JwDong2019
JwDong2019
> 这是Box-Clustering.ipynb聚类后得到的聚类中心值 > [[0.0627878 0.09495271] > [0.11484864 0.23439993] > [0.86874759 0.87105803] > [0.79309616 0.46284657] > [0.29627503 0.71413459] > [0.41545661 0.44281255] > [0.30121263 0.2348154 ] > [0.1761251 0.47797277] > [0.52819415 0.79192799]]...
> @zxy3416391 可能是因为在makefile.config 文件中把使用cudnn 注释掉了，但是这个文件好像是必须要使用cudnn 的 没有注释掉cudnn，也有这个问题，请问您知道解决方案吗？
> I have the same problem. Did you have any solution for this? Have you solved the problem？
I also encounter the problem. I have solved the problem,you can edit the 'read_clip_and_label()' function like this:" lines = open(filename, 'r') read_dirnames = [] rgb_data = [] flow_data = []...
@sudonto Thank you ,I have soved the problem by changing the input_data.read_clip_and_label(num_frames_per_clip=FLAGS.num_frame_per_clib*sample_rate) in the train function, and define the sample_rate in the train function the same as input_data python file.
@sudonto you just change the sample_rate in input_data python file? Can you tell more detailed?
> First,thanks, > How to save time by extracting optical streams from multiple videos at once? Do you know how to install opencv and get optical flow from multiple videos...
@GuangmingZhu Thank you teacher. I see the modified in your code. Now i know why the loss is Nan. Maybe it is because I forget to initializer 'kernel_initializer' and 'recurrent_initializer'...
@GuangmingZhu Yeah, my structure is 3DCNN + ConvLSTM2D. 3DCNN component is prior ConvLSTM2D. num_frames is the count of the frames selected in a video in my case. I see the...
@GuangmingZhu Oh I see. I use the ConvLSTM2D as you say. It mean the 'samples' is correspond to to 'batchsize' and 'time' is correspond to to temporal sequence length? (samples,time,...