-
Are there any runnable demos of using Sparse-QAT/PTQ (2:4) to accelerate inference, such as applying PTQ to a 2:4 sparse LLaMA for inference acceleration? I am curious about the potential speedup rati…
-
Thanks for you repo!
I'm tring to convert pth to onnx. But When I use "python qat/export-camera.py --ckpt=model/resnet50/bevfusion-det.pth --fp16".
I get a error "model = torch.load(args.ckpt).modu…
-
libqat.so is dynamic linked with openssl library, but how to static linked with openssl library?
I changed
MAKE = perl $(with_openssl_dir)/util/mkerr.pl -conf e_qat.ec \
-rebuild -reindex …
-
I trained my model on my own dataset following the guidance from qat_ptq_yolo_nas.md. The model works fine in non-QA situations, but after applying QAT, it fails to detect any objects.
-
Here is the last part of the error from build:
make[4]: Entering directory `/usr/src/kernel/linux-stable'
Building for linux_2.6
CC [M] /usr/src/kernel/drivers/os-tools/qat/qat-1.7/quickassist/l…
-
### 🐛 Describe the bug
import numpy as np
import torch
import os
import torch
# Might not be necessary
os.environ.setdefault("KMP_DUPLICATE_LIB_OK", "True")
#import pytorch_GRU
#import qua…
-
QAT_Engine-0.6.19 ，QAT driver：QAT.L.4.23.0-00001，OpenSSL 1.1.1w
one request print one log into logs/error.log:
2024/04/16 19:20:37 [alert] 38566#0: *408 async del conn epoll_ctl(2, 21) failed (9…
-
I'm building QAT engine twice with:
```
RUN cd /QAT_Engine && \
./autogen.sh && \
./configure \
--enable-ipsec_offload \
--enable-multibuff_offload \
--with-multibuff_inst…
mythi updated
3 years ago
-
I was debugging what is wrong with the engine .so when openssl returns:
```
# openssl engine -c -t -vvv qat-hw
140402102523712:error:260B606D:engine routines:dynamic_load:init failed:../crypto/en…
mythi updated
2 years ago
-
Hi,
I have the C62X QAT coprocessor on my server -
$ lspci | grep -i "quick"
da:00.0 Co-processor: Intel Corporation C62x Chipset QuickAssist Technology (rev 04)
db:00.0 Co-processor: Intel Co…