bcmi / CaGNet-Zero-Shot-Semantic-Segmentation

Code for our ACMMM2020 paper "Context-aware Feature Generation for Zero-shot Semantic Segmentation".
233 stars 14 forks source link

Train and val data path #3

Closed andygong2018 closed 4 years ago

andygong2018 commented 4 years ago

I download the data follow the installation 3, using 'CaGNet_VOC2012_data.tar'. In train.py line 166, I print the train and val data: 微信截图_20200911102303 and the print results are: 微信截图_20200911102359 the datapath is '/home/schoudhu/zss/SBD/dataset/', how can I change the path to mine? Thanks

zhangxgu commented 4 years ago

@andygong2018 Hi, there is no need to change the path. We only use the name of the corresponding data. You can run the code by simply changing the path of VOC in the config file.

andygong2018 commented 4 years ago

@andygong2018 Hi, there is no need to change the path. We only use the name of the corresponding data. You can run the code by simply changing the path of VOC in the config file.

Thanks a lot! Yes, I have found the corresponding in the method set_files.