How vikashishere/yt-mlops-proj1 works
This is an end-to-end vehicle-insurance MLOps application. Training starts from the web app or training entry point, ingests MongoDB data, validates and transforms it, trains and evaluates a model, then pushes an accepted model to S3. Prediction converts submitted vehicle fields into a dataframe and loads the model from S3. The graph preserves the documented web/API and AWS capabilities; several component links rely on the SOURCE INDEX or README because their implementations were not excerpted.
Main entry points: training_pipeline.py Training orchestrator · s3_estimator.py S3 model estimator · app.py Web/API app · model_trainer.py Model trainer
Components 21
Application entry
- Web/API app
app.py
Training workflow
- Training orchestrator
src/pipline/training_pipeline.py - Data ingestion
src/components/data_ingestion.py - Data validation
src/components/data_validation.py - Data transformation
src/components/data_transformation.py - Model trainer
src/components/model_trainer.py - Model evaluation
src/components/model_evaluation.py - Model pusher
src/components/model_pusher.py
Model serving
- Prediction pipeline
src/pipline/prediction_pipeline.py - Vehicle input
src/pipline/prediction_pipeline.py - S3 model estimator
src/entity/s3_estimator.py - Model estimator
src/entity/estimator.py
Data and cloud integrations
- Mongo data access
src/data_access/proj1_data.py - Mongo connection
src/configuration/mongo_db_connection.py - MongoDB
- S3 storage service
src/cloud_storage/aws_storage.py - AWS S3
Pipeline contracts
- Pipeline artifacts
src/entity/artifact_entity.py - Pipeline configuration
src/entity/config_entity.py - File and YAML utilities
src/utils/main_utils.py
Other components
- User
Connections 25
- User → Web/API app: submits data
README.md - Web/API app → Training orchestrator: starts training
app.py - Web/API app → Prediction pipeline: routes prediction
app.py - Training orchestrator → Data ingestion: runs ingestion
src/pipline/training_pipeline.py - Training orchestrator → Data validation: runs validation
src/pipline/training_pipeline.py - Training orchestrator → Data transformation: runs transformation
src/pipline/training_pipeline.py - Training orchestrator → Model trainer: runs training
src/pipline/training_pipeline.py - Training orchestrator → Model evaluation: runs evaluation
src/pipline/training_pipeline.py - Training orchestrator → Model pusher: runs pushing
src/pipline/training_pipeline.py - Data ingestion → Mongo data access: fetches records
src/components/data_ingestion.py - Mongo data access → Mongo connection: uses connection
src/data_access/proj1_data.py - Mongo connection → MongoDB: connects to
README.md - Data transformation → File and YAML utilities: uses utilities
src/components/data_transformation.py - Model trainer → File and YAML utilities: uses utilities
src/components/model_trainer.py - Model trainer → Model estimator: builds model
src/components/model_trainer.py - Model evaluation → S3 model estimator: compares model
src/components/model_evaluation.py - Model pusher → S3 storage service: uploads model
src/components/model_pusher.py - Model pusher → S3 model estimator: uses estimator
src/components/model_pusher.py - S3 model estimator → S3 storage service: loads or saves
src/entity/s3_estimator.py - S3 storage service → AWS S3: reads and writes
README.md - Prediction pipeline → Vehicle input: accepts vehicle fields
README.md - Prediction pipeline → S3 model estimator: requests prediction
src/pipline/prediction_pipeline.py - S3 model estimator → Model estimator: predicts with
src/entity/s3_estimator.py - Training orchestrator → Pipeline artifacts: passes stage results
src/pipline/training_pipeline.py - Training orchestrator → Pipeline configuration: configures stages
src/pipline/training_pipeline.py