def _build_base(self): with tf.variable_scope(self._scope, self._scope): net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def build_base(self): with tf.variable_scope(self.scope, self.scope, reuse=tf.AUTO_REUSE,): net = resnet_utils.conv2d_same(self.image, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): """Bottleneck residual unit variant with BN after convolutions. This is the original residual unit proposed in [1]. See Fig. 1(a) of [2] for its definition. Note that we use here the bottleneck variant which has an extra bottleneck layer. When putting together two consecutive ResNet blocks that use this unit, one should use stride = 2 in the last unit of the first block. Args: inputs: A tensor of size [batch, height, width, channels]. depth: The depth of the ResNet unit output. depth_bottleneck: The depth of the bottleneck layers. stride: The ResNet unit's stride. Determines the amount of downsampling of the units output compared to its input. rate: An integer, rate for atrous convolution. outputs_collections: Collection to add the ResNet unit output. scope: Optional variable_scope. Returns: The ResNet unit's output. """ with variable_scope.variable_scope(scope, 'bottleneck_v1', [inputs]) as sc: depth_in = utils.last_dimension(inputs.get_shape(), min_rank=4) if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = layers.conv2d(inputs, depth, [1, 1], stride=stride, activation_fn=None, scope='shortcut') residual = layers.conv2d(inputs, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = resnet_utils.conv2d_same(residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = layers.conv2d(residual, depth, [1, 1], stride=1, activation_fn=None, scope='conv3') output = nn_ops.relu(shortcut + residual) return utils.collect_named_outputs(outputs_collections, sc.name, output)
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): """Bottleneck residual unit variant with BN before convolutions. This is the full preactivation residual unit variant proposed in [2]. See Fig. 1(b) of [2] for its definition. Note that we use here the bottleneck variant which has an extra bottleneck layer. When putting together two consecutive ResNet blocks that use this unit, one should use stride = 2 in the last unit of the first block. Args: inputs: A tensor of size [batch, height, width, channels]. depth: The depth of the ResNet unit output. depth_bottleneck: The depth of the bottleneck layers. stride: The ResNet unit's stride. Determines the amount of downsampling of the units output compared to its input. rate: An integer, rate for atrous convolution. outputs_collections: Collection to add the ResNet unit output. scope: Optional variable_scope. Returns: The ResNet unit's output. """ with variable_scope.variable_scope(scope, 'bottleneck_v2', [inputs]) as sc: depth_in = utils.last_dimension(inputs.get_shape(), min_rank=4) preact = layers.batch_norm( inputs, activation_fn=nn_ops.relu, scope='preact') if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = layers_lib.conv2d( preact, depth, [1, 1], stride=stride, normalizer_fn=None, activation_fn=None, scope='shortcut') residual = layers_lib.conv2d( preact, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = resnet_utils.conv2d_same( residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = layers_lib.conv2d( residual, depth, [1, 1], stride=1, normalizer_fn=None, activation_fn=None, scope='conv3') output = shortcut + residual return utils.collect_named_outputs(outputs_collections, sc.name, output)
def atrous_spatial_pyramid_pooling(inputs, output_stride, batch_norm_decay, is_training, depth=256): """Atrous Spatial Pyramid Pooling. Args: inputs: A tensor of size [batch, height, width, channels]. output_stride: The ResNet unit's stride. Determines the rates for atrous convolution. the rates are (6, 12, 18) when the stride is 16, and doubled when 8. batch_norm_decay: The moving average decay when estimating layer activation statistics in batch normalization. is_training: A boolean denoting whether the input is for training. depth: The depth of the ResNet unit output. Returns: The atrous spatial pyramid pooling output. """ with tf.variable_scope("aspp"): if output_stride not in [8, 16]: raise ValueError('output_stride must be either 8 or 16.') atrous_rates = [6, 12, 18] if output_stride == 8: atrous_rates = [2*rate for rate in atrous_rates] with tf.contrib.slim.arg_scope(resnet_v2.resnet_arg_scope(batch_norm_decay=batch_norm_decay)): with arg_scope([layers.batch_norm], is_training=is_training): inputs_size = tf.shape(inputs)[1:3] # (a) one 1×1 convolution and three 3×3 convolutions with rates = (6, 12, 18) when output stride = 16. # the rates are doubled when output stride = 8. conv_1x1 = layers_lib.conv2d(inputs, depth, [1, 1], stride=1, scope="conv_1x1") conv_3x3_1 = resnet_utils.conv2d_same(inputs, depth, 3, stride=1, rate=atrous_rates[0], scope='conv_3x3_1') conv_3x3_2 = resnet_utils.conv2d_same(inputs, depth, 3, stride=1, rate=atrous_rates[1], scope='conv_3x3_2') conv_3x3_3 = resnet_utils.conv2d_same(inputs, depth, 3, stride=1, rate=atrous_rates[2], scope='conv_3x3_3') # (b) the image-level features with tf.variable_scope("image_level_features"): # global average pooling image_level_features = tf.reduce_mean(inputs, [1, 2], name='global_average_pooling', keepdims=True) # 1×1 convolution with 256 filters( and batch normalization) image_level_features = layers_lib.conv2d(image_level_features, depth, [1, 1], stride=1, scope='conv_1x1') # bilinearly upsample features image_level_features = tf.image.resize_bilinear(image_level_features, inputs_size, name='upsample') net = tf.concat([conv_1x1, conv_3x3_1, conv_3x3_2, conv_3x3_3, image_level_features], axis=3, name='concat') net = layers_lib.conv2d(net, depth, [1, 1], stride=1, scope='conv_1x1_concat') return net
def build_base(self,conv): with tf.variable_scope(self._resnet_scope, self._resnet_scope): net = resnet_utils.conv2d_same(conv, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def RRB(inputs, is_training=True, depth=512): """Refinement Residual Block. Args: inputs: A tensor of size [batch, height, width, channels]. Returns: output. """ conv_1x1 = layers_lib.conv2d(inputs, depth, [1, 1], stride=1) conv_3x3_1 = resnet_utils.conv2d_same(conv_1x1, depth, 3, stride=1) batch_norm = tf.layers.batch_normalization(conv_3x3_1, training=is_training) relu1 = tf.nn.relu(batch_norm) conv_3x3_2 = resnet_utils.conv2d_same(relu1, depth, 3, stride=1) sum = tf.add_n([conv_1x1, conv_3x3_2]) relu2 = tf.nn.relu(sum) return relu2
def uncompressed_func(): nonlocal uncompressed_net if uncompressed_net is None: #need this as irritatingly the tf.case() calls it twice uncompressed_net = resnet_utils.conv2d_same( inputs, num_output_channels, kernel_size, stride=stride, scope=layer_name) return uncompressed_net
