Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

A little questuon about ENAS #4

Open
ChaliceCooper opened this issue Jan 12, 2020 · 2 comments
Open

A little questuon about ENAS #4

ChaliceCooper opened this issue Jan 12, 2020 · 2 comments

Comments

@ChaliceCooper
Copy link

hi,i'm new in your project,and it is the first time i ask question on github,Please bear with me if there are any mistakes. I find something may be wrong in your code about ENAS.
when i finish a code about ENAS by imitating your "main_tensorflow.py" and i run it and find some error.

Traceback (most recent call last):

File "C:/Users/JssI/Desktop/deep_architect-masterPZJ/ENAS.py", line 322, in main()

File "C:/Users/JssI/Desktop/deep_architect-masterPZJ/ENAS.py", line 312, in main inputs, outputs, _ ,searcher_eval_token = searcher.sample()

File "C:\Users\JssI\Desktop\deep_architect-masterPZJ\dev\enas\searcher\enas_searcher.py", line 104, in sample outputs, list(hs.values()))):
NameError: name 'hs' is not defined

so i click into the "enas_searcher.py" i find that there is one variable is not declared which named "hs"

`

 def sample(self):
    arc = self._sample()
    idx = 0
    hyp_values = {}
    for i in range(1, self.num_layers + 1):
        hyp_values['op_' + str(i)] = arc[idx]
        idx += 1
        for j in range(i - 1):
            hyp_values['H.skip_' + str(j) + '_' + str(i) + '-0'] = arc[idx]
            idx += 1

    inputs, outputs = self.search_space_fn()
    vs = []
    for i, h in enumerate(
            unassigned_independent_hyperparameter_iterator(
                outputs, list(hs.values()))):
        if h.get_name() in hyp_values:
            v = h.vs[hyp_values[h.get_name()]]
            h.assign_value(v)
            vs.append(v)
        else:
            v = random_specify_hyperparameter(h)
            vs.append(v)
    return inputs, outputs, vs, {'arc': arc}

`

What you do is a really nice work and i want to use your code to finish my experiment during my holiday, i will very very appreciate you if you could help me solve the problem. Thank u!

@ChaliceCooper
Copy link
Author

ChaliceCooper commented Jan 12, 2020

by the way , could u write a code to call the enas as a simple example ? Thank u!!!!

@negrinho
Copy link
Owner

I can eventually look at this, as we want to have ENAS implemented. Unfortunately, the code in the dev folder is highly experimental. It is kept there mostly as a reference for when we get around making it part of the main folder. Currently, I can only provide guidance on how to implement ENAS in the current codebase (e.g., where to look to support the operations needed for ENAS). With my current availability, it will be one or two weeks until I can look at this myself (and I'm not making any hard promises right now).

That being said, if you decide to implement ENAS yourself in DeepArchitect, I can provide support.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants