caffe

caffe copied to clipboard

Documentation of output blob?

1

``` modelFile = "res10_300x300_ssd_iter_140000_fp16.caffemodel" configFile = "deploy.prototxt" net = cv2.dnn.readNetFromCaffe(configFile, modelFile) blob = cv2.dnn.blobFromImage(frame, 1.0, (300, 300), [104, 117, 123], False, False) net.setInput(blob) detections = net.forward() ``` ``` # detections.shape...

samuelwhiskeyjohnson

CUDA 13.0 has some incompatible API [changes](https://developer.nvidia.com/blog/whats-new-and-important-in-cuda-toolkit-13-0/#changes_to_cudadeviceprop).

dbermond

Remove obsolete loss layer check

5

With the `n`d blob update, it's no longer true that the `num` dimension has to be the same for both bottoms of every loss layer (e.g., softmax loss layer with...

longjon

De-duplicate Engine Tests

4

A not insignificant amount of the cuDNN integration diff was testing, but these tests are mostly mirrors of the Caffe standard tests with the engine swapped between CAFFE and CUDNN....

shelhamer

Remove unused codes

1

size_, pre_pad_, alpha_, beta_, k_ is defined in the base LRNLayer class.

t-ubukata

Error found when tried to run OpenPose 1.7.0

2

Code: !cd openpose/build && make -j`nproc` Link: https://colab.research.google.com/github/miu200521358/motion_trace_colab/blob/master/OpenposeDemo.ipynb In file included from /content/openpose/3rdparty/caffe/include/caffe/util/device_alternate.hpp:40, from /content/openpose/3rdparty/caffe/include/caffe/common.hpp:19, from /content/openpose/3rdparty/caffe/include/caffe/blob.hpp:8, from /content/openpose/3rdparty/caffe/include/caffe/layers/cudnn_tanh_layer.hpp:6, from /content/openpose/3rdparty/caffe/src/caffe/layers/cudnn_tanh_layer.cpp:4: /content/openpose/3rdparty/caffe/include/caffe/util/cudnn.hpp: In function ‘const char* cudnnGetErrorString(cudnnStatus_t)’: /content/openpose/3rdparty/caffe/include/caffe/util/cudnn.hpp:21:10: warning: enumeration...

c0316yuan