Image Classification Model Development

I have a collection of images that needs to be turned into a reliable predictive system—specifically, an image-classification model. The work begins with exploring and cleaning the dataset, then moves through model selection, training, hyper-parameter tuning, and evaluation. I am open to popular frameworks such as TensorFlow, Keras, or PyTorch as long as you can justify the choice and document the pipeline clearly. Deliverables • Reproducible code (Python notebooks or scripts) • Trained model files and clear instructions for inference on new images • A concise report covering data preparation steps, model architecture, training metrics, and final accuracy/F1-score • Optional: a lightweight API or CLI for running the classifier locally Acceptance criteria • Consistent image classification performance that meets or exceeds the metric targets agreed upon at project kick-off • Code runs end-to-end on my machine with only the documented dependencies • Well-commented source so future adjustments are straightforward If you have previous examples of predictive analysis on visual data, that will speed up the selection process.