def build_base(self): c = np.zeros((3, 5, 5)) c[0] = [[-1, 2, -2, 2, -1], [2, -6, 8, -6, 2], [-2, 8, -12, 8, -2], [2, -6, 8, -6, 2], [-1, 2, -2, 2, -1]] c[0] = c[0] / 12 c[1] = [[0, 0, 0, 0, 0], [0, -1, 2, -1, 0], [0, 0, 2, -4, 2, 0], [0, -1, 2, -1, 0], [0, 0, 0, 0, 0]] c[1] /= 4 #c[2][2][1]=1 #c[2][2][2]=-2 #c[2][2][3]=1 #c[2]=c[2]/2 c[2] = [[0, 0, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, -2, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 0, 0]] c[2] /= 2 Wcnn = np.zeros((5, 5, 3, 3)) for i in xrange(3): #k=i%10+1 #Wcnn[i]=[c[3*k-3],c[3*k-2],c[3*k-1]] Wcnn[:, :, 0, i] = c[i] Wcnn[:, :, 1, i] = c[i] Wcnn[:, :, 2, i] = c[i] with tf.variable_scope('noise'): #kernel = tf.get_variable('weights', #shape=[5, 5, 3, 3], #initializer=tf.constant_initializer(c)) conv = tf.nn.conv2d(self._image, Wcnn, [1, 1, 1, 1], padding='SAME', name='srm') self._layers['noise'] = conv #with tf.variable_scope('noise'): ##kernel = tf.get_variable('weights', #shape=[5, 5, 3, 3], #initializer=tf.constant_initializer(Wcnn)) #conv = tf.nn.conv2d(self.noise, Wcnn, [1, 1, 1, 1], padding='SAME',name='srm') #conv = tf.nn.conv2d(self.noise, kernel, [1, 1, 1, 1], padding='SAME',name='srm') #srm_conv = tf.nn.tanh(conv, name='tanh') with tf.variable_scope(self._resnet_scope, self._resnet_scope): net = resnet_utils.conv2d_same(conv, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def build_base(self, ver='', init=None): with tf.variable_scope(self._resnet_scope, self._resnet_scope): if(ver == 'n'): def truncate_2(x): neg = ((x + 2) + abs(x + 2)) / 2 - 2 return -(2 - neg + abs(2 - neg)) / 2 + 2 # Main network # Layer SRM net = slim.conv2d(self._image, 3, [5, 5], trainable=True, weights_initializer=init, activation_fn=None, padding='SAME', stride=1, scope='srm') net = truncate_2(net) net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1'+ver) net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1'+ver) else: net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1'+ver) net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1'+ver) return net
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): with tf.variable_scope(scope, 'bottleneck_v1', [inputs]) as sc: depth_in = slim.utils.last_dimension(inputs.get_shape(), min_rank=4) if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = slim.conv2d(inputs, depth, [1, 1], stride=stride, activation_fn=None, scope='shortcut') residual = slim.conv2d(inputs, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = resnet_utils.conv2d_same(residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = slim.conv2d(residual, depth, [1, 1], stride=1, activation_fn=None, scope='conv3') output = tf.nn.relu(shortcut + residual) return slim.utils.collect_named_outputs(outputs_collections, sc.original_name_scope, output)
def _build_base(self, reuse=None): with tf.variable_scope(self._scope, self._scope) as scope: if (reuse == True): scope.reuse_variables() net = resnet_utils.conv2d_same(self._image_large, 64, 7, stride=2, scope='conv1') else: net = resnet_utils.conv2d_same(self._image_small, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): with tf.variable_scope(scope, 'bottleneck_v1', [inputs]) as sc: depth_in = slim.utils.last_dimension(inputs.get_shape(), min_rank=4) if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = slim.conv2d(inputs, depth, [1, 1], stride=stride, activation_fn=None, scope='shortcut') residual = slim.conv2d(inputs, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = resnet_utils.conv2d_same(residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = slim.conv2d(residual, depth, [1, 1], stride=1, activation_fn=None, scope='conv3') output = tf.nn.relu(shortcut + residual) return slim.utils.collect_named_outputs(outputs_collections, sc.original_name_scope, output)
def resnet_v1(inputs, blocks, num_classes=None, is_training=True, global_pool=True, output_stride=None, include_root_block=True, reuse=None, scope=None): with variable_scope.variable_scope(scope, 'resnet_v1', [inputs], reuse=reuse) as sc: end_points_collection = sc.original_name_scope + '_end_points' with arg_scope([layers.conv2d, naive, resnet_utils.stack_blocks_dense], outputs_collections=end_points_collection): with arg_scope([layers.batch_norm], is_training=is_training): net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError( 'The output_stride needs to be a multiple of 4.' ) output_stride /= 4 net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = layers_lib.max_pool2d(net, [3, 3], stride=2, scope='pool1') net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) if global_pool: # Global average pooling. net = math_ops.reduce_mean(net, [1, 2], name='pool5', keep_dims=True) if num_classes is not None: net = layers.conv2d(net, num_classes, [1, 1], activation_fn=None, normalizer_fn=None, scope='logits') end_points = utils.convert_collection_to_dict( end_points_collection) if num_classes is not None: end_points['predictions'] = layers_lib.softmax( net, scope='predictions') return net, end_points
def _build_base_hm(self): with tf.variable_scope('hm/' + self._scope, 'hm/' + self._scope): net = resnet_utils.conv2d_same(self._image[..., 3:], 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') self._layers['base_hm'] = net return net
def _build_base(self): with tf.variable_scope(self._scope, self._scope): net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1') # 只填充宽高两个维度,上下左右各填充一层 net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def encoder(self, x): with tf.variable_scope('encoder'): net = resnet_utils.conv2d_same(x, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) x = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') x_features_all, _ = resnet_v1.resnet_v1(x, self._blocks_encoder, global_pool=False, include_root_block=False, scope=self._resnet_scope) x_features_all = tf.reduce_mean(x_features_all, axis=[1, 2]) x_features_labeled, x_features_unlabeled = tf.split(x_features_all, 2) x_features_tiled = tf.tile(x_features_unlabeled, [self._num_classes, 1]) # (100, 256) --> (2100, 256) x_features = tf.concat([x_features_labeled, x_features_tiled], 0) # (2100, 256) --> (2200, 256) return x_features
def mobilenet_v1_base(inputs, conv_defs, starting_layer=0, min_depth=8, depth_multiplier=1.0, output_stride=None, reuse=None, scope=None): # starting_layer 对于RPN为0,对于region classification 是12 depth =lambda d:max(int(d*depth_multiplier),min_depth) end_points = {} if depth_multiplier <= 0: raise ValueError('depth_multiplier is not greater than zero.') with tf.variable_scope(scope,'MobilenetV1',[inputs],reuse=reuse): current_stride = 1 rate = 1 net = inputs for i, conv_def in enumerate(conv_defs): end_point_base = 'Conv2d_%d'%(i+starting_layer) if output_stride is not None and current_stride == output_stride: layer_stride = 1 layer_rate = rate rate *= conv_def.stride else: layer_stride = conv_def.stride layer_rate = 1 current_stride *= conv_def.stride if isinstance(conv_def,Conv): end_point = end_point_base net = resnet_utils.conv2d_same(net,depth(conv_def.depth),conv_def.kernel, stride=conv_def.stride, scope=end_point) elif isinstance(conv_def,DepthSepConv): end_point = end_point_base + '_depthwise' net = separable_conv2d_same(net,conv_def.kernel, stride=layer_stride, rate=layer_rate, scope=end_point) end_point = end_point_base + '_pointwise' net = slim.conv2d(net,depth(conv_def.depth),[1,1], stride=1, scope=end_point) else: raise ValueError('Unknown convolution type %s for layer %d'%(conv_def.ltype,i)) return net
def build_base(self): c = np.zeros((3, 5, 5)) c[0] = [[-1, 2, -2, 2, -1], [2, -6, 8, -6, 2], [-2, 8, -12, 8, -2], [2, -6, 8, -6, 2], [-1, 2, -2, 2, -1]] c[0] = c[0] / 12 c[1][1][1] = -1 c[1][1][2] = 2 c[1][1][3] = -1 c[1][2][1] = 2 c[1][2][2] = -4 c[1][2][3] = 2 c[1][3][1] = -1 c[1][3][2] = 2 c[1][3][3] = -1 c[1] = c[1] / 4 c[2][1][2] = 1 c[2][2][2] = -2 c[2][3][2] = 1 c[2] = c[2] / 2 Wcnn = np.zeros((5, 5, 3, 3)) for i in xrange(3): Wcnn[:, :, 0, i] = c[i] Wcnn[:, :, 1, i] = c[i] Wcnn[:, :, 2, i] = c[i] with tf.variable_scope('noise'): conv = tf.nn.conv2d(self._image, Wcnn, [1, 1, 1, 1], padding='SAME', name='srm') self._layers['noise'] = conv with tf.variable_scope(self._resnet_scope, self._resnet_scope): net = resnet_utils.conv2d_same(conv, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): with variable_scope.variable_scope(scope, 'bottleneck_v2', [inputs]) as sc: depth_in = utils.last_dimension(inputs.get_shape(), min_rank=4) preact = layers.batch_norm( inputs, activation_fn=nn_ops.relu, scope='preact') if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = layers_lib.conv2d( preact, depth, [1, 1], stride=stride, normalizer_fn=None, activation_fn=None, scope='shortcut') residual = preact residual = tf.layers.batch_normalization(residual) residual = tf.nn.relu(residual) residual = layers_lib.conv2d( residual, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = tf.layers.batch_normalization(residual) residual = tf.nn.relu(residual) residual = resnet_utils.conv2d_same( residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = tf.layers.batch_normalization(residual) residual = tf.nn.relu(residual) residual = layers_lib.conv2d( residual, depth, [1, 1], stride=1, normalizer_fn=None, activation_fn=None, scope='conv3') output = shortcut + residual return utils.collect_named_outputs(outputs_collections, sc.name, output)
