Introduction

We often use ssh (Secure SHELL) to connect server via terminal remotely. If you are tired entering password to login the server, just do the following steps to enter passwordless.

Introduction

Last year, someone broke our car’s window parked inside the parking and took some stuff, I though, it is good idea to look at the rubbing data in Montreal where we have been living.

How generate variable name using loop item

If you need to create the variable name using the loop object, use exec:

for i in range(4):
    exec(f'var_{i} = [range(i)]')

How to save all objects

# to save session
dill.dump_session('backup_2021_10_22.db')
# to load 
backup_restore = dill.load_session('backup_2021_10_22.db')

How run apply on array

import numpy as np
x = np.array([[5,2,1,3], [2,1,5]])
fun = lambda t: np.argmax(t)
np.array([fun(xi) for xi in x])

Classification

Data

To show how fit the multiple regression using R and Python, we consider the car data [car] which has the car specifications; HwyMPG, Model, etc,. We fit different regression models to predict Highway MPG using the car specifications.

Logistic Regression

The logistics regression is used when the response variable is a binary variable, such as (0 or 1), (live or die), and (fail or succeed).

Introduction

TensorFlow is a module developed to achieve the machine learning models. It is develped based on manipulating tensors, which are actually multidimensional array. It supports the hardware acceleration (GPU), which makes it suitable for machine learning model that need alot of computation.

Classification

In this short, we show how run several classification methods; XGBoost, Ada Boost, Discriminant Analysis, KNN, random forest, decision theory, Gaussian Process, Logistics regression, Gaussian Mixture Classification, SVM, and LSTM.

Introductory Notes on plot

Optimization

When we do not have a closed form, we can use the optimization to find estimate of parameters. In order to find optimization, you need a loos function and use a procedure to minimize the value of loss function based on the parameter values.

Iteration

Python is equipped with strong tools for the repeat of some commands or produce sequence number.

Data control structure

Function

In the context of programming, a function is a sequence of statements that performs a computation. Functions has three parts; argument, script, and output. Python has two kinds of function: built-in function that is in the core of Python or are collected as package. User-defined function that is written by user.

Data Structures

Python provides a variety of useful data structures, such as lists, sets, and dictionaries, and a new structure define by programmer which called class.

Summarizing data-frame

To see the type, the information and summary of variables in the data-frame, use .dtypes, .describe(), and .info().

pipeline

Pipeline in Pandas allows to build a sequence of function to run in order on data-frame.

Adding new column to data-frame

A column can easily be added to data-frame

df0=pd.DataFrame([38,40,25,33])
df['Ave_hour']=df0

Merging data-frames

Panada is very useful for merging dataset; to merging data consider the following data sets, where ‘id1’ and ‘id2’ include the ids of data.

Manipulating data-frame

To select the data use the name of variable, or specify the indices via .iloc and .loc (link)[http://pandas.pydata.org/pandas-docs/version/0.22/indexing.html]. .iloc is an integer-based select and should be used with integer indies. On contrary, .loc is primarily label based, and may also be used with a boolean array.

Data-frame

Data-frame via pandas is very useful format for working with dataset, its structure is two-dimensional size-mutable, potentially heterogeneous tabular data structure with labeled axes (rows and columns). The following codes create a data-frame from a dictionary.

Crosstab

To show how generate the cross tabulate, let us categorize the columns; consider two continuous variables ( e.g., housing_median_age and total_rooms), categorize them according their .3 and .7 quantiles, and label the elements as L, M, and H. Then find the cross tabulate of them,

Applying a function on row or column

Using df.apply(fun) can apply a function on columns or row:

df.apply(np.sum, axis=0)
df.apply(np.sum, axis=1)

Keras is a deep learning library written in Python which is running on top of TensorFlow, CNTK, or Theano.

Object

• Learn a basic knowledge about the Keras
• Learn classification

Data

To show how fit the multiple regression using R and Python, we consider the car data [car] which has the car specifications; HwyMPG, Model, etc,. We fit different regreesion models to predict Highway MPG using the car specifications.

Contents

SQL At A Glance

This text is a self-study to learn how use SQL to work with the databases.

R-Python-Matlab (Basics)

A brief cheatsheet for those who use R daily, and use other script programing languages, Python and Matlab, occasionally.