Jintao Meng
Jintao Meng
Hello Ghimiredhikura, The code support ARMv7, but we did not provide ARMv7 toolchain for users in the master branch. If you are using Raspberry Pi, it is actually a 64-bit...
Hi Dele, Can you use squeezenet v1.1 first, the model can be download here: http://hpcc.siat.ac.cn/jintao/feathercnn/feathermodels/squeezenet. I am fixing the v1.0 model and will close this issue afterward. Thanks, Jintao
Input_size should be 224 * 224 * 3, it has been fixed. The test program can take several lines of 224 * 224 * 3 floats as its input, here...
您好， 编译feather_convert_caffe使用的脚本是： https://github.com/Tencent/FeatherCNN/blob/booster/tools/build.sh， 这个脚本需要提前您安装protobuf。 feather_convert_caffe可以把caffe模型转换为feathermodel, 具体请参考其命令行参数。 FeatherCNN内部有Layer Fusion，它会把ReLU， Batchnorm， Scale等可以合并到其他Layer里面。所以Layer.size只有33层。 如果您不愿意合并可以将https://github.com/Tencent/FeatherCNN/blob/booster/src/net.cpp 里面第302行到344行注释掉。 金涛
@helloearth012 你可否把model发我看一下？ 我看能不能重现segmentation fault.
@yangyangv8 如果你们的caffemodel需要保密，你这边可否把prototxt发我看一下。
I find my libcaffe2.so is located in /usr/local/lib/python2.7/dist-packages/torch/lib. What should we do to make your project successfully?
Is there still anyone maintaining this project?
Caffe2's Python API works fine, but the C++ API definitely not. MACE's benchmark did not include Caffe2 into their benchmark, I guess they may failed to do so.
There are so many bugs in ChatTTS.