def resnet50V2_reduced(inputs, is_training=True, output_stride=None, include_root_block=True, reuse=None, scope=None): # These are the blocks for resnet 50 blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]), resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 5) ] # Initialize Model with tf.variable_scope(scope, 'resnet_v2_50', [inputs], reuse=reuse): with slim.arg_scope( [slim.conv2d, bottleneck, resnet_utils.stack_blocks_dense]): with slim.arg_scope([slim.batch_norm], is_training=is_training) as scope: net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError( 'The output_stride needs to be a multiple of 4.' ) output_stride /= 4 with slim.arg_scope([slim.conv2d], activation_fn=None, normalizer_fn=None): net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = slim.max_pool2d(net, [3, 3], stride=2, scope='pool1') net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) with slim.arg_scope([slim.batch_norm], is_training=is_training) as scope: net = slim.batch_norm(net, activation_fn=tf.nn.relu, scope='postnorm') return net
def testConv2DSameOdd(self): n, n2 = 5, 3 # Input image. x = create_test_input(1, n, n, 1) # Convolution kernel. w = create_test_input(1, 3, 3, 1) w = array_ops.reshape(w, [3, 3, 1, 1]) variable_scope.get_variable('Conv/weights', initializer=w) variable_scope.get_variable('Conv/biases', initializer=array_ops.zeros([1])) variable_scope.get_variable_scope().reuse_variables() y1 = layers.conv2d(x, 1, [3, 3], stride=1, scope='Conv') y1_expected = math_ops.cast([[14, 28, 43, 58, 34], [28, 48, 66, 84, 46], [43, 66, 84, 102, 55], [58, 84, 102, 120, 64], [34, 46, 55, 64, 30]], dtypes.float32) y1_expected = array_ops.reshape(y1_expected, [1, n, n, 1]) y2 = resnet_utils.subsample(y1, 2) y2_expected = math_ops.cast([[14, 43, 34], [43, 84, 55], [34, 55, 30]], dtypes.float32) y2_expected = array_ops.reshape(y2_expected, [1, n2, n2, 1]) y3 = resnet_utils.conv2d_same(x, 1, 3, stride=2, scope='Conv') y3_expected = y2_expected y4 = layers.conv2d(x, 1, [3, 3], stride=2, scope='Conv') y4_expected = y2_expected with self.cached_session() as sess: sess.run(variables.global_variables_initializer()) self.assertAllClose(y1.eval(), y1_expected.eval()) self.assertAllClose(y2.eval(), y2_expected.eval()) self.assertAllClose(y3.eval(), y3_expected.eval()) self.assertAllClose(y4.eval(), y4_expected.eval())
def build_base(self): #with tf.variable_scope('noise'): ##kernel = tf.get_variable('weights', #shape=[5, 5, 3, 3], #initializer=tf.constant_initializer(Wcnn)) #conv = tf.nn.conv2d(self.noise, Wcnn, [1, 1, 1, 1], padding='SAME',name='srm') #conv = tf.nn.conv2d(self.noise, kernel, [1, 1, 1, 1], padding='SAME',name='srm') #srm_conv = tf.nn.tanh(conv, name='tanh') with tf.variable_scope(self._resnet_scope, self._resnet_scope): net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def build_base(self): with tf.variable_scope(self._resnet_scope, self._resnet_scope): net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1') self._predictions[self._resnet_scope + '/conv1'] = net # end_points_collection = self._resnet_scope + '_end_points' utils.collect_named_outputs(self._end_points_collection, self._resnet_scope + '/conv1', net) net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def testConv2DSameOdd(self): n, n2 = 5, 3 # Input image. x = create_test_input(1, n, n, 1) # Convolution kernel. w = create_test_input(1, 3, 3, 1) w = array_ops.reshape(w, [3, 3, 1, 1]) variable_scope.get_variable('Conv/weights', initializer=w) variable_scope.get_variable('Conv/biases', initializer=array_ops.zeros([1])) variable_scope.get_variable_scope().reuse_variables() y1 = layers.conv2d(x, 1, [3, 3], stride=1, scope='Conv') y1_expected = math_ops.to_float([[14, 28, 43, 58, 34], [28, 48, 66, 84, 46], [43, 66, 84, 102, 55], [58, 84, 102, 120, 64], [34, 46, 55, 64, 30]]) y1_expected = array_ops.reshape(y1_expected, [1, n, n, 1]) y2 = resnet_utils.subsample(y1, 2) y2_expected = math_ops.to_float([[14, 43, 34], [43, 84, 55], [34, 55, 30]]) y2_expected = array_ops.reshape(y2_expected, [1, n2, n2, 1]) y3 = resnet_utils.conv2d_same(x, 1, 3, stride=2, scope='Conv') y3_expected = y2_expected y4 = layers.conv2d(x, 1, [3, 3], stride=2, scope='Conv') y4_expected = y2_expected with self.cached_session() as sess: sess.run(variables.global_variables_initializer()) self.assertAllClose(y1.eval(), y1_expected.eval()) self.assertAllClose(y2.eval(), y2_expected.eval()) self.assertAllClose(y3.eval(), y3_expected.eval()) self.assertAllClose(y4.eval(), y4_expected.eval())
def resnet_v2_50(inputs, num_classes=None, is_training=True, global_pool=False, output_stride=None, reuse=None, scope='resnet_v2_50'): """ResNet-50 model of [1]. See resnet_v2() for arg and return description.""" resnet_v2_block = resnet_v2.resnet_v2_block blocks = [ resnet_v2_block('block1', base_depth=64, num_units=3, stride=2), resnet_v2_block('block2', base_depth=128, num_units=4, stride=2), resnet_v2_block('block3', base_depth=256, num_units=6, stride=2), resnet_v2_block('block4', base_depth=512, num_units=3, stride=1), ] # custom root block, because original root block will use maxpooling with valid padding. with tf.variable_scope(scope) as sc: with slim.arg_scope([slim.conv2d], activation_fn=None, normalizer_fn=None): net = resnet_utils.conv2d_same(inputs, 64, 7, stride=2, scope='conv1') # stride 2 net = slim.max_pool2d(net, [3, 3], stride=2, scope='pool1', padding='SAME') # stride 2 return resnet_v2.resnet_v2(net, blocks, num_classes, is_training=is_training, global_pool=global_pool, output_stride=output_stride, include_root_block=False, reuse=reuse, scope=scope)
def resnet50_reduced(inputs, is_training=True, output_stride=None, include_root_block=True, reuse=None, scope=None): # These are the blocks for resnet 50 blocks = [ resnet_v1_block('block1', base_depth=64, num_units=3, stride=2), resnet_v1_block('block2', base_depth=128, num_units=4, stride=2), resnet_v1_block('block3', base_depth=256, num_units=6, stride=2), # resnet_v1_block('block4', base_depth=512, num_units=3, stride=1), ] # Initialize Model with tf.variable_scope(scope, 'resnet_v1_50', [inputs], reuse=reuse): with slim.arg_scope( [slim.conv2d, bottleneck, resnet_utils.stack_blocks_dense]): with slim.arg_scope([slim.batch_norm], is_training=is_training) as scope: net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError( 'The output_stride needs to be a multiple of 4.' ) output_stride /= 4 net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = slim.max_pool2d(net, [3, 3], stride=2, scope='pool1') net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) return net
