Are you sure you want to delete this access key?
To run:
First, you need to run
conda env create -f conda.yaml
to make the environment and
conda activate aitest
to get into it. Then you need to prepare the data sample by running:
python 0-preparation.py
It downloads the data into data directory. Then you should run:
mlflow run . --no-conda
This will run 1-train.py
with the default parameters to train a model.
In order to run the file with custom parameters, you can run it like:
mlflow run . \
-P run_name='test' \
-P batch_size=64 \
-P epochs=20 \
-P aug_rot=45 \
-P aug_w=0.05 \
-P aug_h=0.05 \
-P aug_zoom=0.05 \
-P model_path='../models' \
--no-conda
Once the code is finished executing, you can view the run's metrics, parameters, and details by running the command
mlflow ui
and navigating to http://localhost:5000.
After all and to predict with the trained model, you can run the User Interface by:
streamlit run 2-predict.py
Press p or to see the previous file or, n or to see the next file
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?