-
**Test Env:**
Chromium Version: nightly build 75.0.3739.0 (1a2331d)
Platform: Linux/Windows/macOS
**Expected Result:**
The DeepLab(without Atrous) performance should be improved.
**Actual Res…
-
Hi folks awesome work with hitting native performance via web.
Coming to my question, To use webml, can we just use the polyfill? it doesn't seem to be working though(Should we build the webml …
-
As title, is there any sample code to show how to
1. load model
2. inference
3. measure time
I want to compare the performance between this proposal (webml) and onnx.js
-
**Chromium build version:** https://github.com/otcshare/chromium-src/commit/f456c6d52199f5c81560e6671a7b9a496d160e6b
**Actual Result:**
MKLDNN backend doesn't work for macOS:
```
$ ./f456c6d…
-
According to [Apple* Machine Learning on Intel® Processor Graphics](https://software.intel.com/en-us/articles/apple-machine-learning-on-intel-processor-graphics), the MobileNet inference performance i…
-
**Test Env:**
Chromium Version: nightly build 70.0.3503.0 (**a5a8547**)
Platform: Mac OS 10.13.6
**Expected Result:**
No crash
**Actual Result:**
```
[8960:775:0130/100151.412558:ERROR:nati…
-
**Test Env:**
Chromium Version: nightly build 75.0.3739.0 (c623981)
Platform: macOS 10.13.6
**Expected Result:**
The example should work when only choosing WebML backend.
**Actual Result:**
…
-
#### feature request
Mediatek p90, kirin 980 are shipping something they are calling APU and NPU(along with GPU), is there a way TF.js can leverage that?
https://www.mediatek.com/blog/googles…
-
**Test Env:**
Chromium Version: nightly build 75.0.3739.0 (c623981)
Platform: Windows (Dell XPS 13, GPU: Intel Graphics 620, CPU: Intel i5-8250U), Android (Google Pixel 2 XL, GPU: Qualcomm Adreno 54…
-
I think this issue is best demonstrated by the screenshot comparing Edge on the left and Firefox on the right. Observe the bad layout of the numbers on the left.
![image](https://user-images.github…