leoluopy

chatglm-6b 最小训练显存多少？ 如果显存不够如何部分加载 chatglm-6b 的权重

2

chatglm-6b 最小训练显存多少？ 如果显存不够如何部分加载 chatglm-6b 的权重

大模型推理中这个推理引擎如何支持 lora，ptuning等私有语料训练插件后的新模型

2

如题： 大模型推理中这个推理引擎如何支持 lora，ptuning等私有语料训练插件后的新模型 有没有guide ？

how the ByteTrack works while in low frame rate 8 frames/s e.g.

2

hi , @ifzhang thanks for great works , and have u tried , how it works in low frame rate situation , 8 frames/s for example . i tried it...

do any one met this problem ? INFO:short_trian:Train Epoch: 1 [0%] INFO:short_trian:[2.9181134700775146, 4.161535739898682, 17.82921028137207, 25.843229293823242, 0.0, 2.7403459548950195, 0, 9.99875e-05] 5%|█████████▍ | 5/100 [00:09

How to add a new backend to onnx-mlir [RISC-V, and DSP SOC]

10

as we can see in the top README has explained , onnx-mlir support backend like : s390x or ppc64le or amd64 . but now i want to add new hardware...