def resnet50_reduced(inputs, is_training=True, output_stride=None, include_root_block=True, reuse=None, scope=None): # These are the blocks for resnet 50 blocks = [ resnet_v1_block('block1', base_depth=64, num_units=3, stride=2), resnet_v1_block('block2', base_depth=128, num_units=4, stride=2), resnet_v1_block('block3', base_depth=256, num_units=6, stride=2), # resnet_v1_block('block4', base_depth=512, num_units=3, stride=1), ] # Initialize Model with tf.variable_scope(scope, 'resnet_v1_50', [inputs], reuse=reuse): with slim.arg_scope([slim.conv2d, bottleneck, resnet_utils.stack_blocks_dense]): with slim.arg_scope([slim.batch_norm], is_training=is_training) as scope: net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError('The output_stride needs to be a multiple of 4.') output_stride /= 4 net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = slim.max_pool2d(net, [3, 3], stride=2, scope='pool1') net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) return net
def testConv2DSameEven(self): n, n2 = 4, 2 # Input image. x = create_test_input(1, n, n, 1) # Convolution kernel. w = create_test_input(1, 3, 3, 1) w = array_ops.reshape(w, [3, 3, 1, 1]) variable_scope.get_variable('Conv/weights', initializer=w) variable_scope.get_variable('Conv/biases', initializer=array_ops.zeros([1])) variable_scope.get_variable_scope().reuse_variables() y1 = layers.conv2d(x, 1, [3, 3], stride=1, scope='Conv') y1_expected = math_ops.to_float([[14, 28, 43, 26], [28, 48, 66, 37], [43, 66, 84, 46], [26, 37, 46, 22]]) y1_expected = array_ops.reshape(y1_expected, [1, n, n, 1]) y2 = resnet_utils.subsample(y1, 2) y2_expected = math_ops.to_float([[14, 43], [43, 84]]) y2_expected = array_ops.reshape(y2_expected, [1, n2, n2, 1]) y3 = resnet_utils.conv2d_same(x, 1, 3, stride=2, scope='Conv') y3_expected = y2_expected y4 = layers.conv2d(x, 1, [3, 3], stride=2, scope='Conv') y4_expected = math_ops.to_float([[48, 37], [37, 22]]) y4_expected = array_ops.reshape(y4_expected, [1, n2, n2, 1]) with self.test_session() as sess: sess.run(variables.global_variables_initializer()) self.assertAllClose(y1.eval(), y1_expected.eval()) self.assertAllClose(y2.eval(), y2_expected.eval()) self.assertAllClose(y3.eval(), y3_expected.eval()) self.assertAllClose(y4.eval(), y4_expected.eval())
def build_base(self): with tf.variable_scope(self._resnet_scope, self._resnet_scope): # with arg_scope( # [slim.conv2d], # weights_regularizer=None, # weights_initializer=None, # trainable=False, # # activation_fn=nn_ops.relu, # activation_fn=None, # normalizer_fn=None, # normalizer_params=None, # biases_initializer=None): #make first layer clean, no BN no biases net = resnet_utils.conv2d_same(self._image, 64, 7, stride=2, scope='conv1') self._predictions[self._resnet_scope+'/conv1'] = net # end_points_collection = self._resnet_scope + '_end_points' utils.collect_named_outputs(self._end_points_collection, self._resnet_scope+'/conv1', net) net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') return net
def resnet_v2(inputs, blocks, num_classes=None, is_training=None, global_pool=True, output_stride=None, include_root_block=True, reuse=None, scope=None): """Generator for v2 (preactivation) ResNet models. This function generates a family of ResNet v2 models. See the resnet_v2_*() methods for specific model instantiations, obtained by selecting different block instantiations that produce ResNets of various depths. Training for image classification on Imagenet is usually done with [224, 224] inputs, resulting in [7, 7] feature maps at the output of the last ResNet block for the ResNets defined in [1] that have nominal stride equal to 32. However, for dense prediction tasks we advise that one uses inputs with spatial dimensions that are multiples of 32 plus 1, e.g., [321, 321]. In this case the feature maps at the ResNet output will have spatial shape [(height - 1) / output_stride + 1, (width - 1) / output_stride + 1] and corners exactly aligned with the input image corners, which greatly facilitates alignment of the features to the image. Using as input [225, 225] images results in [8, 8] feature maps at the output of the last ResNet block. For dense prediction tasks, the ResNet needs to run in fully-convolutional (FCN) mode and global_pool needs to be set to False. The ResNets in [1, 2] all have nominal stride equal to 32 and a good choice in FCN mode is to use output_stride=16 in order to increase the density of the computed features at small computational and memory overhead, cf. http://arxiv.org/abs/1606.00915. Args: inputs: A tensor of size [batch, height_in, width_in, channels]. blocks: A list of length equal to the number of ResNet blocks. Each element is a resnet_utils.Block object describing the units in the block. num_classes: Number of predicted classes for classification tasks. If None we return the features before the logit layer. is_training: whether is training or not. If None, the value inherited from the resnet_arg_scope is used. Specifying value None is deprecated. global_pool: If True, we perform global average pooling before computing the logits. Set to True for image classification, False for dense prediction. output_stride: If None, then the output will be computed at the nominal network stride. If output_stride is not None, it specifies the requested ratio of input to output spatial resolution. include_root_block: If True, include the initial convolution followed by max-pooling, if False excludes it. If excluded, `inputs` should be the results of an activation-less convolution. reuse: whether or not the network and its variables should be reused. To be able to reuse 'scope' must be given. scope: Optional variable_scope. Returns: net: A rank-4 tensor of size [batch, height_out, width_out, channels_out]. If global_pool is False, then height_out and width_out are reduced by a factor of output_stride compared to the respective height_in and width_in, else both height_out and width_out equal one. If num_classes is None, then net is the output of the last ResNet block, potentially after global average pooling. If num_classes is not None, net contains the pre-softmax activations. end_points: A dictionary from components of the network to the corresponding activation. Raises: ValueError: If the target output_stride is not valid. """ with variable_scope.variable_scope( scope, 'resnet_v2', [inputs], reuse=reuse) as sc: end_points_collection = sc.original_name_scope + '_end_points' with arg_scope( [layers_lib.conv2d, bottleneck, resnet_utils.stack_blocks_dense], outputs_collections=end_points_collection): if is_training is not None: bn_scope = arg_scope([layers.batch_norm], is_training=is_training) else: bn_scope = arg_scope([]) with bn_scope: net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError('The output_stride needs to be a multiple of 4.') output_stride /= 4 # We do not include batch normalization or activation functions in # conv1 because the first ResNet unit will perform these. Cf. # Appendix of [2]. with arg_scope( [layers_lib.conv2d], activation_fn=None, normalizer_fn=None): net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = layers.max_pool2d(net, [3, 3], stride=2, scope='pool1') net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) # This is needed because the pre-activation variant does not have batch # normalization or activation functions in the residual unit output. See # Appendix of [2]. net = layers.batch_norm( net, activation_fn=nn_ops.relu, scope='postnorm') if global_pool: # Global average pooling. net = math_ops.reduce_mean(net, [1, 2], name='pool5', keep_dims=True) if num_classes is not None: net = layers_lib.conv2d( net, num_classes, [1, 1], activation_fn=None, normalizer_fn=None, scope='logits') # Convert end_points_collection into a dictionary of end_points. end_points = utils.convert_collection_to_dict(end_points_collection) if num_classes is not None: end_points['predictions'] = layers.softmax(net, scope='predictions') return net, end_points
def resnet_v2(inputs, blocks, num_classes=None, is_training=True, global_pool=True, output_stride=None, include_root_block=True, reuse=None, scope=None): """Generator for v2 (preactivation) ResNet models. This function generates a family of ResNet v2 models. See the resnet_v2_*() methods for specific model instantiations, obtained by selecting different block instantiations that produce ResNets of various depths. Training for image classification on Imagenet is usually done with [224, 224] inputs, resulting in [7, 7] feature maps at the output of the last ResNet block for the ResNets defined in [1] that have nominal stride equal to 32. However, for dense prediction tasks we advise that one uses inputs with spatial dimensions that are multiples of 32 plus 1, e.g., [321, 321]. In this case the feature maps at the ResNet output will have spatial shape [(height - 1) / output_stride + 1, (width - 1) / output_stride + 1] and corners exactly aligned with the input image corners, which greatly facilitates alignment of the features to the image. Using as input [225, 225] images results in [8, 8] feature maps at the output of the last ResNet block. For dense prediction tasks, the ResNet needs to run in fully-convolutional (FCN) mode and global_pool needs to be set to False. The ResNets in [1, 2] all have nominal stride equal to 32 and a good choice in FCN mode is to use output_stride=16 in order to increase the density of the computed features at small computational and memory overhead, cf. http://arxiv.org/abs/1606.00915. Args: inputs: A tensor of size [batch, height_in, width_in, channels]. blocks: A list of length equal to the number of ResNet blocks. Each element is a resnet_utils.Block object describing the units in the block. num_classes: Number of predicted classes for classification tasks. If None we return the features before the logit layer. is_training: whether batch_norm layers are in training mode. global_pool: If True, we perform global average pooling before computing the logits. Set to True for image classification, False for dense prediction. output_stride: If None, then the output will be computed at the nominal network stride. If output_stride is not None, it specifies the requested ratio of input to output spatial resolution. include_root_block: If True, include the initial convolution followed by max-pooling, if False excludes it. If excluded, `inputs` should be the results of an activation-less convolution. reuse: whether or not the network and its variables should be reused. To be able to reuse 'scope' must be given. scope: Optional variable_scope. Returns: net: A rank-4 tensor of size [batch, height_out, width_out, channels_out]. If global_pool is False, then height_out and width_out are reduced by a factor of output_stride compared to the respective height_in and width_in, else both height_out and width_out equal one. If num_classes is None, then net is the output of the last ResNet block, potentially after global average pooling. If num_classes is not None, net contains the pre-softmax activations. end_points: A dictionary from components of the network to the corresponding activation. Raises: ValueError: If the target output_stride is not valid. """ with variable_scope.variable_scope(scope, 'resnet_v2', [inputs], reuse=reuse) as sc: end_points_collection = sc.original_name_scope + '_end_points' with arg_scope( [layers_lib.conv2d, bottleneck, resnet_utils.stack_blocks_dense], outputs_collections=end_points_collection): with arg_scope([layers.batch_norm], is_training=is_training): net = inputs if include_root_block: if output_stride is not None: if output_stride % 4 != 0: raise ValueError( 'The output_stride needs to be a multiple of 4.' ) output_stride /= 4 # We do not include batch normalization or activation functions in # conv1 because the first ResNet unit will perform these. Cf. # Appendix of [2]. with arg_scope([layers_lib.conv2d], activation_fn=None, normalizer_fn=None): net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = layers.max_pool2d(net, [3, 3], stride=2, scope='pool1', padding="SAME") net = resnet_utils.stack_blocks_dense(net, blocks, output_stride) # This is needed because the pre-activation variant does not have batch # normalization or activation functions in the residual unit output. See # Appendix of [2]. net = layers.batch_norm(net, activation_fn=nn_ops.relu, scope='postnorm') if global_pool: # Global average pooling. import tensorflow as tf net = tf.reduce_mean(net, [1, 2], name='pool5', keep_dims=True) if num_classes is not None: net = layers_lib.conv2d(net, num_classes, [1, 1], activation_fn=None, normalizer_fn=None, scope='logits') # Convert end_points_collection into a dictionary of end_points. end_points = utils.convert_collection_to_dict( end_points_collection) if num_classes is not None: end_points['predictions'] = layers.softmax( net, scope='predictions') return net, end_points
def densenet(inputs, num_classes=None, reduction=None, growth_rate=None, num_filters=None, convs_each_num_layers=None, dropout_rate=None, data_format='NHWC', is_training=True, include_root_block=True, reuse=None, base_net=True, global_pool=True, scope=None): assert reduction is not None assert growth_rate is not None assert num_filters is not None assert convs_each_num_layers is not None compression = 1.0 - reduction num_dense_blocks = len(convs_each_num_layers) if data_format == 'NCHW': inputs = tf.transpose(inputs, [0, 3, 1, 2]) with tf.variable_scope(scope, 'densenetxxx', [inputs, num_classes], reuse=reuse) as sc: end_points_collection = sc.name + '_end_points' with slim.arg_scope([slim.batch_norm, slim.dropout], is_training=is_training), \ slim.arg_scope([slim.conv2d, _conv, _conv_block, _dense_block, _transition_block], outputs_collections=end_points_collection), \ slim.arg_scope([_conv], dropout_rate=dropout_rate): net = inputs if include_root_block: # initial convolution # net = slim.conv2d(net, num_filters, 7, stride=2, scope='conv1') # net = slim.batch_norm(net) # net = tf.nn.relu(net) # net = slim.max_pool2d(net, 3, stride=2, padding='SAME') net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1') net = tf.pad(net, [[0, 0], [1, 1], [1, 1], [0, 0]]) net = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID', scope='pool1') if base_net: # blocks for i in range(num_dense_blocks - 1): # dense blocks net, num_filters = _dense_block(net, convs_each_num_layers[i], num_filters, growth_rate, scope='dense_block' + str(i + 1)) # Add transition_block if i == num_dense_blocks - 2: last_one = True else: last_one = False net, num_filters = _transition_block( net, num_filters, compression=compression, last_one=last_one, scope='transition_block' + str(i + 1)) net, num_filters = _dense_block(net, convs_each_num_layers[-1], num_filters, growth_rate, scope='dense_block' + str(num_dense_blocks)) else: for i in range(num_dense_blocks - 1): # dense blocks net, num_filters = _dense_block(net, convs_each_num_layers[i], num_filters, growth_rate) # Add transition_block net, num_filters = _transition_block( net, num_filters, compression=compression) net, num_filters = _dense_block(net, convs_each_num_layers[-1], num_filters, growth_rate) # final blocks if global_pool: with tf.variable_scope('final_block', [inputs]): net = slim.batch_norm(net) net = tf.nn.relu(net) net = _global_avg_pool2d(net, scope='global_avg_pool') if num_classes is not None: net = slim.conv2d(net, num_classes, 1, biases_initializer=tf.zeros_initializer(), scope='logits') end_points = slim.utils.convert_collection_to_dict( end_points_collection) if num_classes is not None: end_points['predictions'] = slim.softmax(net, scope='predictions') return net, end_points
