ran the train script but nothing happened

imatge-upc / rvos

RVOS: End-to-End Recurrent Network for Video Object Segmentation (CVPR 2019)

https://imatge-upc.github.io/rvos/

Other

277 stars 54 forks source link

ran the train script but nothing happened #20

Closed sklf closed 4 years ago

sklf commented 4 years ago

Hello, I cloned your code and tried to train the model. But after I ran the script, nothing happened. My terminal just showed two messages. So what's the problem?

sklf commented 4 years ago

I fixed this issue. Maybe I didn't download both datasets.

Chao86 commented 4 years ago

I fixed this issue. Maybe I didn't download both datasets.

I also met the question, Do I have to download these two databases? Thanks in advance!

sklf commented 4 years ago

I fixed this issue. Maybe I didn't download both datasets.

I also met the question, Do I have to download these two databases? Thanks in advance!

Yeah, you can try to download both datasets. And make sure your GPU is empty for the training process.

Yiyiyimu commented 4 years ago

@sklf should I download all those 60+GB dataset of Youtube-VOC? Is there any part of it that we do not need to use when training? Thanks in advance!

sklf commented 4 years ago

@sklf should I download all those 60+GB dataset of Youtube-VOC? Is there any part of it that we do not need to use when training? Thanks in advance!

No, you only need to download the basic dataset ( the smaller one in the google drive). Because only basic dataset has annotations.

Yiyiyimu commented 4 years ago

@sklf should I download all those 60+GB dataset of Youtube-VOC? Is there any part of it that we do not need to use when training? Thanks in advance!

No, you only need to download the basic dataset ( the smaller one in the google drive). Because only basic dataset has annotations.

Thank you! Just to make sure, is the file name 'train.zip'?

sklf commented 4 years ago

@sklf should I download all those 60+GB dataset of Youtube-VOC? Is there any part of it that we do not need to use when training? Thanks in advance!

No, you only need to download the basic dataset ( the smaller one in the google drive). Because only basic dataset has annotations.

Thank you! Just to make sure, is the file name 'train.zip'?

train.zip and valid.zip

Yiyiyimu commented 4 years ago

@sklf should I download all those 60+GB dataset of Youtube-VOC? Is there any part of it that we do not need to use when training? Thanks in advance!

No, you only need to download the basic dataset ( the smaller one in the google drive). Because only basic dataset has annotations.

Thank you! Just to make sure, is the file name 'train.zip'?

train.zip and valid.zip

Thanks!

Chao86 commented 4 years ago

I fixed this issue. Maybe I didn't download both datasets.

I also met the question, Do I have to download these two databases? Thanks in advance!

Yeah, you can try to download both datasets. And make sure your GPU is empty for the training process.

After solver thr error ,I also met an UnboundLocalError: local variable 'seq_name' referenced before assignment in the dataset.py file, I carefuully browse the file, the variable is not assigned. What dose the variable mean? Have you encountered a similar problem?