,
- Alpine Forget-me-not,
- Beach Evening Primrose,
- Beach Morning Glory,
- Black-eyed Susan,
- Bluebell,
- Columbine,
- Daisy,
- Desert Marigold,
- Desert Sunflower,
- Indian Paintbrush,
- Lady's Slipper Orchid,
- Lupine,
- Ocotillo,
- Pitcher Plant,
- Prairie Clover,
- Prairie Phlox,
- Prickly Pear,
- Queen Anne's Lace,
- Seaside Goldenrod,
- Sunflower,
- Swamp Milkweed,
- Trillium,
- Water Lily,
- Wild Bergamot
Data Collection: Downloaded from DuckDuckGo using term name
DataLoader: Used fastai DataBlock API to set up the DataLoader.
Data Augmentation: fastai provides default data augmentation which operates in GPU.
Details can be found in notebooks/Wildflower_Recognizer_data_prep.ipynb
Training: Fine-tuned a resnet34 model for 3 epochs (3 times) and got upto ~96% accuracy.
Data Cleaning: This part took the highest time. Since I collected data from browser, there were many noises. Also, there were images that contained. I cleaned and updated data using fastai ImageClassifierCleaner. I cleaned the data each time after training or finetuning of the model.
I deployed the model to HuggingFace Spaces Gradio App. The implementation can be found in app folder or here.
The deployed model API is integrated here in GitHub Pages Website. Implementation and other details can be found in docs folder.