Closed szad670401 closed 4 years ago
Hello. I have plans to make the inference of convolution layer so fast as possible. So I will try to implement these methods when I finish current tasks connected with compatibility Synet and Caffe and Tensorflow models.
thank you. I found the opencv dnn have implemented the tensorflow importer. you can refer to its desgin. but the dnn module can't compute with low precison , therefore it run a little bit slow on ARM. I very look forward you will work on low precision computing.
I have seen this importer. It was useful in order to understand inner structure of Tensorflow. Unfortunately it is not enough in order to convert complicated models.
yes. tensorflow has two much ops. some chinese company open source some mobile tiny CNN inference framework. I hope their code will help for you.
Reference: 1.Mace 2.NCNN 3.FeatherCNN
Thank you for information. And I want to note that I have implemented Winograd 3x3 2x2 convolutions. It gives performance improvement in some cases.
A great work. I have seen you have implemented most of common layer. Have you planed to give out some benchmark tests that comparing with other frameworks?
I I periodically do performance comparison between Synet and (Darknet/Caffe/Tensorflow/Opencv:Dnn). Synet contains tests which compare Synet and Darknet. Other comparison tests are not ready to publication. I will try to do they. Now I am experementing with image format. In some cases NHWC is faster then NCHW.
Winograd's algorithms with kernel 3x3 and window 2x2, 3x3, 4x4 were implemented.
the im2col + gemm maybe a little slow.