## melihat jumlah value dari num-of-doors
# df["num-of-doors"].value_counts()

## melihat value yang paling banyak muncul
df['num-of-doors'].value_counts().idxmax()

df["num-of-doors"].replace(np.nan, 'four', inplace=True)

## remove baris jika value price kosong
df.dropna(subset=["price"], axis=0, inplace=True)

## reset index setelah menghapus sebagian baris
df.reset_index(drop=True, inplace=True)
df.head()

df[["bore", "stroke", "price", "peak-rpm"]] = df[["bore", "stroke", "price", "peak-rpm"]].astype("float")
df[["normalized-losses", "horsepower"]] = df[["normalized-losses", "horsepower"]].astype("int")
# df[["price"]] = df[["price"]].astype("float")
# df[["peak-rpm"]] = df[["peak-rpm"]].astype("float")

df.dtypes

# mengubah mile/galon ke L/100km
df["city-mpg"] = 235/df["city-mpg"]
df.rename(columns={"city-mpg": "city-L/100km"}, inplace=True)
df["city-L/100km"].head()

df["highway-mpg"] = 235/df["highway-mpg"]
df.rename(columns={"highway-mpg": "highway-L/100km"}, inplace=True)
df["highway-L/100km"].head()

## menormalisasi width dan length menjadi range 0 - 1(menggunakan simple feature Xold/Xmax)
df["width"] = df["width"]/df["width"].max()
df["height"] = df["height"]/df["height"].max()
df["length"] = df["length"]/df["length"].max()

df[["width", "height", "length"]].head()

import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline 
plt.hist(df["horsepower"])

# set x/y labels and plot title
plt.xlabel("horsepower")
plt.ylabel("count")
plt.title("horsepower bins")

## membuat edge bin 4 value dengan min awal dan max akhir untuk digunakan dalam 3 bin([48-119], [119-190], [190-262])
bins = np.linspace(min(df["horsepower"]), max(df["horsepower"]), 4)
bins

group_names = ['Low', 'Medium', 'High']

## 
df['horsepower-binned'] = pd.cut(df['horsepower'], bins, labels=group_names, include_lowest=True )
df["horsepower-binned"].value_counts()

pyplot.bar(group_names, df["horsepower-binned"].value_counts())

# set x/y labels and plot title
plt.xlabel("horsepower")
plt.ylabel("count")
plt.title("horsepower bins")

# menggunakan historgram "horsepower" dengan bins = 3
plt.pyplot.hist(df["horsepower"], bins = 3)

# set x/y labels and plot title
plt.xlabel("horsepower")
plt.ylabel("count")
plt.title("horsepower bins")

## dummy variabel untuk mengkategorikan tipe bahan bakar
dummy_variable_1 = pd.get_dummies(df["fuel-type"])
dummy_variable_1.head()

dummy_variable_1.rename(columns={'gas':'bensin', 'diesel':'solar'}, inplace=True)
dummy_variable_1.head()

# merge data frame "df" dg "dummy_variable_1" 
df = pd.concat([df, dummy_variable_1], axis=1)

# drop "fuel-type"
df.drop("fuel-type", axis = 1, inplace=True)
df.head()

dummy_aspirasi = pd.get_dummies(df["aspiration"])
dummy_aspirasi.head()

df = pd.concat([df, dummy_aspirasi], axis=1)
df.drop("aspiration", axis=1, inplace=True)

df.head()

## melihat korelasi
df[['bore','stroke','compression-ratio','horsepower']].corr()

import seaborn as sns

# Semakin besar engine semakin mahal
sns.regplot(x="engine-size", y="price", data=df)
plt.ylim(0,)

## test nilai korelasi
df[["engine-size", "price"]].corr()

## korelasi seberapa boros di jalan lepas terhadap harga
## semakin boros semakin murah
sns.regplot(x="highway-L/100km", y="price", data=df)

df[["highway-L/100km", "price"]].corr()

sns.regplot(x="peak-rpm", y="price", data=df)

df[['peak-rpm','price']].corr()

df[["stroke","price"]].corr()

sns.regplot(x="stroke", y="price", data=df)

sns.boxplot(x="body-style", y="price", data=df)

sns.boxplot(x="engine-location", y="price", data=df)

sns.boxplot(x="drive-wheels", y="price", data=df)

df.describe()

df.describe(include=object)
# df["num-of-doors"].value_counts()

df['drive-wheels'].value_counts().to_frame()

