https://github.com/kernelerr/paddlenode

A third-party node.js addon for Paddle Inference.
https://github.com/kernelerr/paddlenode

Last synced: 25 days ago
JSON representation

A third-party node.js addon for Paddle Inference.

• Host: GitHub
• URL: https://github.com/kernelerr/paddlenode
• Owner: KernelErr
• License: apache-2.0
• Created: 2020-07-05T15:03:12.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2021-02-22T06:29:30.000Z (over 3 years ago)
• Last Synced: 2023-03-08T23:54:12.890Z (over 1 year ago)
• Language: C++
• Homepage:
• Size: 1.09 MB
• Stars: 8
• Watchers: 2
• Forks: 5
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# paddle node

[![FOSSA Status](https://app.fossa.com/api/projects/git%2Bgithub.com%2FKernelErr%2Fpaddlenode.svg?type=shield)](https://app.fossa.com/projects/git%2Bgithub.com%2FKernelErr%2Fpaddlenode?ref=badge_shield)

A third-party node.js addon for Paddle Inference. The beginning two versions of paddle node are using Paddle Lite for backbone, but we use Paddle Inference instead as it is more compatible for x86_64 platform.

Tested on Windows 10, Node.js v12.16.3.

Current features:

- Import combined model.

- Infer from float array.

- Image classification from file.

## Methods

#### set_combined_model

```javascript

paddlenode.set_combined_model(ModelDir)

```

#### infer_float

```javascript

paddlenode.infer_float(Data, Size)

```

#### image_file_classification

```javascript

paddlenode.image_file_classification(ImagePath, InputSize, Scalefactor, Mean, swapRB)

```

## Example

### Image Classification

```javascript

var paddlenode = require('./paddlenode');

paddlenode.set_combined_model('./mobilenetv1');

var res = paddlenode.image_file_classfication("test.jpg",[1, 3, 224, 224],0.007843,[224,224],[0.485,0.456,0.406], false)

console.log('Result', res.indexOf(Math.max(...res)));

```

### Sample Vector

```javascript

var paddlenode = require('./paddlenode');

paddlenode.set_combined_model('./mobilenetv1');

var arr = new Array(150528);

for(var i=0; i