Unable to run example code - TypeError('require string label')

[bohoro]

Hi,

I am unable to run the first hyperas example. Specifically this line:

>>> model.add(Dropout({{uniform(0, 1)}}))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/brohoro/Documents/PyProjects/envs/tensorflow/lib/python2.7/site-packages/hyperopt/pyll_utils.py", line 21, in wrapper
    raise TypeError('require string label')

The example without hyperas runs fine.

Here is my env

Python 2.7.13
Name: Keras
Version: 2.0.4
Name: hyperas
Version: 0.3
Name: hyperopt
Version: 0.1

I have tried this both via Jupyter and via the python interpreter.

Thanks

[maxpumperla]

Hi @bohoro, does it work if you install hyperas from master? It's been ages since I got around to release a new version.

[hi-ilkin]

There is also a problem with comments, try again after deleting comments in your code

[maxpumperla]

@huseynlilkin while this is absolutely true, I don't think it relates to this issue.

[nchuramani]

Hi @maxpumperla , I am facing the same problem with uniform and choice. I tried installing hyperas from master i.e. hyperas 0.4 but I am still facing the same issue.

[maxpumperla]

sorry to hear that. will look into the issue once back from vacation

[MMostavi]

Hi
I have the same issue with choice and uniform. Please if anyone has any suggestions, let me know!

[aydindemircioglu]

same for me, though i did not use the example but hacked it directly into my unet network.
btw, is it supported to have images as labels?

[rcox771]

I had the same error earlier today, but mine was caused by not using double curly braces around my choices. Do you have a spot in your code where you're missing an extra set of braces?

before:
model.compile(loss='categorical_crossentropy', optimizer={choice(['rmsprop', 'adam', 'sgd'])}, metrics=['accuracy'])

after:
model.compile(loss='categorical_crossentropy', optimizer={{choice(['rmsprop', 'adam', 'sgd'])}}, metrics=['accuracy'])

[aydindemircioglu]

ok, at least in my case i realized that i call my network directly, did not see that (stupid me).
but i have still problems, a) i use python 2 and "from future import division", which needs to
be handled in extract_imports in utils.py like if 'print_function' in line or 'import division' in line:
b) i still get an error like

Traceback (most recent call last):
File "./hyperopt16bit.py", line 455, in
trials=Trials(15))
File "/usr/local/lib/python2.7/dist-packages/hyperas/optim.py", line 67, in minimize
verbose=verbose)
File "/usr/local/lib/python2.7/dist-packages/hyperas/optim.py", line 133, in base_minimizer
return_argmin=True),
File "/usr/local/lib/python2.7/dist-packages/hyperopt/fmin.py", line 307, in fmin
return_argmin=return_argmin,
File "/usr/local/lib/python2.7/dist-packages/hyperopt/base.py", line 635, in fmin
return_argmin=return_argmin)
File "/usr/local/lib/python2.7/dist-packages/hyperopt/fmin.py", line 314, in fmin
pass_expr_memo_ctrl=pass_expr_memo_ctrl)
File "/usr/local/lib/python2.7/dist-packages/hyperopt/base.py", line 786, in init
pyll.toposort(self.expr)
File "/usr/local/lib/python2.7/dist-packages/hyperopt/pyll/base.py", line 715, in toposort
assert order[-1] == expr
IndexError: list index out of range

will see into it, what i am doing wrong.

[aydindemircioglu]

and the other thing is that my code was calling a function to create the model, like

def getUNet(): 
...

def trainStep():
    model = getUNet()

now the issue is fixed on my side.

[hacheemaster]

Any update on this issue? I'm still getting the 'require string label' error...

[nchuramani]

HI,
For me, only the uniform gives me the error now. I removed uniform and it worked. I have the updated versions of keras and hyperopt and the hyperas version from git master.

[shadiakiki1986]

You can't execute the model creation code directly in python. You need to wrap your code in a def create_model(...): ... function, and then call it from optim.minimize(model=create_model,... like in the example in the readme.

My understanding is that the reason for this is that hyperas works is by doing template replacement of everything in the {{...}} into a separate temporary file, and then running the model with the replaced braces.

[kesarianubhav]

I have a function named model wherein I am creating layers,fitting to data . And i am calling that fnc from opt.minimize . But i am Still getting this error !! Any clues ??

[mbfair]

Hello, I realize this is an older thread, but I am getting the same issue as @aydindemircioglu:

assert order[-1] == expr
IndexError: list index out of range

Do you have insight on what caused / solved this issue? TIA

[jcatanza]

Why is the issue closed?

It doesn't seem to be resolved. I (and others) still cannot run the example code without getting the 'TypeError: require string label' at the line

I have installed the latest versions of hyperopt and hyperas under python 3.6

model.add(Dropout({{uniform(0, 1)}}))

If the issue is resolved, can someone please describe how they are able to run this code without getting the TypError?

[shadiakiki1986]

Where's your code?

[JonnoFTW]

I ran the simple.py example just now and it worked fine for me.

Steps:

git clone https://github.com/maxpumperla/hyperas
cd hyperas
python3 -m venv ./venv
source venv/bin/activate
pip install -e .
pip install tensorflow
cd examples
# edit simple.py line 60 to epochs=1
python simple.py