MFlux and MLflow cheat sheet

A collection of the most common MLflow and MFlux commands.

```python
import mlflow
import mflux_ai
import mlflow.keras
```
_keras_ functions are the same for _sklearn_

|Command|Help|
|---|---|
|**`with mlflow.start_run() as run:`<br>&nbsp;&nbsp;&nbsp;&nbsp;`mlflow.keras.log_model(model, artifact_path)`**|exits after indent|
|**`mlflow.search_runs(experiment_ids=None, filter_string='')`**|_Get a DataFrame of runs_|
|**`mlflow.set_experiment(experiment_name)`**|_Set as active experiment, create if not existing_|
|**`mlflow.experiments.list_experiments()`**|_Get a list of experiments_|
|**`mlflow.log_param(param_name, value)`**||
|**`mlflow.log_params(params)`**|params = {"k1": "v1", "k2": 2} **Similarly for log_metrics, set_tags and log_artifacts. See also mlflow.log_metrics**|
|**`mlflow.log_metric(metric_name, value, step=None)`**|"mse", 123, step=2|
|**`mlflow.set_tag(tag_name, value)`**|key is string, value will be stringified|
|**`mlflow.delete_tag(tag_name)`**|_irreversible_|
|**`mlflow.log_artifact(local_path, artifact_path=None)`**||
|**`mlflow.get_artifact_uri(artifact_path=None)`**||
|**`mlflow.keras.save_model(model, path)`**||
|**`mlflow.keras.log_model(model, artifact_path)`**||
|**`mlflow.keras.load_model(model_uri)`**||
|**`mflux_ai.put_dataset(dataset, "name.pkl")`**|_Upload dataset_|
|**`mflux_ai.get_dataset("name.pkl")`**|_Download dataset_|

This tutorial is open source, if you have suggestions for how this tutorial can be improved, you are welcome to propose a change.