Commit 26de6b23 authored by Laanait, Nouamane's avatar Laanait, Nouamane
Browse files

removing hard-coded parameters from ynet architecture generation

parent 8fa8b888

stemdl/losses.py

+2 −2
Original line number Diff line number Diff line
@@ -108,8 +108,8 @@ def calc_loss(n_net, scope, hyper_params, params, labels, step=None, images=None 
    #Assemble all of the losses.

    losses = tf.get_collection(tf.GraphKeys.LOSSES)

    if hyper_params['network_type'] == 'YNet':

        #losses = [inverter_loss , decoder_loss_re, decoder_loss_im, 0.01 * reg_loss]

        losses = [inverter_loss , decoder_loss_re, decoder_loss_im]

        losses = [inverter_loss , decoder_loss_re, decoder_loss_im, reg_loss]

        # losses = [inverter_loss , decoder_loss_re, decoder_loss_im]

        # losses, prefac = ynet_adjusted_losses(losses, step)

        # tf.summary.scalar("prefac_inverter", prefac)

        # losses = [inverter_loss]

stemdl/network.py

+2 −2
Original line number Diff line number Diff line
@@ -611,7 +611,7 @@ class ConvNet: 
            input= tf.cast(input, tf.float32)

    #    with tf.variable_scope('layer_normalization', reuse=None) as scope:

    #         output = tf.keras.layers.LayerNormalization(trainable=False)(inputs=input)

        mean , variance = tf.nn.moments(input, axes=[2,3], keepdims=True)

        mean , variance = tf.nn.moments(input, axes=[2,3], keep_dims=True)

        output = (input - mean)/ (tf.sqrt(variance) + 1e-7)

        if self.params['IMAGE_FP16']:

            output = tf.cast(output, tf.float16)
@@ -2861,7 +2861,7 @@ class YNet(FCDenseNet, FCNet): 
        # out = tf.transpose(out, perm= [1, 2, 0])

        dim = int(math.sqrt(self.images.shape.as_list()[1]))

        out = tf.reshape(out, [self.params['batch_size'], -1, dim, dim])



        self.print_rank('decoder reshape:', out.shape.as_list())

        with tf.variable_scope('%s_conv_1by1' % subnet, reuse=self.reuse) as scope:

            out, _ = self._conv(input=out, params=conv_1by1) 

            do_bn = conv_1by1.get('batch_norm', False)

stemdl/network_utils.py

+3 −4
Original line number Diff line number Diff line
@@ -403,7 +403,6 @@ def generate_YNet_json(save= True, out_dir='json_files', n_pool=3, n_layers_per_ 
    fc_cvae = OrderedDict({'type': 'fully_connected','weights': fc_dim,'bias': fc_dim, 'activation': activation,

                                'regularize': True})

    cvae_model = OrderedDict({'n_conv_layers': 4, 'n_fc_layers':fc_layers,'fc_params': fc_cvae, 'conv_params':conv_cvae}) 

    init_features = 1024

    freq2space_block = OrderedDict({'type': 'freq2space', 'activation': activation, 'dropout': dropout_prob, 

                                    'init_features':init_features, 'batch_norm': batch_norm})

    freq2space_block['type'] = 'freq2space_CVAE'
@@ -423,7 +422,7 @@ def generate_YNet_json(save= True, out_dir='json_files', n_pool=3, n_layers_per_ 
                            'activation': activation, 'padding': 'SAME', 'batch_norm': batch_norm, 'dropout':dropout_prob})

    deconv_layer_base = OrderedDict({'type': "deconv_2D", 'stride': [2, 2], 'kernel': [4,4], 'features': None, 

                        'padding': 'SAME', 'upsample': pool['kernel'][0]})

    features = 1024

    features = init_features

    rank = 0

    for i in range(n_pool+1):

        deconv_layer = deepcopy(deconv_layer_base)
@@ -449,7 +448,7 @@ def generate_YNet_json(save= True, out_dir='json_files', n_pool=3, n_layers_per_ 
    #############################

    layers_params_list = []

    layers_keys_list = []

    features = 1024

    features = init_features

    rank = 0

    for i in range(n_pool+1):

        deconv_layer = deepcopy(deconv_layer_base)
@@ -483,7 +482,7 @@ def generate_YNet_json(save= True, out_dir='json_files', n_pool=3, n_layers_per_ 
                            'activation': activation, 'padding': 'SAME', 'batch_norm': batch_norm, 'dropout':dropout_prob})

    deconv_layer_base = OrderedDict({'type': "deconv_2D", 'stride': [2, 2], 'kernel': [4,4], 'features': None, 

                        'padding': 'SAME', 'upsample': pool['kernel'][0]})

    features = 1024

    features = init_features

    rank = 0

    for i in range(n_pool+1):

        deconv_layer = deepcopy(deconv_layer_base)