shamangary/SSR-Net

Morph2 dataset website is dead

noelcodes opened this issue · 4 comments

noelcodes commented

Morph2 dataset website is dead. Do you happen to have the dataset for download?

shamangary commented

�Sorry, I cannot provide the dataset due to the morph2 dataset requires an application and some payment.
I have no idea why the dataset website is down. You have to ask the authors of the dataset.

noelcodes commented

Since morph db is dead, therefore i am trying to modify run_ssrnet.sh. I noticed u have nested for looped imdb + wiki with i,j. I am confused, because your demo is only taking either wiki OR imdb weights. I planned to cut down train time by not looping, but not sure if this mess up the model concept.

So I want to clarify, your purpose of the for-loop, is it because you are taking the last train weights from imdb for start weights for wiki, then continued loop using last wiki weight for start imdb and so on and on...? If this is so, then for loop is mandatory.

I was planning to train only this: but not sure.
python SSRNET_train.py --input ../data/imdb_db.npz --db imdb --netType1 1 --netType2 1

shamangary commented

No. The for loop is just for different hyper parameters. Different model from different hyper parameters do not take each others pretrained weight.

The paper uses --netType1 4 --netType2 4.

noelcodes commented

Any idea why?
It throw me an error when I run training without loop, just this cmd>
python SSRNET_train.py --input ../data/imdb_db.npz --db imdb --netType1 1 --netType2 1
Error:
UnicodeDecodeError: 'rawunicodeescape' codec can't decode bytes in position 838-839: truncated \uXXXX

====================================================================================================
Total params: 40,915
Trainable params: 40,531
Non-trainable params: 384

DEBUG:root:Saving model...
Traceback (most recent call last):
File "SSRNET_train.py", line 136, in
main()
File "SSRNET_train.py", line 95, in main
f.write(model.to_json())
File "C:\Users\default.LAPTOP-2CI68M4P\Anaconda3\envs\xrvision2\lib\site-packages\keras\engine\topology.py", line 2618, in to_json
model_config = self._updated_config()
File "C:\Users\default.LAPTOP-2CI68M4P\Anaconda3\envs\xrvision2\lib\site-packages\keras\engine\topology.py", line 2585, in _updated_config
config = self.get_config()
File "C:\Users\default.LAPTOP-2CI68M4P\Anaconda3\envs\xrvision2\lib\site-packages\keras\engine\topology.py", line 2322, in get_config
layer_config = layer.get_config()
File "C:\Users\default.LAPTOP-2CI68M4P\Anaconda3\envs\xrvision2\lib\site-packages\keras\layers\core.py", line 656, in get_config
function = func_dump(self.function)
File "C:\Users\default.LAPTOP-2CI68M4P\Anaconda3\envs\xrvision2\lib\site-packages\keras\utils\generic_utils.py", line 175, in func_dump
code = marshal.dumps(func.code).decode('raw_unicode_escape')
UnicodeDecodeError: 'rawunicodeescape' codec can't decode bytes in position 838-839: truncated \uXXXX