Code Monkey home page Code Monkey logo

cyclical_annealing's People

Contributors

chunyuanli avatar haofuml avatar

Stargazers

avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar

Watchers

avatar avatar avatar

Forkers

namisan yyht qianmuluo zheng-yanan ecvgit sunset-clouds jihunchoi aprilxiaoyanliu mjchku kj141 whoyouwith91 soumyatghosh bsridatta sachinsavale2007 liam0949 trellixvulnteam rafaperpi nkzhangheng

cyclical_annealing's Issues

IndexError: index 1100 is out of bounds for axis 0 with size 1100

Getting below error
IndexError: index 1100 is out of bounds for axis 0 with size 1100
at below line
L[int(i + c * period)] = 1.0 / (1.0 + np.exp(- (v * 12. - 6.)))

it seems that we missed to add second condition at below line in function 'frange_cycle_linear' & 'frange_cycle_sigmoid'
while v <= stop:
should be
while v <= stop and (int(i + c * period) < n_epoch):

word dropout

Are the results in the paper with no word dropout in the decoder?

About implementation

Hi. I am trying to run the source code based on Pytorch 0.4.1.

I encountered a problem when the used model is the sa-vae:

cudnn RNN backward can only be called in training mode

I noticed that you used Pytorch 0.2. Did you try the latest version of Pytorch?

About the design of Cyclical Annealing Schedule

Hi~
have you tried any other kind of annealing schedule? e.g. cyclical cosine or cyclical exponential.
I'm just wondering why you chose this linear one. just for simplicity? or this is actually the best?

control learning rate in train_text_cyc_ptb.py example

Could you explain why you control learning rate in optimizer in the example?

tmp = float((epoch-1)%args.cycle)/args.cycle cur_lr = args.lr*0.5*(1+np.cos(tmp*np.pi)) for param_group in optimizer.param_groups: param_group['lr'] = cur_lr

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.