whitphx commited on
Commit
c162fa7
·
verified ·
1 Parent(s): 2e78a62

Add/update the quantized ONNX model files and README.md for Transformers.js v3

Browse files

## Applied Quantizations

### ✅ Based on `model.onnx` *with* slimming

↳ ❌ `int8`: `model_int8.onnx` (added but JS-based E2E test failed)
```
/home/ubuntu/src/tjsmigration/node_modules/.pnpm/onnxruntime-node@1.21.0/node_modules/onnxruntime-node/dist/backend.js:25
__classPrivateFieldGet(this, _OnnxruntimeSessionHandler_inferenceSession, "f").loadModel(pathOrBuffer, options);
^

Error: Could not find an implementation for ConvInteger(10) node with name '/convnext/embeddings/patch_embeddings/Conv_quant'
at new OnnxruntimeSessionHandler (/home/ubuntu/src/tjsmigration/node_modules/.pnpm/onnxruntime-node@1.21.0/node_modules/onnxruntime-node/dist/backend.js:25:92)
at Immediate.<anonymous> (/home/ubuntu/src/tjsmigration/node_modules/.pnpm/onnxruntime-node@1.21.0/node_modules/onnxruntime-node/dist/backend.js:67:29)
at process.processImmediate (node:internal/timers:485:21)

Node.js v22.16.0
```
↳ ✅ `uint8`: `model_uint8.onnx` (added)
↳ ✅ `q4`: `model_q4.onnx` (added)
↳ ✅ `q4f16`: `model_q4f16.onnx` (added)
↳ ✅ `bnb4`: `model_bnb4.onnx` (added)

README.md CHANGED
@@ -7,15 +7,15 @@ https://huggingface.co/facebook/convnext-base-224 with ONNX weights to be compat
7
 
8
  ## Usage (Transformers.js)
9
 
10
- If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@xenova/transformers) using:
11
  ```bash
12
- npm i @xenova/transformers
13
  ```
14
 
15
  **Example:** Perform image classification with `Xenova/convnext-base-224`.
16
 
17
  ```js
18
- import { pipeline } from '@xenova/transformers';
19
 
20
  // Create image classification pipeline
21
  const classifier = await pipeline('image-classification', 'Xenova/convnext-base-224');
@@ -23,7 +23,7 @@ const classifier = await pipeline('image-classification', 'Xenova/convnext-base-
23
  // Classify an image
24
  const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/tiger.jpg';
25
  const output = await classifier(url);
26
- console.log(output)
27
  ```
28
 
29
  ---
 
7
 
8
  ## Usage (Transformers.js)
9
 
10
+ If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@huggingface/transformers) using:
11
  ```bash
12
+ npm i @huggingface/transformers
13
  ```
14
 
15
  **Example:** Perform image classification with `Xenova/convnext-base-224`.
16
 
17
  ```js
18
+ import { pipeline } from '@huggingface/transformers';
19
 
20
  // Create image classification pipeline
21
  const classifier = await pipeline('image-classification', 'Xenova/convnext-base-224');
 
23
  // Classify an image
24
  const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/tiger.jpg';
25
  const output = await classifier(url);
26
+ console.log(output);
27
  ```
28
 
29
  ---
onnx/model_bnb4.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7000654b0724c287cd4d9a51459285e09ed257ec13fd896037811d055c1349a
3
+ size 66850966
onnx/model_q4.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1a3aab34ff1de4d2721e342afdebe79bffe97752e3e9d41b7cc2a484db8fc8f
3
+ size 72085075
onnx/model_q4f16.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1acfa5abcf79499a6e719ee967cade7cf635c498dde979350bc9f91c89592492
3
+ size 57177632
onnx/model_uint8.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:778a6a7735a59929ff2fe30348ff326d75f65c29025ba8b76cf28742e739a9f2
3
+ size 89681678