mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
a8695178aa
Summary:
To achive this, I modified the blob name scheme defined in a layer.
Before it was scope/fc_w and scope/fc_w_auto_0 (if there is another fc
within the same scope).
Now I change it to scope/fc/w and scope/fc_auto_0/w.
That is, we rely on the uniqueness of the scoped layer name to define
names for blobs.
I also overwrote the create_param method in LayerModelHelper to let it
use the resolved name for blobs given the sharingparameter context.
There are some details such as making the initializer more structured
that I need to finalize.
Reviewed By: kennyhorror
Differential Revision: D5435132
fbshipit-source-id: a0525f5ea0977e255dd5ea765b38913f5951d455
66 lines
2.2 KiB
Python
66 lines
2.2 KiB
Python
## @package position_weighted
|
|
# Module caffe2.python.layers.position_weighted
|
|
from __future__ import absolute_import
|
|
from __future__ import division
|
|
from __future__ import print_function
|
|
from __future__ import unicode_literals
|
|
|
|
import logging
|
|
import numpy as np
|
|
|
|
from caffe2.python import schema
|
|
from caffe2.python.layers.layers import (
|
|
get_categorical_limit,
|
|
ModelLayer,
|
|
)
|
|
|
|
from caffe2.python.layers.tags import Tags
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
class PositionWeighted(ModelLayer):
|
|
def __init__(self, model, input_record, weight_optim=None,
|
|
name="position_weights"):
|
|
super(PositionWeighted, self).__init__(model, name, input_record)
|
|
|
|
assert isinstance(input_record, schema.List), "Incorrect input type"
|
|
length_metadata = input_record.lengths.metadata
|
|
max_length = (length_metadata.categorical_limit if length_metadata is
|
|
not None else None)
|
|
if max_length is not None:
|
|
self.shape = max_length
|
|
else:
|
|
self.shape = get_categorical_limit(input_record)
|
|
logger.warning(
|
|
'{}: categorical_limit of lengths is not available, using '
|
|
'categorical_limit of the keys: {}'.format(
|
|
str(input_record.lengths()), self.shape))
|
|
|
|
self.pos_w = self.create_param(param_name='pos_w',
|
|
shape=[self.shape, ],
|
|
initializer=('ConstantFill', {'value': 1.0}),
|
|
optimizer=weight_optim)
|
|
|
|
self.output_schema = schema.Struct(
|
|
('position_weights',
|
|
schema.Scalar((np.float32, self.shape),
|
|
self.get_next_blob_reference("pos_w_gather")))
|
|
)
|
|
|
|
self.tags.update({Tags.HANDLE_AS_SPARSE_LAYER})
|
|
self.tags.update({Tags.GRADIENT_FROM_PS})
|
|
|
|
def get_memory_usage(self):
|
|
return self.shape
|
|
|
|
def add_ops(self, net):
|
|
inc_seq = net.LengthsRangeFill(
|
|
[self.input_record.lengths()],
|
|
self.input_record.lengths() + '_pos_w_seq'
|
|
)
|
|
|
|
net.Gather(
|
|
[self.pos_w, inc_seq],
|
|
self.output_schema.position_weights.field_blobs())
|