# melihat value engine location
engine_loc_counts = df['engine-location'].value_counts().to_frame()
engine_loc_counts.rename(columns={'engine-location': 'value_counts'}, inplace=True)
engine_loc_counts.index.name = 'engine-location'
engine_loc_counts.head(10)

df['drive-wheels'].unique()

df_gptest = df[['drive-wheels','body-style','price']]
df_group_wheel = df_gptest.groupby(['drive-wheels', 'body-style'],as_index=False).mean()
df_group_wheel

pivot_group = df_group_wheel.pivot(index='drive-wheels',columns='body-style')
pivot_group

## isi NaN dengan 0
pivot_group = pivot_group.fillna(0)
pivot_group

df_body = df[["body-style", "price"]]
df_body = df_body.groupby("body-style", as_index=False).mean()
df_body

plt.pcolor(pivot_group, cmap='RdBu')
plt.colorbar()
plt.show()

fig, ax = plt.subplots()
im = ax.pcolor(pivot_group, cmap='RdBu')

#label names
row_labels = pivot_group.columns.levels[1]
col_labels = pivot_group.index

#move ticks and labels to the center
ax.set_xticks(np.arange(pivot_group.shape[1]) + 0.5, minor=False)
ax.set_yticks(np.arange(pivot_group.shape[0]) + 0.5, minor=False)

#insert labels
ax.set_xticklabels(row_labels, minor=False)
ax.set_yticklabels(col_labels, minor=False)

#rotate label if too long
plt.xticks(rotation=90)

fig.colorbar(im)
plt.show()

from scipy import stats

pearson_coef, p_value = stats.pearsonr(df['wheel-base'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P =", p_value)

pearson_coef, p_value = stats.pearsonr(df['horsepower'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P = ", p_value)

pearson_coef, p_value = stats.pearsonr(df['length'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P = ", p_value)

pearson_coef, p_value = stats.pearsonr(df['width'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P =", p_value )

pearson_coef, p_value = stats.pearsonr(df['curb-weight'], df['price'])
print( "The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P = ", p_value)

pearson_coef, p_value = stats.pearsonr(df['engine-size'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P =", p_value)

pearson_coef, p_value = stats.pearsonr(df['bore'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P =  ", p_value )

pearson_coef, p_value = stats.pearsonr(df['city-L/100km'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P = ", p_value)

pearson_coef, p_value = stats.pearsonr(df['highway-L/100km'], df['price'])
print("The Pearson Correlation Coefficient is", pearson_coef, " with a P-value of P = ", p_value)

df_group_wheel2 = df_gptest[["drive-wheels", "price"]].groupby(["drive-wheels"])
df_group_wheel2.head(2)

df_group_wheel2.get_group('rwd')['price']

# ANOVA
f_val, p_val = stats.f_oneway(df_group_wheel2.get_group('fwd')['price'], df_group_wheel2.get_group('rwd')['price'], df_group_wheel2.get_group('4wd')['price'])  

print( "ANOVA results: F=", f_val, ", P =", p_val)

f_val, p_val = stats.f_oneway(df_group_wheel2.get_group('fwd')['price'], df_group_wheel2.get_group('rwd')['price'])  

print( "ANOVA results: F=", f_val, ", P =", p_val )

f_val, p_val = stats.f_oneway(df_group_wheel2.get_group('4wd')['price'], df_group_wheel2.get_group('rwd')['price'])  

print( "ANOVA results: F=", f_val, ", P =", p_val)

f_val, p_val = stats.f_oneway(df_group_wheel2.get_group('4wd')['price'], df_group_wheel2.get_group('fwd')['price'])  

print("ANOVA results: F=", f_val, ", P =", p_val)