Coder Social home page Coder Social logo

epi-dlmh's People

Contributors

wuzhaobuying avatar xzenglab avatar

Stargazers

 avatar

Watchers

 avatar  avatar

Forkers

lichenbiostat

epi-dlmh's Issues

question about training time between EPI-DLMH and SIMCNN

In both papers,the GPU is 1080 GPU。
In SIMCNN paper, it says that “For the training procedures in sections 2.3.1 and 2.3.2, it took less than 1 hour (for 40 epochs) and about 2.5 hours (including 8 epochs for pretraining and 32 epochs for training), respectively.”
In EPI-DLMH paper,it says that "In training, times were roughly 180 h (SPEID), 120 h (SIMCNN) and 240 h (EPIANN), which are considerably longer than the running time of our model (72 h)."
So I want to learn why the time spending is so diffent.
And I also noticed that in your paper the epoch is 90, in SIMCNN it is 40.It about 2.25 times between them.

Another,SIMCNN is a simple CNN structure, your model structure has bi_gru and matching heuristics except cnn.obviously your model is more complex, but you spend less time. I would like to learn the reason.

Sincerely,PBC.

关于EPI-DLMH和SIMCNN的对比

您好,阅读了您的论文,对于EPI-DLMN和SIMCNN的对比,我有一些疑惑,希望您可以抽空回复,在此感激不尽。

对于训练过程,我有以下疑问:
在您的论文中,您有提到“在每个细胞系中,我们的模型训练90个EPOCH,与现有的几种方法,如SPEID,==SIMCNN==和EPIANN==一致==”。因此,我阅读了SIMCNN的训练过程,其训练过程描述是:3.使用训练集上的增强/平衡训练训练模型4或5个Epoch,以初始化模型。4.继续使用原始的不平衡训练集对模型进行训练,持续32个Epoch。另外,您的训练过程是:“4.在Daug上训练模型。5.对模型进行Dtest检验。 在每个细胞系中,模型训练90个EPOCH”。
所以请问,“与现有的几种方法,如SPEID,SIMCNN和EPIANN一致”,具体是指什么一致?

对于对比结果我也有一些疑问:
在您的论文中:
对于AUROC,SIMCNN一行中给的数据依次是:0.937,0.924,0.941,0.946,0.951,0.967
对于AUPRC,SIMCNN一行中给的数据依次是:0.786,0.642,0.791,0.773,0.805,0.887
在SIMCNN原论文中:
对于AUROC,SIMCNN和SPEID的数据都在0.9以上
对于AUPRC,SPEID的数据在0.76-0.84之间,SIMCNN的数据在0.78-0.86之间

在您的论文中,您的表述是:“为了将我们的模型与现有的其他模型(Speid,SIMCNN,EPIANN)进行比较,我们==使用相同的数据集,遵循相同的训练过程==来构建我们的深度学习模型。”,那我的问题是:
既然数据集一样,训练过程一样,那为什么在您论文中的关于以往模型的结果,部分数据会不同程度地低于原论文中的结果呢?

我想了解您对比试验的具体细节。

希望您能在百忙之中答复我的疑惑,感激不尽。

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.