## Workflows
1. Update config.yaml
2. Update params.yaml
3. Update the entity
4. Update the configuration manager in src config
5. Update the components
6. Update the pipeline
7. Update the dvc.yaml
8. Update the app.py- Git clone the repository and Define template of the project
touch template.py
python3 template.py- define setup.py scripts, Create environment and install dependencies
conda create -n kidney-env python=3.9 -y
conda activate kidney-env
pip install -r requirements.txt-
define logger and utils
-
Data Ingestion
- define config/config.yaml and constant.yaml --> add 01_data_ingestion.ipynb
- entity --> configuration manager --> componenets --> pipeline --> run
dvc repro
- Base Model Section
- define config/config.yaml and params.yaml --> add 02_base_model.ipynb
- entity --> configuration manager --> componenets --> pipeline --> run
dvc repro
- Model Training Section
- define config/config.yaml and params.yaml --> add 03_model_training.ipynb (prepare call backs and train the model)
- entity --> configuration manager --> componenets --> pipeline --> run
dvc repro
- Model Evaluation
- define config/config.yaml and params.yaml --> add 04_model_evaluation.ipynb
- entity --> configuration manager --> componenets --> pipeline --> run
dvc repro - using mlflow and dagshub
export MLFLOW_TRACKING_URI=https://dagshub.com/fraidoon_omarzai/end-to-end-kidney-tumor.mlflow \
export MLFLOW_TRACKING_USERNAME=fraidoon_omarzai \
export MLFLOW_TRACKING_PASSWORD=bc25b16bd5206328d8899cf34377f26ad71d1420 \
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent