Hi,
I'm a total newb to theano and have been struggling with these problem for
days,
how can I add a custom loss function based on cosine similarity as defined
in this paper - github code here
<https://github.com/cvjena/semantic-embeddings/blob/master/utils.py#L46>?
def inv_correlation(y_true, y_pred):
""" Computes 1 minus the dot product between corresponding pairs of
samples in two tensors. """
return 1. - K.sum(y_true * y_pred, axis = -1)
I try doing by coping out the code from the theano nnet but am totally
confused on how to do this properly.
For some reason my code is going to the second if else, the ndim doesn't
match up, and I'm lost after that.
def cosine_similarity(coding_dist, true_dist):
"""
H(p,q) = - \sum_x p(x) \log(q(x))
"""
if true_dist.ndim == coding_dist.ndim: #it's not coming here
return -T.sum(true_dist * T.log(coding_dist),
axis=coding_dist.ndim - 1)
elif true_dist.ndim == coding_dist.ndim - 1: # try to modify this part
but it seems to be a bit way over my head
return cosine_similarity_1hot(coding_dist, true_dist)
else:
raise TypeError('rank mismatch between coding and true
distributions')
class CosineSimilarity1HotGrad(gof.Op):
__props__ = ()
def make_node(self, g_y, coding_dist, true_one_of_n):
return Apply(self, [g_y, coding_dist, true_one_of_n],
[coding_dist.type()])
def perform(self, node, inp, out):
g_y, coding_dist, true_one_of_n = inp
g_coding_strg, = out
g_coding = np.zeros_like(coding_dist)
for i in xrange(len(g_y)):
g_coding[i, true_one_of_n[i]] = (-g_y[i] /
coding_dist[i,
true_one_of_n[i]])
g_coding_strg[0] = g_coding
def infer_shape(self, node, in_shapes):
return [in_shapes[1]]
cosine_similarity_1hot_grad = CosineSimilarity1HotGrad()
class CosineSimilarity1Hot(gof.Op):
"""
Compute the cross entropy between a coding distribution and
a true distribution of the form [0, 0, ... 0, 1, 0, ..., 0].
.. math::
y[i] = - \log(coding_dist[i, one_of_n[i])
Notes
-----
In the case that the coding distribution is the output of a
softmax, an application of this Op will probably be optimized
away in favour of one with a C implementation.
"""
__props__ = ()
def make_node(self, coding_dist, true_one_of_n):
"""
Parameters
----------
coding_dist : dense matrix
true_one_of_n : lvector
Returns
-------
dvector
"""
_coding_dist = tensor.as_tensor_variable(coding_dist)
_true_one_of_n = tensor.as_tensor_variable(true_one_of_n)
if _coding_dist.type.ndim != 2:
raise TypeError('matrix required for argument: coding_dist')
if _true_one_of_n.type not in (tensor.lvector, tensor.ivector):
raise TypeError(
'integer vector required for argument: true_one_of_n'
'(got type: %s instead of: %s)' % (_true_one_of_n.type,
tensor.lvector))
return Apply(self, [_coding_dist, _true_one_of_n],
[tensor.Tensor(dtype=_coding_dist.dtype,
broadcastable=[False])()])
def perform(self, node, inp, out):
coding, one_of_n = inp
y_out, = out
y = np.zeros_like(coding[:, 0])
for i in xrange(len(y)):
y[i] = -np.log(coding[i, one_of_n[i]])
y[i] = 1. - T.sum(true_dist * coding_dist[0], axis= -1)
y_out[0] = y
def infer_shape(self, node, in_shapes):
return [(in_shapes[0][0],)]
def grad(self, inp, grads):
coding, one_of_n = inp
g_y, = grads
return [cosine_similarity_1hot_grad(g_y, coding, one_of_n),
grad_not_implemented(self, 1, one_of_n)]
cosine_similarity_1hot = CosineSimilarity1Hot()
--
---
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].
To view this discussion on the web visit
https://groups.google.com/d/msgid/theano-users/2e16211a-c1a7-42e2-81f8-5fc4490ec85f%40googlegroups.com.
