Skip to content

GitLab

Commit a800e9b1 authored by The-Indian-Chinna's avatar The-Indian-Chinna
Browse files

Kwargs update

parent dd5cc386
Hide whitespace changes
Inline Side-by-side
Showing with 57 additions and 111 deletions
official/vision/beta/projects/yolo/modeling/layers/nn_blocks.py
View file @ a800e9b1
...@@ -266,22 +266,24 @@ class DarkResidual(ks.layers.Layer): ...@@ -266,22 +266,24 @@ class DarkResidual(ks.layers.Layer):
super().__init__(**kwargs) super().__init__(**kwargs)
def build(self, input_shape): def build(self, input_shape):
_dark_conv_args = {"use_bias" : self._use_bias,
"kernel_initializer" : self._kernel_initializer,
"bias_initializer" : self._bias_initializer,
"bias_regularizer" : self._bias_regularizer,
"use_bn" : self._use_bn,
"use_sync_bn" : self._use_sync_bn,
"norm_momentum" : self._norm_moment,
"norm_epsilon" : self._norm_epsilon,
"activation" : self._conv_activation,
"kernel_regularizer" : self._kernel_regularizer,
"leaky_alpha" : self._leaky_alpha
}
if self._downsample: if self._downsample:
self._dconv = DarkConv(filters=self._filters, self._dconv = DarkConv(filters=self._filters,
kernel_size=(3, 3), kernel_size=(3, 3),
strides=(2, 2), strides=(2, 2),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
kernel_regularizer=self._kernel_regularizer,
leaky_alpha=self._leaky_alpha)
else: else:
self._dconv = Identity() self._dconv = Identity()
...@@ -289,32 +291,13 @@ class DarkResidual(ks.layers.Layer): ...@@ -289,32 +291,13 @@ class DarkResidual(ks.layers.Layer):
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
kernel_regularizer=self._kernel_regularizer,
leaky_alpha=self._leaky_alpha)
self._conv2 = DarkConv(filters=self._filters, self._conv2 = DarkConv(filters=self._filters,
kernel_size=(3, 3), kernel_size=(3, 3),
strides=(1, 1), strides=(1, 1),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
kernel_regularizer=self._kernel_regularizer,
leaky_alpha=self._leaky_alpha)
self._shortcut = ks.layers.Add() self._shortcut = ks.layers.Add()
# self._activation_fn = ks.layers.Activation(activation=self._sc_activation) # self._activation_fn = ks.layers.Activation(activation=self._sc_activation)
...@@ -430,21 +413,23 @@ class CSPTiny(ks.layers.Layer): ...@@ -430,21 +413,23 @@ class CSPTiny(ks.layers.Layer):
super().__init__(**kwargs) super().__init__(**kwargs)
def build(self, input_shape): def build(self, input_shape):
_dark_conv_args = {"use_bias" : self._use_bias,
"kernel_initializer" : self._kernel_initializer,
"bias_initializer" : self._bias_initializer,
"bias_regularizer" : self._bias_regularizer,
"use_bn" : self._use_bn,
"use_sync_bn" : self._use_sync_bn,
"norm_momentum" : self._norm_moment,
"norm_epsilon" : self._norm_epsilon,
"activation" : self._conv_activation,
"kernel_regularizer" : self._kernel_regularizer,
"leaky_alpha" : self._leaky_alpha
}
self._convlayer1 = DarkConv(filters=self._filters, self._convlayer1 = DarkConv(filters=self._filters,
kernel_size=(3, 3), kernel_size=(3, 3),
strides=(1, 1), strides=(1, 1),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
leaky_alpha=self._leaky_alpha)
self._convlayer2 = DarkConv(filters=self._filters // 2, self._convlayer2 = DarkConv(filters=self._filters // 2,
kernel_size=(3, 3), kernel_size=(3, 3),
...@@ -468,33 +453,13 @@ class CSPTiny(ks.layers.Layer): ...@@ -468,33 +453,13 @@ class CSPTiny(ks.layers.Layer):
kernel_size=(3, 3), kernel_size=(3, 3),
strides=(1, 1), strides=(1, 1),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
leaky_alpha=self._leaky_alpha)
self._convlayer4 = DarkConv(filters=self._filters, self._convlayer4 = DarkConv(filters=self._filters,
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
padding='same', padding='same',
