Skip to content

palashmoon/Google-Stock-price

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Google-Stock-price

Dependencies

1. pip install numpy
2. pip install panda
3. pip install matplotlib
4. pip install scikit-learn
5. pip install keras 

Import libraries

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

Importing training set

dataset_train = pd.read_csv('Google_Stock_Price_Train.csv')
training_set = dataset_train.iloc[: , 1:2].values

we are interseted only in the open coloumn of the Google_stock_price csv file thats the reason we extracted only open coloumn from our dataset.

Feature Scaling

1.why feature scaling?

If a feature in the dataset is big in scale compared to others then in algorithms where Euclidean distance is measured this big scaled feature becomes dominating and needs to be normalized.

Examples of Algorithms where Feature Scaling matters

  1. K-Means uses the Euclidean distance measure here feature scaling matters.
  2. K-Nearest-Neighbours also require feature scaling.
  3. Principal Component Analysis (PCA): Tries to get the feature with maximum variance, here too feature scaling is required.
  4. Gradient Descent: Calculation speed increase as Theta calculation becomes faster after feature scaling.

Note: Naive Bayes, Linear Discriminant Analysis, and Tree-Based models are not affected by feature scaling. In Short, any Algorithm which is Not Distance based is Not affected by Feature Scaling.

from sklearn.preprocessing import MinMaxScaler
sc = MinMaxScaler(feature_range = (0,1))
training_set_scale = sc.fit_transform(training_set)

We have two choice for feature scaling 1. standard deviation 2. Normalization here we are using Normalization bcoz Normalization is more suited in RNN and in the activation fuction where Sigmoid function is use at the output.

Create a data Structure

y_train=[]
for i in range(60 , 1258):
    X_train.append(training_set_scale[i-60:i , 0])
    y_train.append(training_set_scale[i , 0])

X_train , y_train = np.array(X_train) , np.array(y_train)

X_train=[] contains the list of all 60 previous openings of google stock market. y_train=[] it is the opening of 60+1 st financial day of google stock market.

X_train.append(training_set_scale[i-60:i , 0]) it contains all the opening of all 60days for coloumn 0 which is for opening coloumn. y_train.append(training_set_scale[i , 0]) 60+1 st opening

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages