Putting It All Together

In this challenge, you will apply everything you learned throughout the course. Here are the steps you need to take:

Remove the rows that hold too little information;
Encode the y;
Split the dataset into training and test sets;
Build a pipeline with all the preprocessing steps and the GridSearchCV as the final estimator to find the best hyperparameters;
Train the model using the pipeline;
Evaluate the model using the pipeline;
Predict the target for X_new and decode it using the LabelEncoder's .inverse_transform().

Let's get to w̵o̵r̵k̵ code!

Завдання

Swipe to start coding

Encode the target using LabelEncoder.
Split the data so that 33% is used for a test set and the rest – for a training set.
Make a ColumnTransformer to encode only the 'island' and 'sex' columns. Make the others remain untouched. Use a proper encoder for nominal data.
Fill the gaps in a param_grid to try the following values for the number of neighbors: [1, 3, 5, 7, 9, 12, 15, 20, 25].
Create a GridSearchCV object with the KNeighborsClassifier as a model.
Make a pipeline with ct as a first step and grid_search as a final estimator.
Train the model using a pipeline on the training set.
Evaluate the model on the test set. (Print its score)
Get a predicted target for X_test.
Print the best estimator found by grid_search.

Рішення

import pandas as pd

from sklearn.impute import SimpleImputer

from sklearn.preprocessing import OneHotEncoder, LabelEncoder, StandardScaler

from sklearn.compose import make_column_transformer

from sklearn.pipeline import make_pipeline

from sklearn.neighbors import KNeighborsClassifier

from sklearn.model_selection import GridSearchCV, train_test_split

df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/a65bbc96-309e-4df9-a790-a1eb8c815a1c/penguins.csv')

# Removing rows with more than 1 null

df = df[df.isna().sum(axis=1) < 2]

# Assigining X, y variables

X, y = df.drop('species', axis=1), df['species']

# Encode the target

label_enc = LabelEncoder()

y = label_enc.fit_transform(y)

# Make a train-test split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)

# Create the ColumnTransformer for encoding features

ct = make_column_transformer((OneHotEncoder(), ['island', 'sex']),

remainder='passthrough')

# Make a param_grid for the grid search and initialize the GridSearchCV object

param_grid = {'n_neighbors': [1, 3, 5, 7, 9, 12, 15, 20, 25],

'weights': ['distance', 'uniform'],

'p': [1, 2]

}

grid_search = GridSearchCV(KNeighborsClassifier(), param_grid)

# Make a Pipeline of ct, SimpleImputer, and StandardScaler

pipe = make_pipeline(ct,

SimpleImputer(strategy='most_frequent'),

StandardScaler(),

grid_search

)

# Train the model

pipe.fit(X_train, y_train)

# Print score

Все було зрозуміло?

Дякуємо за ваш відгук!

Секція 4. Розділ 10

import pandas as pd

from sklearn.impute import SimpleImputer

from sklearn.preprocessing import OneHotEncoder, LabelEncoder, StandardScaler

from sklearn.compose import make_column_transformer

from sklearn.pipeline import make_pipeline

from sklearn.neighbors import KNeighborsClassifier

from sklearn.model_selection import GridSearchCV, train_test_split

df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/a65bbc96-309e-4df9-a790-a1eb8c815a1c/penguins.csv')

# Removing rows with more than 1 null

df = df[df.isna().sum(axis=1) < 2]

# Assigining X, y variables

X, y = df.drop('species', axis=1), df['species']

# Encode the target

label_enc = LabelEncoder()

y = label_enc.___(y)

# Make a train-test split

X_train, X_test, y_train, y_test = ___(X, y, ___=0.33)

# Create the ColumnTransformer for encoding features

ct = make_column_transformer((___(), ['island', 'sex']),

remainder='___')

# Make a param_grid for the grid search and initialize the GridSearchCV object

param_grid = {'n_neighbors': ___,

'weights': ['distance', 'uniform'],

'p': [1, 2]

}

grid_search = ___(KNeighborsClassifier(), ___)

# Make a Pipeline of ct, SimpleImputer, and StandardScaler

pipe = make_pipeline(___,

SimpleImputer(strategy='most_frequent'),

StandardScaler(),

___

)

# Train the model

pipe.___(___, ___)

# Print score

print(pipe.___(X_test, ___))

# Print predictions

y_pred = pipe.___(X_test) # Get encoded predictions

print(label_enc.inverse_transform(y_pred[:5])) # Decode predictions and print 5 first

# Print the best estimator

print(grid_search.___)