- Update config.yaml
- Update secrets.yaml [optional] # if you have any then only
- Update params.yaml
- Update the entity
- Update the configuration manager in src config.
- Update the components
- Update the pipelines
- Test run pipeline stage
- Run tox for testing your package
- Update the dvc.yaml
- run "dvc repro" for running all the stages in pipeline
STEP 1: Set the env variable | Get it from Dagshub -> remote tab -> mlflow tab
MLFLOW_TRACKING_URI=https://dagshub.com/AmitSingh-DataScientist/Deep_CNN_Classifier.mlflow
MLFLOW_TRACKING_USERNAME=AmitSingh-DataScientist
MLFLOW_TRACKING_PASSWORD=<> \
STEP 2: Install mlflow
STPE 3: Set remote URI
STEP 4: Use context manager to mlflow to start run and then log metrics, params, and model
FROM python:3.8-slim # download pyton image with bare minimum requirement of os to run
WORKDIR /app # create app folder inside root directory
COPY . . # copy all the files in app folder
RUN pip install -r requirements.txt
CMD ["streamlit", "run", "app.py"]
Run this in terminal
export DOCKER_BUILDKIT=0 export COMPOSE_DOCKER_CLI_BUILD=0
This will fix that error and you will be able to build your docker image.
https://raw.githubusercontent.com/c17hawke/raw_data/main/sample_data.zip