zeakey / skeleton

Home Page:http://kaizhao.net/deepsk

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Question about running training.py

yeLer opened this issue · comments

I have done the data agumentation by the tips of your website ,and I get the result like this.
image
In this folder,i can't find a txt file named list.txt.So when I run the training.py,I meet the problem like this.

I1008 16:43:02.081221 25387 net.cpp:76] Memory required for data: 0
I1008 16:43:02.081353 25387 layer_factory.hpp:76] Creating layer data
I1008 16:43:02.081411 25387 net.cpp:111] Creating Layer data
I1008 16:43:02.081424 25387 net.cpp:434] data -> data
I1008 16:43:02.081471 25387 net.cpp:434] data -> label
I1008 16:43:02.081509 25387 image_labelmap_data_layer.cpp:40] Opening file data/SK-LARGE/list.txt
I1008 16:43:02.081531 25387 image_labelmap_data_layer.cpp:50] Shuffling data
I1008 16:43:02.081581 25387 image_labelmap_data_layer.cpp:55] A total of 0 images.
Segmentation fault (core dumped)

I find the train prtotxt file fsds_train.prototxt,the source is "data/SK-LARGE/list.txt".
image
I think the problem is missing this txt file,could you add the txt file in this repo, or give me some suggestion about list.txt file.Thank you very much!

As shown in https://github.com/SHUCV/SK-LARGE/blob/master/augmentation.m#L17train_pair.lst
is the txt you are looking for.

I get it,Thank you again!