LICENSE

Copyright 2018 Google LLC.

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at

https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

42.89. Introduction

Climate Prediction-Random Forest is a model that uses a combination of climate variables and machine learning algorithms to predict future climate conditions. The model is trained on a large dataset of climate observations and uses a random forest approach to generate predictions. The predictions are based on the relationships between the climate variables and the random forest algorithm is able to capture complex patterns in the data.

42.89.1. Importing Libraries

# Pandas is used for data manipulation
import pandas as pd

# Use numpy to convert to arrays
import numpy as np

# Import tools needed for visualization

import matplotlib.pyplot as plt
%matplotlib inline

42.89.2. Data Exploration

# Reading the data to a dataframe 
df = pd.read_csv('https://static-1300131294.cos.ap-shanghai.myqcloud.com/data/classification/temps.csv')

# displaying first 5 rows
df.head(5)

	year	month	day	week	temp_2	temp_1	average	actual	friend
0	2019	1	1	Fri	45	45	45.6	45	29
1	2019	1	2	Sat	44	45	45.7	44	61
2	2019	1	3	Sun	45	44	45.8	41	56
3	2019	1	4	Mon	44	41	45.9	40	53
4	2019	1	5	Tues	41	40	46.0	44	41

# the shape of our features
df.shape

(348, 9)

# column names
df.columns

Index(['year', 'month', 'day', 'week', 'temp_2', 'temp_1', 'average', 'actual',
       'friend'],
      dtype='object')

# checking for null values
df.isnull().sum()

year       0
month      0
day        0
week       0
temp_2     0
temp_1     0
average    0
actual     0
friend     0
dtype: int64

There are no null values

42.89.3. One-Hot Encoding

A one hot encoding allows the representation of categorical data to be more expressive.

# One-hot encode categorical features
df = pd.get_dummies(df)
df.head(5)

	year	month	day	temp_2	temp_1	average	actual	friend	week_Fri	week_Mon	week_Sat	week_Sun	week_Thurs	week_Tues	week_Wed
0	2019	1	1	45	45	45.6	45	29	True	False	False	False	False	False	False
1	2019	1	2	44	45	45.7	44	61	False	False	True	False	False	False	False
2	2019	1	3	45	44	45.8	41	56	False	False	False	True	False	False	False
3	2019	1	4	44	41	45.9	40	53	False	True	False	False	False	False	False
4	2019	1	5	41	40	46.0	44	41	False	False	False	False	False	True	False

print('Shape of features after one-hot encoding:', df.shape)

Shape of features after one-hot encoding: (348, 15)

42.89.4. Features and Labels

# Labels are the values we want to predict
labels = df['actual']

# Remove the labels from the features
df = df.drop('actual', axis = 1)

# Saving feature names for later use
feature_list = list(df.columns)

42.89.5. Train Test Split

# Using Skicit-learn to split data into training and testing sets
from sklearn.model_selection import train_test_split

# Split the data into training and testing sets
train_features, test_features, train_labels, test_labels = train_test_split(df,
                                                                            labels,
                                                                            test_size = 0.20,
                                                                            random_state = 42)

print('Training Features Shape:', train_features.shape)
print('Training Labels Shape:', train_labels.shape)
print('Testing Features Shape:', test_features.shape)
print('Testing Labels Shape:', test_labels.shape)

Training Features Shape: (278, 14)
Training Labels Shape: (278,)
Testing Features Shape: (70, 14)
Testing Labels Shape: (70,)

42.89.6. Training the Forest

# Import the model we are using
from sklearn.ensemble import RandomForestRegressor

# Instantiate model 
rf = RandomForestRegressor(n_estimators= 1000, random_state=42)

# Train the model on training data
rf.fit(train_features, train_labels);

42.89.7. Make Predictions on Test Data

# Use the forest's predict method on the test data
predictions = rf.predict(test_features)

# Calculate the absolute errors
errors = abs(predictions - test_labels)

# Print out the mean absolute error (mae)
print('Mean Absolute Error:', round(np.mean(errors), 2), 'degrees.')

Mean Absolute Error: 3.78 degrees.

# Calculate mean absolute percentage error (MAPE)
mape = 100 * (errors / test_labels)

# Calculate and display accuracy
accuracy = 100 - np.mean(mape)
print('Accuracy:', round(accuracy, 2), '%.')

Accuracy: 94.02 %.

42.89.8. Visualizing a Single Decision Tree

42.89.9. Your turn! 🚀

You can practice your random-forest skills by following the assignment Climate Prediction-Random Forest.

42.89.10. Acknowledgments

Thanks to Kaggle for creating the open source course Climate Prediction-Random Forest. It contributes some of the content in this chapter.