mcmingchang

Results 61 comments of mcmingchang

max_batch_size doesn't work as expected

I have the same problem

max_batch_size doesn't work as expected

import torch from torch2trt import torch2trt from torchvision.models.alexnet import alexnet # create some regular pytorch model... model = alexnet(pretrained=False).eval().cuda() x = torch.ones((4, 3, 224, 224)).cuda() model_trt = torch2trt(model, [x], max_batch_size=4,...

simOTA提速建议

感觉没什么变化

ConvTunerSimple_tune_and_cache.cc -> can't find suitable algorithm for 0

> 我遇到了同样的问题。FP32 在训练和推理期间都能正常工作。FP16 在训练期间工作正常，但在推理时会出现此错误。 use model.eval() and dont use fp16 dont use model.eval() and use fp16

MinkowskiConvolutionFunction cannot support multi-gpu?

same problem

about the traindata

You can retrain the ccpd dataset

about the traindata

https://github.com/mcmingchang/yolox_keypoint_segment/blob/main/exps/default/yolox_kp.py

I Can not run the demo.py

I've fixed the code

I Can not run the demo.py

you can use -d 1 -b 8，i will fix this bug today

I Can not run the demo.py

-d 1 -b 8 num_worker =4 -d 2 -b 16 num_worker =4 -d 3 -b 24 num_worker =4 -d 4 -b 32 num_worker =2 I suspect that the thread is...