prototype conditional pipeline

load_models.py

@@ -18,11 +18,11 @@ def load_generator(g_params=None, is_g_clone=False, ckpt_dir=None, custom_cuda=T
         }
 
     test_latent = tf.ones((1, g_params['z_dim']), dtype=tf.float32)
-    test_labels = tf.ones((1, g_params['labels_dim']), dtype=tf.float32)
+    test_conditions = (tf.ones((1, 72), dtype=tf.float32), tf.ones((1, 11), dtype=tf.float32), tf.ones((1, 3, 512, 256), dtype=tf.float32))
 
     # build generator model
     generator = Generator(g_params)
-    _ = generator([test_latent, test_labels])
+    _ = generator([test_latent, test_conditions])
 
     if ckpt_dir is not None:
         if is_g_clone:
@@ -51,11 +51,10 @@ def load_discriminator(d_params=None, ckpt_dir=None, custom_cuda=True):
 
     res = d_params['resolutions'][-1]
     test_images = tf.ones((1, 3, res, int(res/2)), dtype=tf.float32)
-    test_labels = tf.ones((1, d_params['labels_dim']), dtype=tf.float32)
 
     # build discriminator model
     discriminator = Discriminator(d_params)
-    _ = discriminator([test_images, test_labels])
+    _ = discriminator(test_images)
 
     if ckpt_dir is not None:
         ckpt = tf.train.Checkpoint(discriminator=discriminator)