https://github.com/hermeskeng/mnist
Get and solve the handwriting dataset from MNIST
https://github.com/hermeskeng/mnist
handwritten-digits machine-learning mnist-dataset python sklearn
Last synced: 8 months ago
JSON representation
Get and solve the handwriting dataset from MNIST
- Host: GitHub
- URL: https://github.com/hermeskeng/mnist
- Owner: HermesKeng
- Created: 2017-10-01T08:12:13.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2017-10-01T15:23:52.000Z (about 8 years ago)
- Last Synced: 2024-12-26T11:07:45.148Z (10 months ago)
- Topics: handwritten-digits, machine-learning, mnist-dataset, python, sklearn
- Language: Python
- Size: 3.91 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# MNIST
> This fuction sets is refer to https://gist.github.com/akesling/5358964 to implement reading the MNIST dataset
- Background: I'm the begin to learn Machine Learning through [Hands-On Machine Learning with Scikit-Learn and TensorFlow](http://shop.oreilly.com/product/0636920052289.do). There are some problems about example codes, when I start to learn in Classification section (CH3). The original method in sklearn method can't download MNIST dataset, so I start to refer to other solution to write the program to download the dataset, and show the data in program.
- Descrition: There are four function in the MNIST.py, and I am goint t give you some introduction.
- download_mnist:
Go to the [THE MNIST DATABASE](http://yann.lecun.com/exdb/mnist/) to download and decompress the dataset.
- load_file:
Load the file which have already downloaded and transform the gz file to binary file
- load_mnist:
Read the Binary file a and get the handwriting dataset and label
- showDigit:
Show the specified digit when you give the image in the dataset
- Image Size: The image size is 28 * 28, so the total size of image is 784 pixel
- Train Set : 60000 images and label
- Test Set : 10000 images and label
- Details : In the [THE MNIST DATABASE](http://yann.lecun.com/exdb/mnist/) document, it tells us for each compress file is a binary file. We need to transform it to corresponding data type, and we can know the binary stroe format at the end of the document
** These code is my first try to decompress image, if there is any advice to my code, please give me some note , Thanks