def bottleneck(inputs, depth, depth_bottleneck, stride, rate=1, outputs_collections=None, scope=None): """Bottleneck residual unit variant with BN before convolutions. This is the full preactivation residual unit variant proposed in [2]. See Fig. 1(b) of [2] for its definition. Note that we use here the bottleneck variant which has an extra bottleneck layer. When putting together two consecutive ResNet blocks that use this unit, one should use stride = 2 in the last unit of the first block. Args: inputs: A tensor of size [batch, height, width, channels]. depth: The depth of the ResNet unit output. depth_bottleneck: The depth of the bottleneck layers. stride: The ResNet unit's stride. Determines the amount of downsampling of the units output compared to its input. rate: An integer, rate for atrous convolution. outputs_collections: Collection to add the ResNet unit output. scope: Optional variable_scope. Returns: The ResNet unit's output. """ with variable_scope.variable_scope(scope, 'bottleneck_v2', [inputs]) as sc: depth_in = utils.last_dimension(inputs.get_shape(), min_rank=4) preact = layers.batch_norm(inputs, activation_fn=nn_ops.relu, scope='preact') if depth == depth_in: shortcut = resnet_utils.subsample(inputs, stride, 'shortcut') else: shortcut = layers_lib.conv2d(preact, depth, [1, 1], stride=stride, normalizer_fn=None, activation_fn=None, scope='shortcut') residual = layers_lib.conv2d(preact, depth_bottleneck, [1, 1], stride=1, scope='conv1') residual = resnet_utils.conv2d_same(residual, depth_bottleneck, 3, stride, rate=rate, scope='conv2') residual = layers_lib.conv2d(residual, depth, [1, 1], stride=1, normalizer_fn=None, activation_fn=None, scope='conv3') random_range = 0.12 #if FLAGS.optimal else FLAGS.random_range weight = tf.random_uniform((depth, ), minval=1 - random_range, maxval=1 + random_range) # weight = tf.random_uniform((depth,), minval=5 - random_range, maxval=5 + random_range) output = weight * shortcut + residual # output = weight * shortcut + 0.5*residual return utils.collect_named_outputs(outputs_collections, sc.name, output)
def build_network(self, sess, is_training=True): # select initializers if cfg.TRAIN.TRUNCATED: initializer = tf.truncated_normal_initializer(mean=0.0, stddev=0.01) initializer_bbox = tf.truncated_normal_initializer(mean=0.0, stddev=0.001) else: #initializer = tf.random_normal_initializer(mean=0.0, stddev=0.01) initializer = tf.contrib.layers.xavier_initializer() initializer_bbox = tf.random_normal_initializer(mean=0.0, stddev=0.001) bottleneck = resnet_v1.bottleneck # choose different blocks for different number of layers if self._num_layers == 50: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 5 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] elif self._num_layers == 101: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 22 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] elif self._num_layers == 152: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 7 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 35 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] else: # other numbers are not supported raise NotImplementedError assert (0 <= cfg.RESNET.FIXED_BLOCKS < 4) if cfg.RESNET.FIXED_BLOCKS == 3: with slim.arg_scope(resnet_arg_scope(is_training=False)): net = self.build_base() net_conv4, _ = resnet_v1.resnet_v1(net, blocks[0:cfg.RESNET.FIXED_BLOCKS], global_pool=False, include_root_block=False, scope=self._resnet_scope) elif cfg.RESNET.FIXED_BLOCKS > 0: with slim.arg_scope(resnet_arg_scope(is_training=False)): net = self.build_base() net, _ = resnet_v1.resnet_v1(net, blocks[0:cfg.RESNET.FIXED_BLOCKS], global_pool=False, include_root_block=False, scope=self._resnet_scope) with slim.arg_scope(resnet_arg_scope(is_training=is_training)): net_conv4, _ = resnet_v1.resnet_v1(net, blocks[cfg.RESNET.FIXED_BLOCKS:-1], global_pool=False, include_root_block=False, scope=self._resnet_scope) else: # cfg.RESNET.FIXED_BLOCKS == 0 with slim.arg_scope(resnet_arg_scope(is_training=is_training)): net = self.build_base() net_conv4, _ = resnet_v1.resnet_v1(net, blocks[0:-1], global_pool=False, include_root_block=False, scope=self._resnet_scope) self._act_summaries.append(net_conv4) self._layers['head'] = net_conv4 c=np.zeros((3,5,5)) c[0]=[[-1,2,-2,2,-1],[2,-6,8,-6,2],[-2,8,-12,8,-2],[2,-6,8,-6,2],[-1,2,-2,2,-1]] c[0]=c[0]/12 c[1][1][1]=-1 c[1][1][2]=2 c[1][1][3]=-1 c[1][2][1]=2 c[1][2][2]=-4 c[1][2][3]=2 c[1][3][1]=-1 c[1][3][2]=2 c[1][3][3]=-1 c[1]=c[1]/4 c[2][2][1]=1 c[2][2][2]=-2 c[2][2][3]=1 c[2]=c[2]/2 Wcnn=np.zeros((5,5,3,3)) for i in range(3): #k=i%10+1 #Wcnn[i]=[c[3*k-3],c[3*k-2],c[3*k-1]] Wcnn[:,:,0,i]=c[i] Wcnn[:,:,1,i]=c[i] Wcnn[:,:,2,i]=c[i] if True: with tf.variable_scope('noise'): #kernel = tf.get_variable('weights', #shape=[5, 5, 3, 3], #initializer=tf.constant_initializer(c)) conv = tf.nn.conv2d(self.noise, Wcnn, [1, 1, 1, 1], padding='SAME',name='srm') self._layers['noise']=conv with slim.arg_scope(resnet_arg_scope(is_training=is_training)): #srm_conv = tf.nn.tanh(conv, name='tanh') noise_net = resnet_utils.conv2d_same(conv, 64, 7, stride=2, scope='conv1') noise_net = tf.pad(noise_net, [[0, 0], [1, 1], [1, 1], [0, 0]]) noise_net = slim.max_pool2d(noise_net, [3, 3], stride=2, padding='VALID', scope='pool1') #net_sum=tf.concat(3,[net_conv4,noise_net]) noise_conv4, _ = resnet_v1.resnet_v1(noise_net, blocks[0:-1], global_pool=False, include_root_block=False, scope='noise') with tf.variable_scope(self._resnet_scope, self._resnet_scope): # build the anchors for the image self._anchor_component() # rpn rpn = slim.conv2d(net_conv4, 512, [3, 3], trainable=is_training, weights_initializer=initializer, scope="rpn_conv/3x3") self._act_summaries.append(rpn) rpn_cls_score = slim.conv2d(rpn, self._num_anchors * 2, [1, 1], trainable=is_training, weights_initializer=initializer, padding='VALID', activation_fn=None, scope='rpn_cls_score') # change it so that the score has 2 as its channel size rpn_cls_score_reshape = self._reshape_layer(rpn_cls_score, 2, 'rpn_cls_score_reshape') rpn_cls_prob_reshape = self._softmax_layer(rpn_cls_score_reshape, "rpn_cls_prob_reshape") rpn_cls_prob = self._reshape_layer(rpn_cls_prob_reshape, self._num_anchors * 2, "rpn_cls_prob") rpn_bbox_pred = slim.conv2d(rpn, self._num_anchors * 4, [1, 1], trainable=is_training, weights_initializer=initializer, padding='VALID', activation_fn=None, scope='rpn_bbox_pred') if is_training: rois, roi_scores = self._proposal_layer(rpn_cls_prob, rpn_bbox_pred, "rois") rpn_labels = self._anchor_target_layer(rpn_cls_score, "anchor") # Try to have a determinestic order for the computing graph, for reproducibility with tf.control_dependencies([rpn_labels]): rois, _ = self._proposal_target_layer(rois, roi_scores, "rpn_rois") else: if cfg.TEST.MODE == 'nms': rois, _ = self._proposal_layer(rpn_cls_prob, rpn_bbox_pred, "rois") elif cfg.TEST.MODE == 'top': rois, _ = self._proposal_top_layer(rpn_cls_prob, rpn_bbox_pred, "rois") else: raise NotImplementedError # rcnn if cfg.POOLING_MODE == 'crop': pool5 = self._crop_pool_layer(net_conv4, rois, "pool5") self._layers['pool5']=pool5 #pool5 = self._crop_pool_layer(net_sum, rois, "pool5") else: raise NotImplementedError if True: noise_pool5 = self._crop_pool_layer(noise_conv4, rois, "noise_pool5") with slim.arg_scope(resnet_arg_scope(is_training=is_training)): noise_fc7, _ = resnet_v1.resnet_v1(noise_pool5, blocks[-1:], global_pool=False, include_root_block=False, scope='noise') with