In this repository, we will do a hands-on on:
- Tracking ML experiments with MLflow tracking
- ML model management with MLflow Model Registry
We will use the Cycle Power Plant dataset
in the different practical works, it is already in the data folder.
The dataset contains 9568 data points collected from a Combined Cycle Power Plant over 6 years (2006-2011). To goal is to predict the net hourly electrical energy output (EP) of the plant.
Features consist of hourly average ambient variables:
- T: Temperature
- AP: Ambient Pressure
- RH: Relative Humidity
- V: Exhaust Vacuum
- EP: Net hourly electrical energy output
The averages are taken from various sensors located around the plant that record the ambient variables every second.
The dataset in the data folder is a concatenation of all the data in the raw excel file.
This was done with the following code:
# install the version 1.2.0 of xlrd (pip install xlrd==1.2.0)
import pandas as pd
df_raw = pd.read_excel('/excel/file/path.xlsx', sheet_name=None)
df = pd.concat(df_raw, ignore_index=True)
df.to_csv("data/power_plants.csv", index=False)