XdpAreKid

Where can i download the ResNet18 pretrained model??

9

Hello, I can't the way to download the pretrained model of resnet18 which is converted from gluon

请教一下，调试方面的问题

1

在您的博客中，我看到有把断点打在glsl内请问有这方面的调试资料吗？？ 还有我在linux下，有没有其他的非Nsight的调试方法？？(intel显卡)

add some fuse conv with binaryop

1

add some fuse conv with binaryop

export style model error

4

hi, i use your project to export style model and face this error. /media/disk_docker/miniconda3/envs/YOLOX/bin/python /media/disk_docker/XdpWork/vkProcess/GFPGANv1.3-to-ncnn/GFPGANv1.3-to-ncnn/Export_StyleGAN2GeneratorClean_to_ONNX.py /media/disk_docker/XdpWork/vkProcess/GFPGANv1.3-to-ncnn/GFPGANv1.3-to-ncnn/gfpgan/archs/stylegan2_clean_arch.py:321: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be...

转换模型报错，如何定位

3

## error log | 日志或报错信息 | ログ ./pnnx gfpganv1_clean_model.pt inputshape=[1,3,512,512] fp16=0 pnnxparam = gfpganv1_clean_model.pnnx.param pnnxbin = gfpganv1_clean_model.pnnx.bin pnnxpy = gfpganv1_clean_model_pnnx.py pnnxonnx = gfpganv1_clean_model.pnnx.onnx ncnnparam = gfpganv1_clean_model.ncnn.param ncnnbin = gfpganv1_clean_model.ncnn.bin ncnnpy...

Conv层结果不一致

2

## error log | 日志或报错信息 | ログ value not match at c:0 d:0 h:31 w:24 expect 2.745680 but got 2.427734 test_layer_gpu failed test_layer Convolution failed use_packing_layout=1 use_fp16_packed=1 use_fp16_storage=0 use_fp16_arithmetic=0 use_shader_pack8=0...

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment： Linux 6.1.55-1-MANJARO - 版本号/Version：2.7 Paddle： PaddleOCR：manylinux_cpu_avx_mkl_gcc8.2: 2.3.2版本 问题相关组件/Related components： 增加画框操作, 画出来的框是乱的 - 运行指令/Command Code： ./build/ppocr --layout_model_dir=picodet_lcnet_x1_0_fgd_layout_cdla_infer \...