slim.arg_scope(resnet_arg_scope(is_training=is_training)): fc7, _ = resnet_v1.resnet_v1(pool5, blocks[-1:], global_pool=False, include_root_block=False, scope=self._resnet_scope) self._layers['fc7']=fc7 with tf.variable_scope('noise_pred'): bilinear_pool=compact_bilinear_pooling_layer(fc7,noise_fc7,2048*8,compute_size=16,sequential=False) fc7=tf.Print(fc7,[tf.shape(fc7)],message='Value of %s' % 'fc', summarize=4, first_n=1) bilinear_pool=tf.reshape(bilinear_pool, [-1,2048*8]) bilinear_pool=tf.Print(bilinear_pool,[tf.shape(bilinear_pool)],message='Value of %s' % 'Blinear', summarize=4, first_n=1) bilinear_pool=tf.multiply(tf.sign(bilinear_pool),tf.sqrt(tf.abs(bilinear_pool)+1e-12)) bilinear_pool=tf.nn.l2_normalize(bilinear_pool,dim=1) noise_cls_score = slim.fully_connected(bilinear_pool, self._num_classes, weights_initializer=tf.contrib.layers.xavier_initializer(), trainable=is_training, activation_fn=None, scope='cls_score') cls_prob = self._softmax_layer(noise_cls_score, "cls_prob") fc7 = tf.reduce_mean(fc7, axis=[1, 2]) bbox_pred = slim.fully_connected(fc7, self._num_classes * 4, weights_initializer=initializer_bbox, trainable=is_training, activation_fn=None, scope='bbox_pred') self._predictions["rpn_cls_score"] = rpn_cls_score self._predictions["rpn_cls_score_reshape"] = rpn_cls_score_reshape self._predictions["rpn_cls_prob"] = rpn_cls_prob self._predictions["rpn_bbox_pred"] = rpn_bbox_pred self._predictions["cls_score"] = noise_cls_score self._predictions["cls_prob"] = cls_prob self._predictions["bbox_pred"] = bbox_pred self._predictions["rois"] = rois self._score_summaries.update(self._predictions) return rois, cls_prob,bbox_pred
def mobilenet_v1_base(inputs, conv_defs, starting_layer=0, min_depth=8, depth_multiplier=1.0, output_stride=None, reuse=None, scope=None): """Mobilenet v1. Constructs a Mobilenet v1 network from inputs to the given final endpoint. Args: inputs: a tensor of shape [batch_size, height, width, channels]. starting_layer: specifies the current starting layer. For region proposal network it is 0, for region classification it is 12 by default. min_depth: Minimum depth value (number of channels) for all convolution ops. Enforced when depth_multiplier < 1, and not an active constraint when depth_multiplier >= 1. depth_multiplier: Float multiplier for the depth (number of channels) for all convolution ops. The value must be greater than zero. Typical usage will be to set this value in (0, 1) to reduce the number of parameters or computation cost of the model. conv_defs: A list of ConvDef named tuples specifying the net architecture. output_stride: An integer that specifies the requested ratio of input to output spatial resolution. If not None, then we invoke atrous convolution if necessary to prevent the network from reducing the spatial resolution of the activation maps. scope: Optional variable_scope. Returns: tensor_out: output tensor corresponding to the final_endpoint. Raises: ValueError: if depth_multiplier <= 0, or convolution type is not defined. """ depth = lambda d: max(int(d * depth_multiplier), min_depth) end_points = {} # Used to find thinned depths for each layer. if depth_multiplier <= 0: raise ValueError('depth_multiplier is not greater than zero.') with tf.variable_scope(scope, 'MobilenetV1', [inputs], reuse=reuse): # The current_stride variable keeps track of the output stride of the # activations, i.e., the running product of convolution strides up to the # current network layer. This allows us to invoke atrous convolution # whenever applying the next convolution would result in the activations # having output stride larger than the target output_stride. current_stride = 1 # The atrous convolution rate parameter. rate = 1 net = inputs for i, conv_def in enumerate(conv_defs): end_point_base = 'Conv2d_%d' % (i + starting_layer) if output_stride is not None and current_stride == output_stride: # If we have reached the target output_stride, then we need to employ # atrous convolution with stride=1 and multiply the atrous rate by the # current unit's stride for use in subsequent layers. layer_stride = 1 layer_rate = rate rate *= conv_def.stride else: layer_stride = conv_def.stride layer_rate = 1 current_stride *= conv_def.stride if isinstance(conv_def, Conv): end_point = end_point_base net = resnet_utils.conv2d_same(net, depth(conv_def.depth), conv_def.kernel, stride=conv_def.stride, scope=end_point) elif isinstance(conv_def, DepthSepConv): end_point = end_point_base + '_depthwise' net = separable_conv2d_same(net, conv_def.kernel, stride=layer_stride, rate=layer_rate, scope=end_point) end_point = end_point_base + '_pointwise' net = slim.conv2d(net, depth(conv_def.depth), [1, 1], stride=1, scope=end_point) else: raise ValueError('Unknown convolution type %s for layer %d' % (conv_def.ltype, i)) return net
def mobilenet_v1_base(inputs, conv_defs, starting_layer=0, min_depth=8, depth_multiplier=1.0, output_stride=None, reuse=None, scope=None): """Mobilenet v1. Constructs a Mobilenet v1 network from inputs to the given final endpoint. Args: inputs: a tensor of shape [batch_size, height, width, channels]. starting_layer: specifies the current starting layer. For region proposal network it is 0, for region classification it is 12 by default. min_depth: Minimum depth value (number of channels) for all convolution ops. Enforced when depth_multiplier < 1, and not an active constraint when depth_multiplier >= 1. depth_multiplier: Float multiplier for the depth (number of channels) for all convolution ops. The value must be greater than zero. Typical usage will be to set this value in (0, 1) to reduce the number of parameters or computation cost of the model. conv_defs: A list of ConvDef named tuples specifying the net architecture. output_stride: An integer that specifies the requested ratio of input to output spatial resolution. If not None, then we invoke atrous convolution if necessary to prevent the network from reducing the spatial resolution of the activation maps. scope: Optional variable_scope. Returns: tensor_out: output tensor corresponding to the final_endpoint. Raises: ValueError: if depth_multiplier <= 0, or convolution type is not defined. """ depth = lambda d: max(int(d * depth_multiplier), min_depth) end_points = {} # Used to find thinned depths for each layer. if depth_multiplier <= 0: raise ValueError('depth_multiplier is not greater than zero.') with tf.variable_scope(scope, 'MobilenetV1', [inputs], reuse=reuse): # The current_stride variable keeps track of the output stride of the # activations, i.e., the running product of convolution strides up to the # current network layer. This allows us to invoke atrous convolution # whenever applying the next convolution would result in the activations # having output stride larger than the target output_stride. current_stride = 1 # The atrous convolution rate parameter. rate = 1 net = inputs for i, conv_def in enumerate(conv_defs): end_point_base = 'Conv2d_%d' % (i + starting_layer) if output_stride is not None and current_stride == output_stride: # If we have reached the target output_stride, then we need to employ # atrous convolution with stride=1 and multiply the atrous rate by the # current unit's stride for use in subsequent layers. layer_stride = 1 layer_rate = rate rate *= conv_def.stride else: layer_stride = conv_def.stride layer_rate = 1 current_stride *= conv_def.stride if isinstance(conv_def, Conv): end_point = end_point_base net = resnet_utils.conv2d_same(net, depth(conv_def.depth), conv_def.kernel, stride=conv_def.stride, scope=end_point) elif isinstance(conv_def, DepthSepConv): end_point = end_point_base + '_depthwise' net = separable_conv2d_same(net, conv_def.kernel, stride=layer_stride, rate=layer_rate, scope=end_point) end_point = end_point_base + '_pointwise' net = slim.conv2d(net, depth(conv_def.depth), [1, 1], stride=1, scope=end_point) else: raise ValueError('Unknown convolution type %s for layer %d' % (conv_def.ltype, i)) return net
def build_network(self, sess, is_training=True): # select initializers if cfg.TRAIN.TRUNCATED: initializer = tf.truncated_normal_initializer(mean=0.0, stddev=0.01) initializer_bbox = tf.truncated_normal_initializer(mean=0.0, stddev=0.001) else: initializer = tf.random_normal_initializer(mean=0.0, stddev=0.01) initializer_bbox = tf.random_normal_initializer(mean=0.0, stddev=0.001) bottleneck = resnet_v1.bottleneck # choose different blocks for different number of layers if self._num_layers == 50: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 5 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] elif self._num_layers == 101: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 3 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 22 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] elif self._num_layers == 152: blocks = [ resnet_utils.Block('block1', bottleneck, [(256, 64, 1)] * 2 + [(256, 64, 2)]), resnet_utils.Block('block2', bottleneck, [(512, 128, 1)] * 7 + [(512, 128, 2)]), # Use stride-1 for the last conv4 layer resnet_utils.Block('block3', bottleneck, [(1024, 256, 1)] * 35 + [(1024, 256, 1)]), resnet_utils.Block('block4', bottleneck, [(2048, 512, 1)] * 3) ] else: # other numbers are not supported raise NotImplementedError assert (0 <= cfg.RESNET.FIXED_BLOCKS < 4) if cfg.RESNET.FIXED_BLOCKS == 3: with slim.arg_scope(resnet_arg_scope(is_training=False)): net = self.build_base() net_conv4, _ = resnet_v1.resnet_v1( net, blocks[0:cfg.RESNET.FIXED_BLOCKS], global_pool=False, include_root_block=False, scope=self._resnet_scope) elif cfg.RESNET.FIXED_BLOCKS > 0: with slim.arg_scope(resnet_arg_scope(is_training=False)): net = self.build_base() net, _ = resnet_v1.resnet_v1(net, blocks[0:cfg.RESNET.FIXED_BLOCKS], global_pool=False, include_root_block=False, scope=self._resnet_scope) with slim.arg_scope(resnet_arg_scope(is_training=is_training)): net_conv4, _ = resnet_v1.resnet_v1( net, blocks[cfg.RESNET.FIXED_BLOCKS:-1], global_pool=False, include_root_block=False, scope=self._resnet_scope) else: # cfg.RESNET.FIXED_BLOCKS == 0 with slim.arg_scope(resnet_arg_scope(is_training=is_training)): net = self.build_base() net_conv4, _ = resnet_v1.resnet_v1(net, blocks[0:-1], global_pool=False, include_root_block=False, scope=self._resnet_scope) self._act_summaries.append(net_conv4) self._layers['head'] = net_conv4 if False: with tf.variable_scope('noise'): #kernel = tf.get_variable('weights', #shape=[5, 5, 3, 3], #initializer=tf.constant_initializer(c)) conv = tf.nn.conv2d(self.noise, Wcnn, [1, 1, 1, 1], padding='SAME', name='srm') self._layers['noise'] = conv with slim.arg_scope(resnet_arg_scope(is_training=is_training)): #srm_conv = tf.nn.tanh(conv, name='tanh') noise_net = resnet_utils.conv2d_same(conv, 64, 7, stride=2, scope='conv1') noise_net = tf.pad(noise_net, [[0, 0], [1, 1], [1, 1], [0, 0]]) noise_net = slim.max_pool2d(noise_net, [3, 3], stride=2, padding='VALID', scope='pool1') #net_sum=tf.concat(3,[net_conv4,noise_net]) noise_conv4, _ = resnet_v1.resnet_v1(noise_net, blocks[0:-1], global_pool=False, include_root_block=False, scope='noise') with tf.variable_scope(self._resnet_scope, self._resnet_scope): # build the anchors for the image self._anchor_component() # rpn rpn = slim.conv2d(net_conv4, 512, [3, 3], trainable=is_training, weights_initializer=initializer, scope="rpn_conv/3x3") self._act_summaries.append(rpn) rpn_cls_score = slim.conv2d(rpn, self._num_anchors * 2, [1, 1], trainable=is_training, weights_initializer=initializer, padding='VALID', activation_fn=None, scope='rpn_cls_score') # change it so that the score has 2 as its channel size rpn_cls_score_reshape = self._reshape_layer( rpn_cls_score, 2, 'rpn_cls_score_reshape') rpn_cls_prob_reshape = self._softmax_layer(rpn_cls_score_reshape, "rpn_cls_prob_reshape") rpn_cls_prob = self._reshape_layer(rpn_cls_prob_reshape, self._num_anchors * 2, "rpn_cls_prob") rpn_bbox_pred = slim.conv2d(rpn, self._num_anchors * 4, [1, 1], trainable=is_training, weights_initializer=initializer, padding='VALID', activation_fn=None, scope='rpn_bbox_pred') if is_training: rois, roi_scores = self._proposal_layer( rpn_cls_prob, rpn_bbox_pred, "rois") rpn_labels = self._anchor_target_layer(rpn_cls_score, "anchor") # Try to have a determinestic order for the computing graph, for reproducibility with tf.control_dependencies([rpn_labels]): rois, _ = self._proposal_target_layer( rois, roi_scores, "rpn_rois") else: if cfg.TEST.MODE == 'nms': rois, _ = self._proposal_layer(rpn_cls_prob, rpn_bbox_pred, "rois") elif cfg.TEST.MODE == 'top': rois, _ = self._proposal_top_layer(rpn_cls_prob, rpn_bbox_pred, "rois") else: raise NotImplementedError # rcnn if cfg.POOLING_MODE == 'crop': pool5 = self._crop_pool_layer(net_conv4, rois, "pool5") #pool5 = self._crop_pool_layer(net_sum, rois, "pool5") else: raise NotImplementedError if False: noise_pool5 = self._crop_pool_layer(noise_conv4, rois, "noise_pool5") with slim.arg_scope(resnet_arg_scope(is_training=is_training)): noise_fc7, _ = resnet_v1.resnet_v1(noise_pool5, blocks[-1:], global_pool=False, include_root_block=False, scope='noise') with slim.arg_scope(resnet_arg_scope(is_training=is_training)): fc7, _ = resnet_v1.resnet_v1(pool5, blocks[-1:], global_pool=False, include_root_block=False, scope=self._resnet_scope) self._layers['fc7'] = fc7 with tf.variable_scope(self._resnet_scope, self._resnet_scope): #pdb.set_trace() #noise_fc7 = tf.reduce_mean(noise_fc7, axis=[1, 2]) #bilinear_pool=compact_bilinear_pooling_layer(fc7,noise_fc7,2048*4,compute_size=16,sequential=False) #bilinear_pool=tf.reshape(bilinear_pool, [-1,2048*4]) fc7 = tf.reduce_mean(fc7, axis=[1, 2]) cls_score = slim.fully_connected(fc7, self._num_classes, weights_initializer=initializer, trainable=is_training, activation_fn=None, scope='cls_score') #pdb.set_trace() #noise_cls_score = slim.fully_connected(bilinear_pool, self._num_classes, weights_initializer=initializer, #trainable=is_training, activation_fn=None, scope='noise_cls_score') cls_prob = self._softmax_layer(cls_score, "cls_prob") bbox_pred = slim.fully_connected( fc7, self._num_classes * 4, weights_initializer=initializer_bbox, trainable=is_training, activation_fn=None, scope='bbox_pred') #with tf.variable_scope(self._resnet_scope, self._resnet_scope): # Average pooling done by reduce_mean #fc7 = tf.reduce_mean(fc7, axis=[1, 2]) #fc_con=tf.concat(1,[fc7,noise_fc]) #cls_score = slim.fully_connected(fc7, self._num_classes, weights_initializer=initializer, #trainable=False, activation_fn=None, scope='cls_score') #cls_score1=cls_score+10*noise_cls_score #cls_prob = self._softmax_layer(noise_cls_score, "cls_prob") #bbox_pred = slim.fully_connected(fc7, self._num_classes * 4, weights_initializer=initializer_bbox, #trainable=False, #activation_fn=None, scope='bbox_pred') self._predictions["rpn_cls_score"] = rpn_cls_score self._predictions["rpn_cls_score_reshape"] = rpn_cls_score_reshape self._predictions["rpn_cls_prob"] = rpn_cls_prob self._predictions["rpn_bbox_pred"] = rpn_bbox_pred self._predictions["cls_score"] = cls_score self._predictions["cls_prob"] = cls_prob self._predictions["bbox_pred"] = bbox_pred self._predictions["rois"] = rois self._score_summaries.update(self._predictions) return rois, cls_prob, bbox_pred