wrong model folders

[harshitwalia]

error in line 76 to 81 in master/training_and_testing/SSRNET_train.py
if db_name == "wiki":
weight_file = "wiki_models/"+save_name+"/"+save_name+".h5"
model.load_weights(weight_file)
elif db_name == "imdb":
weight_file = "imdb_models/"+save_name+"/"+save_name+".h5"
model.load_weights(weight_file)
elif db_name == "morph":
weight_file = "morph_models/"+save_name+"/"+save_name+".h5"
model.load_weights(weight_file)

[shamangary]

Hello. It is not wrong.

As we point out in our paper section 4.3,
"Since the IMDB dataset is much larger, we trained our model on IMDB first and used the trained model as the starting point for training on the WIKI dataset."

In our section 4.4, "Before training with MORPH2, we used the IMDB-WIKI dataset for pre-training."

When we want to train morph2, we use the trained model from WIKI as a starting point for the new training since the model already contain the information of IMDB-WIKI.

[harshitwalia]

Have you prepared any notes for this project. Can you share them with me. It will be of great help.

…

-- Thanks & Regards Harshit Walia 9899644816 Go Green - "Don't print this mail unless it is absolutely necessary " Save Earth – Use Recycled Products” Disclaimer: The information in this mail is confidential and is intended solely for addressee. Access to this mail by anyone else is unauthorised. Copying or further distribution beyond the original recipient may be unlawful. Any opinion expressed in this mail is that of sender .The e-mail has been successfully scanned for presence of virus. It is recommended to scan the attachment before launching.

On Thu 5 Jul, 2018 5:31 pm Tsun-Yi Yang, ***@***.***> wrote: Hello. It is not wrong. As we point out in our paper section 4.3, "Since the IMDB dataset is much larger, we trained our model on IMDB first and used the trained model as the starting point for training on the WIKI dataset." In our section 4.4, "Before training with MORPH2, we used the IMDB-WIKI dataset for pre-training." When we want to train morph2, we use the trained model from WIKI as a starting point for the new training. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AT57cibsEKev_1gVkkgLXT4XXRXHfxC4ks5uDgAigaJpZM4VDubB> .

[shamangary]

https://github.com/shamangary/SSR-Net/blob/master/ijcai18_ssrnet_pdfa_2b.pdf

We only have our paper and the Readme file in this repository for now. Thank you for your appreciation.