nerddd / ssrnet-caffe Goto Github PK

hello，would U like to share the keras2caffe scope U have modified？initial scope lacked a lot of layer，ex：multiply，tanh，merge。。。。waiting your reply and discussion，3Q！

Could you please share me? I will very appreciate.

layer: dense_1
Traceback (most recent call last):
File "convert_ssrnet.py", line 29, in
keras2caffe.convert(model, 'ssrnet.prototxt', 'ssrnet.caffemodel')
File "/home/lwp/beednprojects/keras2caffe/keras2caffe-3/keras2caffe.py", line 403, in convert
caffe_model.params[layer][n].data[...] = net_params[layer][n]
ValueError: could not broadcast input array from shape (3,160) into shape (3,250)

按照你的思路转是转出来了，但caffe模型和keras模型得到的9个值并不同，你能得到相同结果吗？

我也在用回归做人脸质量，研究了你的博客，也遇到了回归loss不降的问题。
我也是拿Face Image Quality Assessment Based on Learning to Rank生成的label。
谢谢。

Hi , thanks for the great job! Can you provide an example to deploy , please?

你好
请问转换成caffe model 后 大家都是使用什么框架进行前向推断? ncnn? Tengine?还是 opencv dnn?

你好，转化一直不成功，能不能共享一下转化完成的caffemodel和deploy.prototxt 非常感谢！

caffe model转好了，可以提供一下怎么使用的test.py吗？