Skip to content

Latest commit

 

History

History
 
 

mnist-node

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 
data.js
data.js
 
 
main.js
main.js
 
 
model.js
model.js
 
 
package-lock.json
package-lock.json
 
 
package.json
package.json
 
 
yarn.lock
yarn.lock
 
 

README.md

TensorFlow.js Example: Training MNIST with Node.js

This example shows you how to train MNIST (using the layers API) under Node.js.

This model will compute accuracy after one pass through the training dataset (60,000 samples) and evaluate 50 images from the test data set for accuracy after each epoch.

Prepare the node environment:

$ npm install
# Or
$ yarn

Run the training script:

$ node main.js

If you are running on a Linux system that is CUDA compatible, try installing the GPU package and replacing the require statement:

$ npm install @tensorflow/tfjs-node-gpu
# Or
$ yarn add @tensorflow/tfjs-node-gpu

After installing the package, replace require('@tensorflow/tfjs-node') with require('@tensorflow/tfjs-node-gpu'); in main.js