Chap 3: Capital Share Bikeshare Predictor: Follow the Book

May 05, 2022

Import

Data Preprocess

Load & View the data

make 1-hot & binary input features

check total number of classes to separate the binary and those has more than 2 classes.

make the 1-hot features

include the binary features into the dataframe of input.

process the quantitative data

normalize the quantitative data.

add them into the input & target dataframe.

make the input & target tensors

beware the values.reshape(len(df_tgt),1). For the output is only 1 dimensional, such reshape help avoid the warning of dimensional mismatch.

make dataset objects

define & make model

make train & test function

Train & Test

Cycle 1 on train data, looks not good enough; 1 nore cycle.

Cycle 2 looks similar to cycle 1 on train data. check test data. fair performance.

Cycle 3, train more epoches. performance on train data looks similar as before. also similar on test data.