How jarrydstringy/transfer-learning-flower-classifier works
This repository is a script-based flower image classifier using ImageNet transfer learning. Its principal workflow prepares image data, adapts a frozen pretrained model to five flower classes, trains it, and evaluates or plots results; additional tasks compare optimizer settings. The sampled excerpts show the script’s functions and some main-program calls, but omit portions of the runner and task implementations, so unsampled call relationships are not inferred.
Main entry points: FlowerClassifier.py Model training
Components 11
Image data
- Task runner
FlowerClassifier.py - Dataset setup
FlowerClassifier.py - Image preparation
FlowerClassifier.py
Transfer model
- ImageNet base
FlowerClassifier.py - Five-class model
FlowerClassifier.py
Training and analysis
- Model training
FlowerClassifier.py - Model evaluation
FlowerClassifier.py - Results plots
FlowerClassifier.py - Optimizer experiments
FlowerClassifier.py
Other components
- Script runner
- Flower images
Connections 10
- Script runner → Task runner: runs
FlowerClassifier.py - Task runner → Dataset setup: calls
FlowerClassifier.py - Flower images → Image preparation: image files
FlowerClassifier.py - Image preparation → ImageNet base: prepares inputs
FlowerClassifier.py - ImageNet base → Five-class model: provides base
FlowerClassifier.py - Image preparation → Model training: provides splits
FlowerClassifier.py - Five-class model → Model training: trains
FlowerClassifier.py - Model training → Model evaluation: evaluates
FlowerClassifier.py - Model training → Results plots: plots history
FlowerClassifier.py - Model training → Optimizer experiments: compares settings
FlowerClassifier.py