-
I was trying to train real-time segmenters on ADE20k. However, models such as ICNet, mobilenet_v3, fast scnn gave me `CUDA error: an illegal memory access was encountered` error. The full trace is att…
-
I have trained a segementation model on my personal computer with GPU. However I need to depoly the segmentation module to the server which has no GPU. Since the server has no GPU, I can only install …
-
Hello, I was using MMDeploy to convert the fastscnn pytorch into a TensorRT engine, which works very well. Testing the TRT model on the MMDeploy tools/test.py results in 10 tasks/second, while testing…
-
环境：Win10 64位，Python3.8.5，PaddlePaddle-gpu 2.3.3，PaddleSeg2.6，CUDA_10.1.105_418.96_win10，CUDNN 7.6.5.32
———————————————————————————
参考执行步骤：https://github.com/PaddlePaddle/PaddleSeg/blob/release/2.6/d…
-
Hello, I found a problem while training my own dataset with mmsegmentation. The input channel of the network is 3, which can be trained normally. However, when I modify the number of input channels of…
-
环境：windows10 64位 cpu部署 paddlelite 2.10rc0 paddlepaddle-gpu 0.0.0.post102 paddleseg 2.3.0
模型：paddle seg 下 fast_scnn stdc1 等
问题：训练好的模型，opt转nb未报错，推理时失效。具体为无法检测出待测物体、分割区域混乱、分割区域周期样重复。对训练集的图片进…
Nonmy updated
2 years ago
-
请问fast scnn模型有paddle lite框架 c++的推理demo吗
-
Hi, I have a general question:
I want to fine-tune the Fast-SCNN model trained in this repo.
Before I started my own training, I wanted to try if the ONNX export works with the pretrained model, …
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. The bug has not been fixed in the latest version.
…
-
CCF BDCI中训练
数据集样本类别数目为7，外加未标注
yaml文件中设置样本类数目为7，经训练，除了unet，deeplab、fast-scnn-pet、hrnet、训练后验证得到的第六个IOU全部为0，如下所示，请问是要把样本数目类别设置成8吗？如果不是，问题出在哪儿？
epoch 58:
[EVAL]#image=2919 acc=0.7806 IoU=0.3270
[EVAL…