maybe canvas can not support arm CPU ? .921Z Install :

Thanks, but I can't rebuild tensorflow print: <div class="highli

Can't install at Raspberry Pi 4 about node-red-contrib-face-recognition HOT 5 CLOSED

thebigpotatoe commented on August 27, 2024

Can't install at Raspberry Pi 4

from node-red-contrib-face-recognition.

Comments (5)

thebigpotatoe commented on August 27, 2024

Seems like there arn't any pre-compiled binaries for the raspberry pi 4 with your operating system;

2019-12-26T20:06:58.470Z [err] Pre-built binaries not found for [email protected] and [email protected] (node-v64 ABI, glibc) (falling back to source compile with node-gyp)

You may have to compile canvas yourself and its dependencies in your .node-red folder. I haven't installed it on a Pi yet as mine has not been playing ball recently, so i'm not 100% sure exactly how to get canvas going exactly but this may help.

from node-red-contrib-face-recognition.

kent1026 commented on August 27, 2024

Thanks, it's work. but when I use npm install @tensorflow/tfjs-node
then node-red debug print :
[Face-api.js : 75ff9e47.e428c : Child Node] - Failed to load in TensorFlow.js for Node.js in child_process: TypeError: tfjs_core_1.registerKernel is not a function"

and Node-Red print error message:

Flows stopped due to missing node types.
face-api-compute
face-api-input

How can i do ?
Thanks.

from node-red-contrib-face-recognition.

thebigpotatoe commented on August 27, 2024

To clarify it works without tfjs-node?

With tfjs-node though, having a look on the tfjs-node GitHub page it states you need to rebuild the package on raspberry pi using rebuild @tensorflow/tfjs-node --build-from-source

from node-red-contrib-face-recognition.

kent1026 commented on August 27, 2024

Thanks, but I can't rebuild tensorflow

print:

pi@raspberrypi:~/.node-red$ npm rebuild @tensorflow/tfjs-node --build-from-source                                                                               │··········
                                                                                                                                                                │··········
> @tensorflow/[email protected] install /home/pi/.node-red/node_modules/@tensorflow/tfjs-node                                                                     │··········
> node scripts/install.js                                                                                                                                       │··········
                                                                                                                                                                │··········
CPU-linux-1.5.1.tar.gz                                                                                                                                          │··········
* Downloading libtensorflow                                                                                                                                     │··········
(node:4484) UnhandledPromiseRejectionWarning: Error: Unsupported system: cpu-linux-arm                                                                          │··········
    at getPlatformLibtensorflowUri (/home/pi/.node-red/node_modules/@tensorflow/tfjs-node/scripts/install.js:95:11)                                             │··········
    at downloadLibtensorflow (/home/pi/.node-red/node_modules/@tensorflow/tfjs-node/scripts/install.js:129:7)                                                   │··········
    at async run (/home/pi/.node-red/node_modules/@tensorflow/tfjs-node/scripts/install.js:190:5)                                                               │··········
(node:4484) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch │··········
block, or by rejecting a promise which was not handled with .catch(). (rejection id: 1)                                                                         │··········
(node:4484) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the│··········
 Node.js process with a non-zero exit code.                                                                                                                     │··········
@tensorflow/[email protected] /home/pi/.node-red/node_modules/@tensorflow/tfjs-node

and I follow this Running face-api.js or tfjs-node on a raspberry pi and node.js
it can't work too

from node-red-contrib-face-recognition.

thebigpotatoe commented on August 27, 2024

These errors are not unique to this repo, and as I have stated, I have not attempted to run this on a raspberry pi 4. So I will be closing this issue as this space should be used for tracking issues unique to this repo, not issues related to getting tfjs installed on a Pi. If you have an issue during running or using this repo after a successful installation feel free to pop open another one.

Some help though; I am still earning JS and how to manage it through NPM myself, but your error messages seem to suggest that the pi and the operating system you are running are not supported by the required libraries (yet?). My suggestion is to browse the tfjs-node repo and face-api.js repo for raspberry pi issues and answers.

I have noticed that you are running node 10.18.0 from the above messages, so I would suggest updating that for a start. Then you may be able to follow those steps in the link you supplied.

When I was building this, I found that the basic Face-api.js worked out of the box easily, but getting tfjs-node to run as well was painful. I am using windows, and as in the readme, I found a specific setup that worked for me. You will need to do the same as I have not used a pi for testing.

from node-red-contrib-face-recognition.

Can't install at Raspberry Pi 4 about node-red-contrib-face-recognition HOT 5 CLOSED

Comments (5)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent