Hi,
When you want to create a theano variable over a numpy array, you could try
using shared variable, something like
In [1]: import theano
In [2]: import numpy as np
In [3]: sh1 = theano.shared(np.asarray(0.12))
In [4]: sh1.eval()
Out[4]: array(0.12)
In [5]: sh1.eval().dtype
Out[5]: dtype('float64')
Cheers,
Ramana
On Saturday, March 18, 2017 at 11:10:14 PM UTC+5:30, Archit Vora wrote:
>
> Is there anyway to specify this explicitly ?
>
> TypeError: We expected inputs of types '[TensorType(int8, scalar),
> TensorType(int8, scalar), TensorType(float64, scalar)]' but got types
> '[TensorType(float64, scalar), TensorType(float64, scalar),
> TensorType(float64, scalar)]'
>
>
> Thanks !
> -Archit
>
--
---
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.
