======  Tracking Machine Learning Experiments with ML Flow ======

How-to-Guide for setting up and connecting to ML Flow, tracking experiments and analysing them.

ML Flow Structure:

{{:cluster:usage_tips:mlflow_with_sleepvae.png?600|}}

==== Installing ML Flow ====

1) Install MLflow to lukaenv (conda activate your-custom-environment)

<code>
pip install mlflow
</code>

2) Run the following commands to setup a server on localhost: 

<code>
ARTIFACT_ROOT="file://home/your-username@sleep.ru.is/your-directory"
</code> 

Choose any port number e.g. 5001 (you cannot choose the same port number as other users on the cluster)

<code>
mlflow server --host 0.0.0.0 --port=5001 --default-artifact-root ARTIFACT_ROOT
</code>    

3) Run the MLflow version of the autoencoder (or setup your own experimental codes with whatever you want to track; see quick instructions "How to use MLflow" below)

4) Create a tunnel from the cluster localhost:5000 to your own localhost port to view and compare the experiments through your own browser:
    
Run the following code on your computers command prompt to map the localhost ports:
    
<code>
ssh -L localhost:5000:130.208.209.4:your-portnumber your-username@130.208.209.4
</code>
    
5) Open localhost:5000 through your browser

==== How to use ML Flow ====

1) Configure run in the beginning of training code:

<code>
import mlflow

mlflow_tracking_uri = 'http://localhost:your-port'
mlflow_experiment = 'test'

mlflow.set_tracking_uri(mlflow_tracking_uri)
mlflow.set_experiment(mlflow_experiment)
</code>

2) Add "with mlflow.start_run():" before your training code


3) Start the tracking

Track parameters by putting these lines in the code of your model:

<code>
mlflow.log_param('testparameter', testparam)
</code>

Track metrics (can change every epoch):

<code>
mlflow.log_metric('testparameter', testparam, step=epoch)
</code>

Track artefacts (figures, codes, etc. need to be saved before logging) using:
<code>
mlflow.log_artifact('/tmp/test_fig.png')
</code>