JarrydWannenburg

import altair as alt
import pandas as pd

# Ignore size limits
alt.data_transformers.enable('default', max_rows=None)

# Create a function that takes the dataset and column name and returns an interactive histogram
def chart(dataset, column_name, target_var):
    w = 500
    single = alt.selection_single()

JarrydWannenburg

# Try/Except to append multiple altair charts to a list
myl = []
for col in dataset.columns:
    try:
        myl.append(chart(dataset, col, target_var))
    except:
        pass
    else:
        pass

JarrydWannenburg

import altair as alt
import pandas as pd

source = pd.DataFrame({
    'Language': ['Python', 'C++', 'Java', 'Perl', 'Scala', 'Lisp'],
    'Usage': [10,8,6,4,2,1]
})

alt.Chart(source, title = "Programming language usage").mark_bar().encode(
    x=alt.X('Language', sort='-y'),

JarrydWannenburg

import matplotlib.pyplot as plt; plt.rcdefaults()
import numpy as np
import matplotlib.pyplot as plt

objects = ('Python', 'C++', 'Java', 'Perl', 'Scala', 'Lisp')
y_pos = np.arange(len(objects))
performance = [10,8,6,4,2,1]

plt.bar(y_pos, performance, align='center', alpha=0.5)
plt.xticks(y_pos, objects)

JarrydWannenburg

import pandas as pd
import numpy as np
from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.compose import ColumnTransformer, make_column_selector as Selector
from sklearn.decomposition import PCA
from sklearn.experimental import enable_iterative_imputer
from sklearn.impute import IterativeImputer, SimpleImputer
from sklearn.svm import SVC
from sklearn.feature_selection import SelectKBest, chi2
from sklearn.model_selection import train_test_split, GridSearchCV

JarrydWannenburg

class CategoricalTransformer(BaseEstimator, TransformerMixin):
    def __init__(self, sep_passengerId = True, sep_cabin = True, convert_bools = True, excluded_features = ['Name']):
        self.sep_passengerId = sep_passengerId
        self.sep_cabin = sep_cabin
        self.convert_bools = convert_bools
        self.excluded_features = excluded_features
    
    def fit(self, X, y=None):
        return self
    

JarrydWannenburg

class CatFeatureEngineerTransformer(BaseEstimator, TransformerMixin):
    def __init__(self, home_dest = True):
        self.home_dest = home_dest
        
    def fit(self, X, y=None):
        return self
    
    def transform(self, X, y=None):
        if self.home_dest:
            X['HomeDest'] = X.HomePlanet + ':' + X.Destination

JarrydWannenburg

class Nominal_Columns(BaseEstimator, TransformerMixin):
    def __init__(self, exclude=['Deck','Deck2']): # These would be your ordinal columns. Be sure to understand the order too.
        self.exclude = exclude

    def fit(self, X, y=None):
        return self

    def transform(self, X, y=None):
        _X = X.copy()
        _X.drop(columns=self.exclude, inplace=True)

JarrydWannenburg

class Ordinal_Columns(BaseEstimator, TransformerMixin):
    def __init__(self, names=['Deck','Deck2']): #,'Deck2'
        self.names = names

    def fit(self, X, y=None):
        return self

    def transform(self, X, y=None):
        if len(self.names) > 0:
            if len(self.names) == 1:

JarrydWannenburg

cat_pipe = Pipeline([
    ('CategoricalTransformer', CategoricalTransformer()), # Manually included fillna(mode) similar to simple_imputer
    ('CatFeatureEngineer', CatFeatureEngineerTransformer()),
    ('Subtypes', FeatureUnion([
            ('ordinal', Pipeline([
                ('cols', Ordinal_Columns()),
                ('encoder', OrdinalEncoder(categories=[['T','G','F', 'E', 'D', 'C', 'B', 'A'],# Need row of values (low to high)
                                                      ['G','F', 'E', 'D', 'C', 'B', 'A']], # can use auto instead of categories
                        # We see 11 total records with 'T' but external research doesn't support that value.
                         handle_unknown='use_encoded_value',