Hyperparameter file for DLKcat

[wangjingjing52]

Hello Yuan,

I am Jingjing Wang, a PhD from the Central South University and research on enzyme activity prediction. I have the honor to learn your work DLkcat, which is a very valuable work for reference. I am trying to run the “run_model.py” under the DeeplearningApproach module. However, the value of the hyperparameter is missing such as "ValueError: not enough values to unpack (expected 14, got 0)" in line 184. Could you please share the Hyperparameter file or help me find it in somewhere? Thank you very much!

[le-yuan]

Hello Jingjing,

Thanks for your interest about our work. If you want to run the python script "run_model.py",

I would suggest you to write a small bash script like that:

DATASET=all
radius=2
ngram=3
...... 

python run_model.py $DATASET $radius $ngram $other hyperparameters

Then you can run the small bash script with various hyperparameters to try the model training process.

[wangjingjing52]

Hello Yuan,

Thank you very much！And apologies for my late reply！

Yours Sincerely， Jingjing Wang

@. @.

 

------------------ 原始邮件 ------------------ 发件人: "SysBioChalmers/DLKcat" @.>; 发送时间: 2023年4月2日(星期天) 凌晨5:18 @.>; @.**@*.**@*.***>; 主题: Re: [SysBioChalmers/DLKcat] Hyperparameter file for DLKcat (Issue #18)

Hello Jingjing,

Thanks for your interest about our work. If you want to run the python script "run_model.py",

I would suggest you to write a small bash script like that: DATASET=all radius=2 ngram=3 ...... python run_model.py $DATASET $radius $ngram $other hyperparameters
Then you can run the small bash script with various hyperparameters to try the model training process.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[wangjingjing52]

Dear Yuan，

Thank you very much for your patience in answering my questions last time. May I proceed to ask you a further question? We would greatly appreciate your patience and understanding!

According to our understanding of the paper (Deep learning-based Kcat prediction enables improved enzyme-constrained model reconstruction), DLKcat uses 80% of the dataset as the training set. After the model is trained, the dataset is divided into two test sets, such as wild-type and mutant, to test the model respectively. We would like to know how you consider the data duplication between the test sets and the training set. Thank you!

 

Best Wishes

 

Jingjing Wang

@. @.

 

------------------ 原始邮件 ------------------ 发件人: "SysBioChalmers/DLKcat" @.>; 发送时间: 2023年4月2日(星期天) 凌晨5:18 @.>; @.**@*.**@*.***>; 主题: Re: [SysBioChalmers/DLKcat] Hyperparameter file for DLKcat (Issue #18)

Hello Jingjing,

Thanks for your interest about our work. If you want to run the python script "run_model.py",

I would suggest you to write a small bash script like that: DATASET=all radius=2 ngram=3 ...... python run_model.py $DATASET $radius $ngram $other hyperparameters
Then you can run the small bash script with various hyperparameters to try the model training process.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>