create_industrial_use_case_and_model.py

# -*- coding: utf-8 -*-

# We start with the import of standard ML librairies
import pandas as pd
import numpy as np
import math

from sklearn.datasets import make_regression
from sklearn.ensemble import RandomForestRegressor

# We add all Plotly and Dash necessary librairies
import plotly.graph_objects as go

import dash
import dash_core_components as dcc
import dash_html_components as html
import dash_daq as daq
from dash.dependencies import Input, Output


# We start by creating a virtual regression use-case
X, y = make_regression(n_samples=1000, n_features=8, n_informative=5, random_state=22)

# We rename columns as industrial parameters
col_names = ["Temperature","Viscosity","Pressure", "pH","Inlet_flow", "Rotating_Speed","Particles_size","Color_density"]

df = pd.DataFrame(X, columns=col_names)

# We change the most important features ranges to make them look like actual figures
df["pH"]=6.5+df["pH"]/4
df["Pressure"]=10+df["Pressure"]
df["Temperature"]=20+df["Temperature"]
df["Y"] = 90+y/20

# We train a simple RF model
model = RandomForestRegressor()
model.fit(df.drop("Y", axis=1), df["Y"])

# We create a DataFrame to store the features' importance and their corresponding label
df_feature_importances = pd.DataFrame(model.feature_importances_*100,columns=["Importance"],index=col_names)
df_feature_importances = df_feature_importances.sort_values("Importance", ascending=False)