From 9fbd09386c8beb145c8ac04dfa812e0cbfccda78 Mon Sep 17 00:00:00 2001 From: Hunor Laczko Date: Tue, 21 Feb 2023 21:08:05 +0000 Subject: [PATCH] Revert "fixed changed reverse?" This reverts commit ac273198602ee7a2a046833cf55b920d3b9b6b02. --- configs/config_abl_all_post.py | 14 +++++++------- convnext/models/convnext_tf.py | 29 +++++++++++++---------------- 2 files changed, 20 insertions(+), 23 deletions(-) diff --git a/configs/config_abl_all_post.py b/configs/config_abl_all_post.py index 2f3373a..954a55f 100644 --- a/configs/config_abl_all_post.py +++ b/configs/config_abl_all_post.py @@ -19,9 +19,9 @@ config['only_curvatures'] = False ablation_count = [config['use_skinning'], config['remove_global_orientation'], config['use_curvature'], config['use_extra_losses'], config['use_offsets']].count(False) assert ablation_count == 1 or ablation_count == 0 -config['gpu'] = '5' -config['name'] = 'iccv_weights_test_8' -config['notes'] = 'smaller weight loss coeff, 1e-6, convolutional weight learning, changed convnext reverse' +config['gpu'] = '4' +config['name'] = 'iccv_weights_test_6' +config['notes'] = 'smaller weight loss coeff, 1e-6, convolutional weight learning, continued from _5' config['avg_garment_path'] = 'custom_data/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv if config['use_skinning']: @@ -53,7 +53,7 @@ if config['is_cape']: config['smpl_res'] = 3638 -config['pretrained_model'] = '' +config['pretrained_model'] = 'checkpoints/post/iccv_weights_test_5' ############### config['lr'] = 0.0001# 0.0001 @@ -91,9 +91,9 @@ config['cvae']['noise_amount'] = 0.00001 config['cvae']['kernel_size'] = 7 config['cvae']['encoder_depths'] = [3, 3, 3, 9, 3] config['cvae']['encoder_dims'] = [32, 64, 128, 256, 64] -config['cvae']['decoder_depths'] = [3, 9, 3, 3, 3, 3] -config['cvae']['decoder_dims'] = [3, 32, 128, 128, 256, 512] -config['cvae']['decoder_weights_depths'] = [3, 9, 3, 3, 3] +config['cvae']['decoder_depths'] = config['cvae']['encoder_depths'] +config['cvae']['decoder_dims'] = config['cvae']['encoder_dims'] +config['cvae']['decoder_weights_depths'] = [3, 3, 3, 5, 3] config['cvae']['decoder_weights_dims'] = [24, 32, 64, 128, 256] config['cond_ae_static'] = dict() diff --git a/convnext/models/convnext_tf.py b/convnext/models/convnext_tf.py index f9380bf..e0fc3a4 100644 --- a/convnext/models/convnext_tf.py +++ b/convnext/models/convnext_tf.py @@ -225,28 +225,26 @@ class ConvNeXtReverse(tf.keras.Model): drop_path_rate=0., layer_scale_init_value=1e-6, head_init_scale=1., learn_weights=False, ): super().__init__() - self.upsample_layers = [] - + self.upsample_layers = [] # stem and 3 intermediate downsampling conv layers stem = tf.keras.Sequential([ - LayerNormalization(epsilon=1e-6), - Conv2DTranspose(dims[0], kernel_size=2, strides=2, padding='same'), - ]) + Conv2DTranspose(dims[0], kernel_size=4, strides=4, padding='same'), + # UpSampling2D((4, 4), interpolation='nearest'), + # Conv2D(dims[0], kernel_size=4, strides=1, padding='same'), + # Dense(dims[0]), + LayerNormalization(epsilon=1e-6)] + ) self.upsample_layers.append(stem) - - for i in range(len(depths) - 2): + for i in range(len(depths) - 1): downsample_layer = tf.keras.Sequential([ LayerNormalization(epsilon=1e-6), - Conv2DTranspose(dims[i+1], kernel_size=2, strides=2, padding='same'), + Conv2DTranspose(dims[i+1], kernel_size=2, strides=2, padding='same') + # UpSampling2D((2,2), interpolation='nearest'), + # Conv2D(dims[i+1], kernel_size=2, strides=1, padding='same'), + # Dense(dims[i+1]), ]) self.upsample_layers.append(downsample_layer) - downsample_layer = tf.keras.Sequential([ - Conv2DTranspose(dims[i+2], kernel_size=2, strides=2, padding='same'), - LayerNormalization(epsilon=1e-6), - ]) - self.upsample_layers.append(downsample_layer) - - self.stages = [] + self.stages = [] # 4 feature resolution stages, each consisting of multiple residual blocks dp_rates = [x for x in np.linspace(0, drop_path_rate, sum(depths))] cur = 0 for i in range(len(depths)): @@ -259,7 +257,6 @@ class ConvNeXtReverse(tf.keras.Model): if not learn_weights: self.last = Dense(3, activation='tanh') - # self.last = Activation('tanh') else: self.last = Activation('softmax') # self.last = Dense(24, activation='softmax')