How kallappa2005/mlops_kidney_disease_cnn works
This repository is a kidney CT image classifier with a training workflow and Flask inference service. Training runs through ingestion, base-model preparation, training, then evaluation; serving accepts an uploaded image, decodes it, and runs the saved model for a Normal/Tumor result. Configuration and shared utilities support these paths, while DVC versioning and MLflow experiment tracking are documented capabilities. Edges reflect the excerpts, source-index references, or README; the internal implementations of several components were not sampled.
Main entry points: configuration.py Configuration manager · main.py Training entry · stage_01_data_ingestion.py Data ingestion stage · stage_02_prepare_base_model.py Base preparation stage
Components 24
Training workflow
- Training entry
main.py - Data ingestion stage
src/cnnClassifier/pipeline/stage_01_data_ingestion.py - Base preparation stage
src/cnnClassifier/pipeline/stage_02_prepare_base_model.py - Training stage
src/cnnClassifier/pipeline/stage_03_model_training.py - Evaluation stage
src/cnnClassifier/pipeline/stage_04_model_evaluation.py
Model lifecycle
- Dataset ingestion
src/cnnClassifier/components/data_ingestion.py - Base model preparation
src/cnnClassifier/components/prepare_base_model.py - CNN training
src/cnnClassifier/components/model_training.py - Model evaluation
src/cnnClassifier/components/model_evaluation_mlflow.py - CT image dataset
- Saved model
model/model.h5
Prediction service
- Flask routes
app.py - Upload UI
templates/index.html - Image decoding
src/cnnClassifier/utils/common.py - Prediction pipeline
src/cnnClassifier/pipeline/prediction.py
Configuration and tracking
- Configuration manager
src/cnnClassifier/config/configuration.py - Stage config entities
src/cnnClassifier/entity/config_entity.py - Shared utilities
src/cnnClassifier/utils/common.py - Project settings
config/config.yaml - Training parameters
params.yaml - DVC versioning
dvc.yaml - MLflow tracking
Other components
- Pipeline operator
- Image submitter
Connections 28
- Pipeline operator → Training entry: runs
README.md - Training entry → Data ingestion stage: runs
main.py - Training entry → Base preparation stage: runs
main.py - Training entry → Training stage: runs
main.py - Training entry → Evaluation stage: runs
main.py - Data ingestion stage → Configuration manager: gets config
src/cnnClassifier/pipeline/stage_01_data_ingestion.py - Data ingestion stage → Dataset ingestion: invokes
src/cnnClassifier/pipeline/stage_01_data_ingestion.py - Base preparation stage → Configuration manager: gets config
src/cnnClassifier/pipeline/stage_02_prepare_base_model.py - Base preparation stage → Base model preparation: invokes
src/cnnClassifier/pipeline/stage_02_prepare_base_model.py - Training stage → Configuration manager: gets config
src/cnnClassifier/pipeline/stage_03_model_training.py - Training stage → CNN training: invokes
src/cnnClassifier/pipeline/stage_03_model_training.py - Evaluation stage → Configuration manager: gets config
src/cnnClassifier/pipeline/stage_04_model_evaluation.py - Evaluation stage → Model evaluation: invokes
src/cnnClassifier/pipeline/stage_04_model_evaluation.py - Configuration manager → Stage config entities: builds
src/cnnClassifier/config/configuration.py - Configuration manager → Shared utilities: reads settings
src/cnnClassifier/config/configuration.py - Configuration manager → Project settings: loads
README.md - Configuration manager → Training parameters: loads
README.md - Dataset ingestion → CT image dataset: downloads and extracts
src/cnnClassifier/components/data_ingestion.py - CNN training → CT image dataset: trains on
- Model evaluation → Saved model: loads
src/cnnClassifier/components/model_evaluation_mlflow.py - Model evaluation → MLflow tracking: logs metrics
src/cnnClassifier/components/model_evaluation_mlflow.py - DVC versioning → CT image dataset: versions data
README.md - DVC versioning → Saved model: versions outputs
README.md - Image submitter → Upload UI: opens
README.md - Upload UI → Flask routes: submits image
README.md - Flask routes → Image decoding: decodes upload
app.py - Flask routes → Prediction pipeline: requests prediction
app.py - Prediction pipeline → Saved model: loads model
src/cnnClassifier/pipeline/prediction.py