Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 13-15 (Chapter 14 Feature Understanding and Selection (14.3…
Chapter 13-15
Chapter 14 Feature Understanding and Selection
Interpret the data
Data types
Descriptive data
Feature content
Missing data
14.1 Descriptive data
Rows w/ names
Unique
Var type
click race
Feature data will expand
Mean
std dev
max & min
showing 7 bins
Right skewed distrubution
14.2 Data types
Binary
Categoricial
NUMERIC
Integers
Decimals
14.3 Evaluations of Feature Content
Evaluations of Feature Content
Frequent Values
Target
Too many values
"Customer names"
14.4 Missing Values
can avoid being categorized as multiple unique
Nulls
Employee ID
inf
INF
Inf
14.5 Exercises
Search
Chapter 15 Build Candidate
15.1 Starting the Process
Use as Target
Offers the option of which metric to optimize
True/false
LogLoss(Accuracy)
15.2 Advanced Options(not recommended for first time users
Show advanced options
Random vs Stratisfied
Partition feature
Group
Date/Time
15.3 Starting the Analytical Process
Autopilot
Quick
Manual
Quick run
Importance
15.4 Model Selction Process
Running algorithms
15.4.1 Tournament Round 1:16%
.2 Round 2
4.3
Round 3
.5 Cross Validation
Leads to blending
Chapter 13. Startup Processes
13.1 Uploading Data
Local File
Shared data
Raw data
Small data
recommended
Hospital diabetes readmission
URL
ODBC
HDFS
Hospital Excercise
Select Local Files
10k Diabetes
Unititled project
Create new project
manage project
1 more item...
13.2 Excercise