search

ceccocats / tkDNN

Deep neural network library and toolkit to do high performace inference on NVIDIA jetson platforms
GNU General Public License v2.0
717 stars 209 forks source link

Error reading file yolo4tiny/layers/c2.bin with n of float: 36864 seek: 0 size: 147456 #164

Closed winterxx closed 3 years ago

winterxx commented 3 years ago

I change the yolov4-tiny-custom.cfg input size to 608x608 and train my yolo4tiny-custom model, export weights is correct, but creat the .rt file is error by running: ./test_yolo4tiny. the error: New NETWORK (tkDNN v0.5, CUDNN v7.604) Reading weights: I=3 O=32 KERNEL=3x3x1 Reading weights: I=32 O=64 KERNEL=3x3x1 Reading weights: I=64 O=64 KERNEL=3x3x1 Error reading file yolo4tiny/layers/c2.bin with n of float: 36864 seek: 0 size: 147456

/media/xx/data1/projects/project2020/yolo/yolotools/tkDNN/src/utils.cpp:58 Aborting...

winterxx commented 3 years ago

I change the yolov4-tiny-custom.cfg input size to 608x608 and train my yolo4tiny-custom model, export weights is correct, but creat the .rt file is error by running: ./test_yolo4tiny. the error: New NETWORK (tkDNN v0.5, CUDNN v7.604) Reading weights: I=3 O=32 KERNEL=3x3x1 Reading weights: I=32 O=64 KERNEL=3x3x1 Reading weights: I=64 O=64 KERNEL=3x3x1 Error reading file yolo4tiny/layers/c2.bin with n of float: 36864 seek: 0 size: 147456

/media/xx/data1/projects/project2020/yolo/yolotools/tkDNN/src/utils.cpp:58 Aborting...

poornimajd commented 3 years ago

I change the yolov4-tiny-custom.cfg input size to 608x608 and train my yolo4tiny-custom model, export weights is correct, but creat the .rt file is error by running: ./test_yolo4tiny. the error: New NETWORK (tkDNN v0.5, CUDNN v7.604) Reading weights: I=3 O=32 KERNEL=3x3x1 Reading weights: I=32 O=64 KERNEL=3x3x1 Reading weights: I=64 O=64 KERNEL=3x3x1 Error reading file yolo4tiny/layers/c2.bin with n of float: 36864 seek: 0 size: 147456

/media/xx/data1/projects/project2020/yolo/yolotools/tkDNN/src/utils.cpp:58 Aborting...

@winterxx How did you solve it?