fuzzy-rules-generator/dataset.ipynb

In [2]:

import pandas as pd

density_train = pd.read_csv("data/density_train.csv", sep=";", decimal=",")
density_test = pd.read_csv("data/density_test.csv", sep=";", decimal=",")
d_data = pd.concat([density_train, density_test])
d_data.info()

<class 'pandas.core.frame.DataFrame'>
Index: 55 entries, 0 to 16
Data columns (total 4 columns):
 #   Column   Non-Null Count  Dtype  
---  ------   --------------  -----  
 0   T        55 non-null     int64  
 1   Al2O3    55 non-null     float64
 2   TiO2     55 non-null     float64
 3   Density  55 non-null     float64
dtypes: float64(3), int64(1)
memory usage: 2.1 KB

In [3]:

import matplotlib.pyplot as plt

d_data.hist(bins=30, figsize=(10, 10))
plt.show()

In [4]:

import seaborn as sns

sns.catplot(
    y="value",
    data=d_data.melt(value_vars=d_data.columns), # type: ignore
    col="variable",
    kind="box",
    col_wrap=2,
    sharex=False,
    sharey=False,
)

Out[4]:

<seaborn.axisgrid.FacetGrid at 0x123f26c30>

In [18]:

sns.heatmap(d_data.corr(), annot=True, fmt=".1%")

Out[18]:

<Axes: >

In [6]:

viscosity_train = pd.read_csv("data/viscosity_train.csv", sep=";", decimal=",")
viscosity_test = pd.read_csv("data/viscosity_test.csv", sep=";", decimal=",")

v_data = pd.concat([viscosity_train, viscosity_test])
v_data.info()

<class 'pandas.core.frame.DataFrame'>
Index: 55 entries, 0 to 16
Data columns (total 4 columns):
 #   Column     Non-Null Count  Dtype  
---  ------     --------------  -----  
 0   T          55 non-null     int64  
 1   Al2O3      55 non-null     float64
 2   TiO2       55 non-null     float64
 3   Viscosity  55 non-null     float64
dtypes: float64(3), int64(1)
memory usage: 2.1 KB

In [7]:

v_data.hist(bins=30, figsize=(10, 10))
plt.show()

In [8]:

sns.catplot(
    y="value",
    data=v_data.melt(value_vars=v_data.columns),  # type: ignore
    col="variable",
    kind="box",
    col_wrap=2,
    sharex=False,
    sharey=False,
)

Out[8]:

<seaborn.axisgrid.FacetGrid at 0x122dc5850>

In [19]:

sns.heatmap(v_data.corr(), annot=True, fmt=".1%")

Out[19]:

<Axes: >