Dataset
Classifier
We will be using a stump as our classifier.
A stump finds an outcome using one condition.
Next ->
There are many conditions we can use to classify images. For this demo, we will use partition pixel intensity.
Next ->
Our stumps will first partition each image into two sets.
Next ->
The set with the highest average pixel intensity will determine the output.
Next ->
We will "train" stumps by trying 10 partitions and selecting the one with the most significant result.
Next ->
AdaBoost
This section is best viewed on a device with a wider screen
Ok ->
Train Classifier -> Test Classifier -> Get Importance -> Update Distribution -> Add to Forest
Sample XXXX
?????
W: 0.1234
Sample XXXX
?????
W: 0.1234
Sample XXXX
?????
W: 0.1234
Sample XXXX
?????
W: 0.1234
Sample XXXX
?????
W: 0.1234
Sample XXXX
?????
W: 0.1234
Classifier #0
Training: Not Started
Testing: Not Started
Correct: ?
Incorrect: ?
Accuracy: ?
Importance: ?
Forest
Evaluation: ?
Forest Status
Accuracy: ?
Members: ?
Credits
Created with lots of ☕ by piman51277
To Top ->