couldn't parse yolov4-tiny

[mltloveyy]

Due to the different environment, I created a new project on VS2015 in windows10, and added your and tensorRT code to the project, the program was compiled and passed When I tried to load a trained model based on yolov4-tiny, there was a memory out-of-bounds problem in line 1211 of trt_utils.cpp, and I tried to use yolov4-tiny and its pre-trained model yolov4-tiny.conv.29, the same problem occurred. All prompt errors in the 11th convolutional layer https://github.com/enazoe/yolo-tensorrt/blob/322a1310adf28fb911f1e10415aa5de6432126aa/modules/trt_utils.cpp#L1211 File does not exist : D:/Code/yolo-tensorrt/configs/yolov4-tiny-kHALF-batch1.engine Loading pre-trained weights... Invalid network type Loading complete! layer inp_size out_size weightPtr (1) conv-bn-leaky 3 x 416 x 416 32 x 208 x 208 992 (2) conv-bn-leaky 32 x 208 x 208 64 x 104 x 104 19680 (3) conv-bn-leaky 64 x 104 x 104 64 x 104 x 104 56800 (4) chunk 64 x 104 x 104 32 x 104 x 104 56800 (5) conv-bn-leaky 32 x 104 x 104 32 x 104 x 104 66144 (6) conv-bn-leaky 32 x 104 x 104 32 x 104 x 104 75488 (7) route - 64 x 104 x 104 75488 (8) conv-bn-leaky 64 x 104 x 104 64 x 104 x 104 79840 (9) route - 128 x 104 x 104 79840 (10) maxpool 128 x 104 x 104 128 x 52 x 52 79840

[mltloveyy]

I tried yolov4, and the same error was reported but in the 12th convolutional layer

[Nuzhny007]

Do you change net_type in config?