diff --git a/.gitignore b/.gitignore
index 8eb1899..9fe5204 100644
--- a/.gitignore
+++ b/.gitignore
@@ -11,6 +11,7 @@ custom
 ffhq
 afhq
 logs
+abandon
 
 # Byte-compiled / optimized / DLL files
 __pycache__/
diff --git a/inference.py b/inference.py
index 0f890c9..87e1e4c 100644
--- a/inference.py
+++ b/inference.py
@@ -25,25 +25,35 @@ def __init__(self,
         self.ckpt = tf.train.Checkpoint(generator_clone=self.Gs)
         print(f'Loading network from {checkpoint_path}...')
         self.ckpt.restore(tf.train.latest_checkpoint(checkpoint_path)).expect_partial()
-  
+        tf.config.optimizer.set_experimental_options({"auto_mixed_precision": True})    # Enable mixed precision
+
+
+    def _get_labels(self, label, length):
+        if self.num_labels > 0: # Contain labels
+            labels_indice = [label]*length if label is not None else tf.random.uniform([length], 0, self.num_labels, dtype=tf.int32)
+            labels = tf.one_hot(labels_indice, self.num_labels)
+        else:  
+            labels_indice = [0]*length
+            labels = tf.zeros([length, 0], tf.float32)
+
+        return labels, labels_indice
+
 
     def genetate_example(self, num_examples, batch_size=1, label=None):
         create_dir(f'{self.result_path}/example')
         print('Generating images...')
         for begin in range(0, num_examples, batch_size):
             latents = tf.random.normal([batch_size, 512])
-            labels_indice = [label]*batch_size if label is not None else tf.random.uniform([batch_size], 0, self.num_labels, dtype=tf.int32)
-            labels = tf.one_hot(labels_indice, self.num_labels) if self.num_labels > 0 else tf.zeros([batch_size, 0], tf.float32)
+            labels, labels_indice = self._get_labels(label, batch_size)
             images = self.Gs([latents, labels], self.truncation_psi, training=False)
-            for idx, (image, label) in enumerate(zip(images, labels_indice)):
-                imsave(image, f'{self.result_path}/example/{begin+idx}_label-{label}.jpg')
+            for idx, (image, indice) in enumerate(zip(images, labels_indice)):
+                imsave(image, f'{self.result_path}/example/{begin+idx}_label-{indice}.jpg')
 
 
     def style_mixing_example(self, row_seeds, col_seeds, label=None, col_styles='0-6'):
         create_dir(f'{self.result_path}/mixing')
         all_seeds = list(set(row_seeds + col_seeds))
-        all_labels_indice = [label]*len(all_seeds) if label is not None else tf.random.uniform([len(all_seeds)], 0, self.num_labels, dtype=tf.int32)
-        all_labels = tf.one_hot(all_labels_indice, self.num_labels) if self.num_labels > 0 else tf.zeros([len(all_seeds), 0], tf.float32)
+        all_labels, all_labels_indices = self._get_labels(label, len(all_seeds))
         all_z = tf.stack([tf.random.normal([512], seed=seed) for seed in all_seeds])    # [minibatch, component]
 
         print('Generating images...')
@@ -88,8 +98,8 @@ def generate_gif(self, output='test', label=None, num_rows=2, num_cols=3, resolu
         output_seq = []
         batch_size = num_rows * num_cols
         latents = [tf.random.normal([batch_size, 512]) for _ in range(num_phases)]
-        labels_indice = [label]*batch_size if label is not None else tf.random.uniform([batch_size], 0, self.num_labels, dtype=tf.int32)
-        labels = [tf.one_hot(labels_indice, self.num_labels) if self.num_labels > 0 else tf.zeros([batch_size, 0], tf.float32) for _ in range(num_phases)]
+        labels, labels_indice = self._get_labels(label, batch_size)
+        labels = tf.repeat(tf.expand_dims(labels, axis=0), repeats=num_phases, axis=0)
 
         def to_image_grid(outputs):
             outputs = (outputs + 1) * 127.5
diff --git a/main.py b/main.py
index 70d8857..1afc45c 100644
--- a/main.py
+++ b/main.py
@@ -26,7 +26,7 @@ def main():
     parser.add_argument('--total_img', help='Training length of images', type=int, default=25000000)
     parser.add_argument('--impl', help="(Faster)Custom op use:'cuda'; (Slower)Tensorflow op use:'ref'", type=str, default='ref', choices=['ref','cuda'])
     # Inference
-    parser.add_argument('--label', help='Inference label', type=int, default=0)
+    parser.add_argument('--label', help='Inference label', type=int, default=None)
     parser.add_argument('--num_labels', help='Number of labels', type=int, default=0)
     parser.add_argument('--truncation_psi', help='Inference truncation psi', type=float, default=0.5)
     parser.add_argument('--mode', help="Inference mode be one of: 'example', 'gif', 'mixing'", type=str, default='example', choices=['example','gif','mixing'])
@@ -71,7 +71,6 @@ def main():
         assert 0.0 <= args.truncation_psi <= 1.0, 'Error: Inference truncation_psi needs to be between 0 and 1.'
         assert args.res >= 4
         assert args.num_labels >= 0
-        assert 0 <= args.label <= max(0, args.num_labels-1)
 
         parameters = {
             'resolution' : args.res,
diff --git a/modules/generator.py b/modules/generator.py
index 334f177..87406fe 100644
--- a/modules/generator.py
+++ b/modules/generator.py
@@ -205,7 +205,7 @@ def truncation_trick(self, w_latents, truncation_psi):
         truncated_w = self.w_avg + (w_latents - self.w_avg) * layer_psi  
         return truncated_w
 
-
+    @tf.function
     def call(self, inputs, truncation_psi=0.5, return_latents=False, training=None):
         latents_in, labels_in = inputs