You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+2-1Lines changed: 2 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -13,14 +13,15 @@ Create a `model.json` file and add your model's information to it, for example
13
13
{
14
14
"name": "mnist-example",
15
15
"version": "0.0.1",
16
+
"modelDownload": false,
16
17
"model": "MNIST.onnx",
17
18
"preprocess": "preprocess.py",
18
19
"postprocess": "postprocess.py",
19
20
"packages": ["numpy", "opencv-python"]
20
21
}
21
22
```
22
23
23
-
Where `model`, `preprocess`, and `postprocess` are all the paths to the model, preprocess, and postprocess files respectively.
24
+
Where `model`, `preprocess`, and `postprocess` are all the paths to the model, preprocess, and postprocess files respectively. If your `model.json` has `modelDownload` set to a URL instead of false you do not need to upload your model, instead you must have a web server hosting the model that workers can fetch from. See `example-server.js` for an example server. DistributiveWorkers intended to work on these slices must have the URL added to their allowOrigins object.
24
25
25
26
Packages are all of the python libraries your pre and post processing scripts require. The list of supported packages is:
0 commit comments