Scribd
Upload
76 views5 pages

3-Month Data Science Learning Plan

The document outlines a 3-month learning plan for Data Science and Machine Learning, detailing weekly topics and resources such as Python, statistics, machine learning algorithms, and deep learning. It includes links to online courses and projects on platforms like EdX, Khan Academy, and Kaggle. Additionally, it emphasizes practical applications, feature engineering, and coding libraries in Python.

Uploaded by

Daniel Wu
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
76 views5 pages

3-Month Data Science Learning Plan

The document outlines a 3-month learning plan for Data Science and Machine Learning, detailing weekly topics and resources such as Python, statistics, machine learning algorithms, and deep learning. It includes links to online courses and projects on platforms like EdX, Khan Academy, and Kaggle. Additionally, it emphasizes practical applications, feature engineering, and coding libraries in Python.

Uploaded by

Daniel Wu
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Learn Data Science in 3 Months

6/24

Week 1 - Learn Python - EdX [Link]


- Siraj Raval [Link]

Week 2 - Statistics & Probability - KhanAcademy [Link]

Week 3 - Data Pre-processing, Data Vis, Exploratory Data Analysis - EdX


[Link]

Week 4 - Kaggle Project #1

Week 5-6 - Algorithms & Machine Learning - Columbia [Link]

Week 7 - Deep Learning - Part 1 and 2 of DL Book [Link]


- Siraj Raval [Link]

Week 8 - Kaggle Project #2 Week 9 - Databases (SQL + NoSQL) - Udacity


[Link]
- EdX [Link]

Week 10 - Hadoop & Map Reduce + Spark - Udacity [Link]


- Spark Workshop [Link]

Week 11 - Data Storytelling - Edx [Link]

Week 12- Kaggle Project #3


Learn Machine Learning in 3 Months
第 1 个月

第一周 线性代数 [Link]

第二周 微积分 [Link]

第三周 [Link]

第四周 算法 [Link]

第 2 个月

第一周 learn python for data science [Link]

Math of Intelligence [Link]

Intro to Tensorflow [Link]

第二周 Intro to ML (Udacity) [Link]

第三四周 ML Project Ideas [Link]

第 3 个月（深度学习）

第一周 Intro to Deep Learning [Link] 第二周 Deep Learning by


[Link] [Link]

第三四周 按照我的 github 重新实现深度学习项目 [Link]

 Linear regression
 Logistic regression
 Random forest
 Gradient boosting
 PCA
 k-mean clustering
 k nearest neighbors
 Natural language processing (2 sessions)
 Exploratory data analysis
 Python web APIs
 Feature engineering (2 sessions)
 Object-oriented programming
 Forecasting
 Linear regression
 Logistic regression
 SVM
 Random forest
 Gradient boosting
 PCA
 k-means
 Collaborative filtering
 kNN
 ARIMA

Business use case -> Domain expertise

Data gathering from vary data source (balance vs. unbalance dataset)

Whether the data is in the right format cleansing, wrangling, exploring EDA and how to handle the
missing value, to better put into ML algorithm. (Feature Engineering -> also apply some stats knowledge
to check Mean, Median, Mode)

Feature selection (regression back elimination, p-value)

Modeling (ML, DL algorithm select 1. accuracy. 2. Confusion matrix. 3. Cross validation

Coding library
Python:
The Inplace parameter

The inplace parameter is commonly used with the following methods:

 dropna()

 drop_duplicates()

 fillna()

 query()

 rename()

 reset_index()

 sort_index()

 sort_values()

import itertools
import numpy as np
import [Link] as plt
%matplotlib inline

import warnings
[Link]("ignore")

[Link]('fivethirtyeight')

import [Link] as sm
import [Link]['[Link]'] = 14
[Link]['[Link]'] = 12
[Link]['[Link]'] = 12
[Link]['[Link]'] = 'k'

for p in p_values:

for d in d_values:

for q in q_values:

order = (p,d,q)

train, test = shampoo [0:25], shampoo[25:36]

prediction = list()

for i in range(len(test)):

try:

model = ARIMA(train, order)

model_fit = [Link](disp=0)

pred_y = model_fit.forecast()[0]

[Link](pred_y)

error = mean_squared_error(test,predictions)

print('ARIMA%s MSE = %.2f'%(order,error))

except:

continue

You might also like