Hi Fred,
I uninstalled theano, then, verified that import command failed. Next,
I installed the latest development version and tried my program. It failed at
the same line with the same error.
My program implements a hybrid DBN-RNN network. The line that failed is
when gradient for the combined network is being formed. The error happened
when computing the outputs of the scan operator for the RNN network. It seems
to have something to do with the derivative of the output of the scan operator
not having the right broadcastable.
This does not happen when I train the only the RNN. In this case, the
derivative of the scan operator has the right broadcastable.
Thanks a lot.
Chi
From: [email protected] [mailto:[email protected]] On
Behalf Of Frédéric Bastien
Sent: Tuesday, November 15, 2016 3:11 PM
To: theano-users
Subject: Re: [theano-users] TypeError: Cannot convert Type TensorType(float64,
3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False,
True, False)). You can try to manually convert Subtensor{:int64:}.0 into a
TensorType(float64, (False, Tr...
To be safer, uninstall Theano a few time before and make sure you can't import
it before reinstalling.
Le 15 nov. 2016 16:20, "Pascal Lamblin"
<[email protected]<mailto:[email protected]>> a écrit :
On Tue, Nov 15, 2016, Chi Ku wrote:
> Hi Pascal,
>
> How would I find out which version of Theano I have?
You can try to print theano.version.version, that should be generated when you
call "pip install"
> Can I use the following command to install the latest development version?
>
> <sudo> pip install <--user> <--no-deps>
> git+https://github.com/Theano/Theano
I think so.
>
> Thanks.
>
> Chi
>
>
>
>
>
>
> -----Original Message-----
> From: [email protected]<mailto:[email protected]>
> [mailto:[email protected]<mailto:[email protected]>]
> On Behalf Of Pascal Lamblin
> Sent: Monday, November 14, 2016 12:37 PM
> To: [email protected]<mailto:[email protected]>
> Subject: Re: [theano-users] TypeError: Cannot convert Type
> TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type
> TensorType(float64, (False, True, False)). You can try to manually convert
> Subtensor{:int64:}.0 into a TensorType(float64, (False, Tr...
>
> Which version of Theano are you using?
> If you are using the 0.8.2 release, can you try the latest development
> version?
>
> On Sat, Nov 12, 2016, [email protected]<mailto:[email protected]> wrote:
> >
> > I got the following errors when calling tensor.grad() to compute the
> > symbolic gradient of the finetune_cost of a hybrid DBN-RNN model.
> > I tried changing the way this expression is formed in several ways without
> > any success. I need help from experts. A tar file of the source code
> > and data is attached here.
> >
> > The pretraining code for file hybrid_array.py ran for 0.48m
> > ... getting the finetuning functions
> > Traceback (most recent call last):
> > File "/usr/lib/python2.7/pdb.py", line 1314, in main
> > pdb._runscript(mainpyfile)
> > File "/usr/lib/python2.7/pdb.py", line 1233, in _runscript
> > self.run(statement)
> > File "/usr/lib/python2.7/bdb.py", line 400, in run
> > exec cmd in globals, locals
> > File "<string>", line 1, in <module>
> > File "hybrid_array.py", line 494, in <module>
> > test_DBN()
> > File "hybrid_array.py", line 412, in test_DBN
> > learning_rate=finetune_lr
> > File "hybrid_array.py", line 253, in build_finetune_functions
> > gparams = T.grad(self.finetune_cost, self.params)
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 561, in grad
> > grad_dict, wrt, cost_name)
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1324, in _populate_grad_dict
> > rval = [access_grad_cache(elem) for elem in wrt]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_gra term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 973, in access_term_cache
> > output_grads = [access_grad_cache(var) for var in node.outputs]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1279, in access_grad_cache
> > term = access_term_cache(node)[idx]
> > File "/usr/local/lib/python2.7/dist-packages/theano/gradient.py", line
> > 1113, in access_term_cache
> > input_grads = node.op.grad(inputs, new_output_grads)
> > File
> > "/usr/local/lib/python2.7/dist-packages/theano/scan_module/scan_op.py",
> > line 2523, in grad
> > outputs = local_op(*outer_inputs)
> > File "/usr/local/lib/python2.7/dist-packages/theano/gof/op.py", line 611,
> > in __call__
> > node = self.make_node(*inputs, **kwargs)
> > File
> > "/usr/local/lib/python2.7/dist-packages/theano/scan_module/scan_op.py",
> > line 430, in make_node
> > new_inputs.append(format(outer_seq, as_var=inner_seq))
> > File
> > "/usr/local/lib/python2.7/dist-packages/theano/scan_module/scan_op.py",
> > line 422, in format
> > rval = tmp.filter_variable(rval)
> > File "/usr/local/lib/python2.7/dist-packages/theano/tensor/type.py", line
> > 233, in filter_variable
> > self=self))
> > TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable
> > Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)).
> > You can try to manually convert Subtensor{:int64:}.0 into a
> > TensorType(float64, (False, True, False)).
> > Uncaught exception. Entering post mortem debugging
> > Running 'cont' or 'step' will restart the program
> > >
> > /usr/local/lib/python2.7/dist-packages/theano/tensor/type.py(233)filter_variable()
> > -> self=self))
> >
> >
> >
> > --
> >
> > ---
> > You received this message because you are subscribed to the Google Groups
> > "theano-users" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to
> > [email protected]<mailto:theano-users%[email protected]>.
> > For more options, visit https://groups.google.com/d/optout.
>
>
>
> --
> Pascal
>
> --
>
> ---
> You received this message because you are subscribed to a topic in the Google
> Groups "theano-users" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/theano-users/8uSF6ub-drA/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected]<mailto:theano-users%[email protected]>.
> For more options, visit https://groups.google.com/d/optout.
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
> "theano-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to
> [email protected]<mailto:theano-users%[email protected]>.
> For more options, visit https://groups.google.com/d/optout.
--
Pascal
--
---
You received this message because you are subscribed to the Google Groups
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
[email protected]<mailto:theano-users%[email protected]>.
For more options, visit https://groups.google.com/d/optout.
--
---
You received this message because you are subscribed to a topic in the Google
Groups "theano-users" group.
To unsubscribe from this topic, visit
https://groups.google.com/d/topic/theano-users/8uSF6ub-drA/unsubscribe.
To unsubscribe from this group and all its topics, send an email to
[email protected]<mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout.
--
---
You received this message because you are subscribed to the Google Groups
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