use_bias=self._use_bias, **_dark_conv_args)
kernel_initializer=self._kernel_initializer,
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._conv_activation,
leaky_alpha=self._leaky_alpha)
self._maxpool = tf.keras.layers.MaxPool2D(pool_size=2, self._maxpool = tf.keras.layers.MaxPool2D(pool_size=2,
strides=2, strides=2,
...@@ -599,43 +564,29 @@ class CSPDownSample(ks.layers.Layer): ...@@ -599,43 +564,29 @@ class CSPDownSample(ks.layers.Layer):
self._norm_epsilon = norm_epsilon self._norm_epsilon = norm_epsilon
def build(self, input_shape): def build(self, input_shape):
_dark_conv_args = {"kernel_initializer" : self._kernel_initializer,
"bias_initializer" : self._bias_initializer,
"bias_regularizer" : self._bias_regularizer,
"use_bn" : self._use_bn,
"use_sync_bn" : self._use_sync_bn,
"norm_momentum" : self._norm_moment,
"norm_epsilon" : self._norm_epsilon,
"activation" : self._activation,
"kernel_regularizer" : self._kernel_regularizer,
}
self._conv1 = DarkConv(filters=self._filters, self._conv1 = DarkConv(filters=self._filters,
kernel_size=(3, 3), kernel_size=(3, 3),
strides=(2, 2), strides=(2, 2),
kernel_initializer=self._kernel_initializer, **_dark_conv_args)
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._activation)
self._conv2 = DarkConv(filters=self._filters // self._filter_reduce, self._conv2 = DarkConv(filters=self._filters // self._filter_reduce,
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
kernel_initializer=self._kernel_initializer, **_dark_conv_args)
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._activation)
self._conv3 = DarkConv(filters=self._filters // self._filter_reduce, self._conv3 = DarkConv(filters=self._filters // self._filter_reduce,
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
kernel_initializer=self._kernel_initializer, **_dark_conv_args)
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._activation)
def call(self, inputs): def call(self, inputs):
x = self._conv1(inputs) x = self._conv1(inputs)
...@@ -701,31 +652,26 @@ class CSPConnect(ks.layers.Layer): ...@@ -701,31 +652,26 @@ class CSPConnect(ks.layers.Layer):
self._norm_epsilon = norm_epsilon self._norm_epsilon = norm_epsilon
def build(self, input_shape): def build(self, input_shape):
_dark_conv_args = {
"kernel_initializer" : self._kernel_initializer,
"bias_initializer" : self._bias_initializer,
"bias_regularizer" : self._bias_regularizer,
"use_bn" : self._use_bn,
"use_sync_bn" : self._use_sync_bn,
"norm_momentum" : self._norm_moment,
"norm_epsilon" : self._norm_epsilon,
"activation" : self._activation,
"kernel_regularizer" : self._kernel_regularizer,
}
self._conv1 = DarkConv(filters=self._filters // self._filter_reduce, self._conv1 = DarkConv(filters=self._filters // self._filter_reduce,
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
kernel_initializer=self._kernel_initializer, **_dark_conv_args)
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._activation)
self._concat = ks.layers.Concatenate(axis=-1) self._concat = ks.layers.Concatenate(axis=-1)
self._conv2 = DarkConv(filters=self._filters, self._conv2 = DarkConv(filters=self._filters,
kernel_size=(1, 1), kernel_size=(1, 1),
strides=(1, 1), strides=(1, 1),
kernel_initializer=self._kernel_initializer, **_dark_conv_args)
bias_initializer=self._bias_initializer,
bias_regularizer=self._bias_regularizer,
kernel_regularizer=self._kernel_regularizer,
use_bn=self._use_bn,
use_sync_bn=self._use_sync_bn,
norm_momentum=self._norm_moment,
norm_epsilon=self._norm_epsilon,
activation=self._activation)
def call(self, inputs): def call(self, inputs):
x_prev, x_csp = inputs x_prev, x_csp = inputs
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment