@tensorflow/tfjs-node installation failed(windows)

Step-by-step guide to fixing @tensorflow/tfjs-node installation errors on Windows

Having trouble installing @tensorflow/tfjs-node on Windows? You're not alone — the installation can be frustrating due to build dependencies and platform differences. Here's a step-by-step guide for you based on what worked for me:

1. Downgrade Node.js, Python

   Node.js: v20.19.0

   Python: 3.9.13

   Download Node.js and Python file from those link:

   Download Node.js

   Download Python

   

   +++node version 18.16.1 enables tfjs-node download with binary files in single folder. This may help you prevent DLOPEN_ERROR when running your project mentioned below.

2. Download from Visual Studio Build Tools

   First, download Visual Studio Build Tools from the link below:

   Download Visual Studio Build Tools

   From Visual Studio installer, select “Desktop development with C++”

   

   From installation details on the right side of window, select

   • MSVC v143

   • Windows 11 SDK(Depends on window version)

   • C++ CMake tools for windows

     

   and restart your pc

3. Update/Create .npmrc file

   run:

   echo msvs_version=2022 > C:\Users\[username on your pc]\.npmrc on powershell(Run as Admin)

   ⚠️Make sure to replace [username on your pc] to the directory name appears on your pc from C:\Users

   

4. Clean cache and install tensorflow

   from project folder, run:

   npm cache clean --force
   
   npm install @tensorflow/tfjs-node

   

Try running your project.

+++DLOPEN_FAILED Error Solution

Please go through this steps if the error below is shown when running your project.

1. Remove node_modules, package-lock.json

2. Clean cache and re-install packages by running the commands below:

   npm cache clean --force
   
   npm install

3. Check directory of .dll file and binding file

   run

   dir node_modules\@tensorflow\tfjs-node\lib\napi-v9\

   and

   dir node_modules\@tensorflow\tfjs-node\lib\napi-v8\

   tensorflow.dll file and binding file should be in the same folder. if they are in the separate folder,

4. Copy .dll file into the folder that has binding file

   run:

   copy node_modules\@tensorflow\tfjs-node\lib\napi-v9\tensorflow.dll node_modules\@tensorflow\tfjs-node\lib\napi-v8\

   or

   copy node_modules\@tensorflow\tfjs-node\lib\napi-v8\tensorflow.dll node_modules\@tensorflow\tfjs-node\lib\napi-v9\

   (Depending on .dll file location)

5. Test if you’re able to import tensorflow.

   Test command below to see if tensorflow works:

   node
   const tf = require('@tensorflow/tfjs-node'); 
   console.log('TensorFlow version:', tf.version.tfjs);

   

   If the Tensorflow version is shown, congratulations it is working🎉

   Go ahead and run your project!