-
I found that if I have a device on a MUSE controller that is using the ```ipcomm.tcp.client``` or ```ipcomm.tcp.client.persistent``` drivers to connect to a TCP server and I delete the device from MUS…
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
PyTorch version: 2.4.0+cu121
Is debug build: False
CUDA used to build PyTorch…
-
I upload the AMX flies and they do not show...
"Please copy some .AMX files to the USB drive :)"
I have tried a number of times from liux...
Also
Can see DSO in windows 8 that another problem need h…
NA7KR updated
10 years ago
-
### 🐛 Describe the bug
In the four variants of `nn.init.zeros_` at the bottom: the expected behavior is to see some zeros in the printed matrix. Howver, when slicing a partial matrix `[:2]`, none o…
-
### 🐛 Describe the bug
Hi @leslie-fang-intel
The compiled model's generation latency is slower than the eager mode. Regression happens from torch 0806 - torch 0807.
transformers version: 4.44.…
-
There are some unique features of AMX which don't work in other SM versions:
Reverse Grade - Reverses the judgement system, for example, what was a marvelous becomes a miss and vice versa.
More mods f…
-
`RemoteProtocolError: API communication encountered an error: peer closed connection without sending complete message body`
Full error logs are attached. Setup of the project is based on instructio…
-
### 🐛 Describe the bug
with TP
without TP
```
cat google-credentials.json
```
Empty.
`h…