-
trying to quantize and no model is generated
my hardware is amd
```python quantize.py --checkpoint_path checkpoints/$MODEL_REPO/model.pth --mode int8
Loading model ...
Quantizing model weights f…
-
Hey, I'm trying to train my custom dataset using mmseg. I'm simply trying to segment the foreground object from the image. I have already trained this model on my own YOLACT repo and it works well the…
-
A previous thought that the AC dongles were under powered and that's what gave them such poor performance/functionality
Found this reference on the internet:
>I just had the idea to set max_usb_cu…
-
Hi,
trying to apply the GstSatbilizer plugin to a bunch of jpg images, I got the following error:
ERREUR : de l'élément /GstPipeline:pipeline0/flow_finder+OpticalFlowFinder:flow_finder+opticalflowfind…
-
If I block this code, there's no problem, so how do I fix it? I used it on the android side and wanted to train my model to make the substitution
```
DenoiseState *st;
st = rnnoise_create…
-
_I'm splitting out the various discussions from #29 so they're easier to keep track of separately. Sorry some of the formatting isn't as pretty, I opted for clarify in who said what. Please don't hesi…
ssfrr updated
6 years ago
-
Models with uint8 inputs are not supported by TRT execution provider. This is fine, but the node should fallback to CUDA, instead of failing to to the conversion in onnxruntime.
**System informat…
-
Not an actual issue. Just storing notes on various components work.
----
When I ran my test task, I get this log:
```
> python tools/joey.py test --no-display --initial-delay 0s --log-level=debu…
-
```
Is not possible convert completely a video file format MATROX, because the
output file duration is smaller than the input file.
I noticed that in the input description the duration of the video …
-
I wonder how to tackle the following problem, which arise when I train on batches of padded sequences:
- [relative positional encoding](https://github.com/espnet/espnet/blob/master/espnet/nets/pytorc…