-
这个训练框架无法在BM1684X芯片上面跑吧?
-
Mautic stored the countries in plain text (not iso3)
I wrote a small wrapper to covert the iso3 to the mautic plain text equivalent
` // based on : app/bundles/CoreBundle/Assets/json/countries.…
robov updated
1 month ago
-
In the GTC talk, it is mentioned that the fp8 quantized weights are precomputed (cached) only for microbatch case such as pipelining.
However, why not just write out fp8 quantized weights during th…
-
**Describe the bug**
A clear and concise description of what the bug is.
**To Reproduce**
Steps to reproduce the behavior:
1. In TestForm Start the Ribbon option.
2. Click the QAT Dropdown menu…
-
### 🐛 Describe the bug
I tried to load from checkpoint using the file initially downloaded by the models.get() call. It looks like there's a leftover TODO in the repo code that's a reminder to remove…
-
## Description
I have seen two quantization librares built by Nvidia: a [TRT modelopt](https://github.com/NVIDIA/TensorRT-Model-Optimizer) and a[ pytorch-quantization](https://docs.nvidia.com/d…
-
Modern kernels provide hardware TLS transmission offload, so we need to setup TLS offloading for a socket after TLS handshake and implement zero-encryption (with necessary memory provisions) to let a …
-
### What happened?
We are interested in running Kepler on an ARM Ampere Altra Max machine (BM). We managed to successfully build the Kepler image from the Dockerfiles available in the build/ folder o…
-
@drmint80 , isiqlandir bizleri zehmet olmasa. Harada hansini istifade etmak yaxsidir? Use case-lerle soylesen ela olardi
-
求助!把examples/quantization/onnx_qat.py中的模型替换成自己的模型时报错,代码如下
[onnx_qat_wt.zip](https://github.com/user-attachments/files/15511238/onnx_qat_wt.zip)
模型就是whisper-tiny的encoder加了自己的分类器,为了方便直接把所有类定义都放在同一个文件了…