hufangjian

convalution havn't activate linear

10

你好用了你百度网盘上的yolov4.pth ,运行demo.py 报“convalution havn't activate linear“，并且无输出。

Function names and buildId information is missing for some frames due to unreadable libraries.

2

**1. 环境（environment）** - Build OS and Version: Ubuntu - RunTime DEVICE: ARM **2. Github版本** - branch： - commit(optional): **3. 编译方式(compile method)** CMake完整编译参数(full cmake arguments) **4. 编译日志(build log)** ```txt 将日志粘贴在这里 Paste...

TNN/source/tnn/device/opencl/opencl_context.cc][Line 245] save program cache failed, ret: 24577, msg: code: 0x6001 msg: check build option size error

2

**1. 环境（environment）** - Build OS and Version: Ubuntu - RunTime OS Version: Android - RunTime DEVICE: OPENCL **2. Github版本** - branch：最新今天的master版本 - commit(optional): **3. 编译方式(compile method)** CMake完整编译参数(full cmake arguments) build_android.sh...

pytoch->onnx->tensorflow how to conevrt from NCHW(onnx) to NHWC(tensorflow lite)

14

I convert a model from pytorch to onnx and than to tflite, the tflite shape is NCHW.run slowly on android , but the NHWC shape tflite is faster. the same...

yolov4转换后识别有问题。

2

您好，用您的代码转换yolov4后识别框是乱的。我把caffe（yolov4.caffemodel）和pytorch 运行(yolov4.weight)的网络输入数据设置成全1，三层输出得featureMap caffe 和 pytorch完全不一样。

yolov4测试精度比较低

4

你好，我这边从darknet转到caffe后，测试时发现精度比较低。有一些莫名其妙的框。

转换后，少yolo layer上一层的 19x19 38x38 76x76的卷积。

TypeError: show_results() got an unexpected keyword argument 'show'

3

I run： python tools/test.py configs/centerpoint/centerpoint_01voxel_second_secfpn_dcn_circlenms_4x8_cyclic_20e_nus.py work_dir/centerpoint/centerpoint_01voxel_second_secfpn_dcn_circlenms_4x8_cyclic_20e_nus.pth --show --show-dir eval_results/ get this error: Traceback (most recent call last): File "tools/test.py", line 217, in main() File "tools/test.py", line 187, in main outputs...