Added counting of Dense mapping layer

Counting Dense mapping layer and convert generator and g_ema accordingly

Author: levindabhi

convert_weight.py

@@ -137,10 +137,10 @@ def discriminator_fill_statedict(statedict, vars, size):
     return statedict
 
 
-def fill_statedict(state_dict, vars, size):
+def fill_statedict(state_dict, vars, size, n_mlp):
     log_size = int(math.log(size, 2))
 
-    for i in range(8):
+    for i in range(n_mlp):
         update(state_dict, convert_dense(vars, f"G_mapping/Dense{i}", f"style.{i + 1}"))
 
     update(
@@ -237,9 +237,15 @@ def fill_statedict(state_dict, vars, size):
 
     size = g_ema.output_shape[2]
 
-    g = Generator(size, 512, 8, channel_multiplier=args.channel_multiplier)
+    n_mlp = 0
+    mapping_layers_names = g_ema.__getstate__()['components']['mapping'].list_layers()
+    for layer in mapping_layers_names:
+        if layer[0].startswith('Dense'):
+            n_mlp += 1
+
+    g = Generator(size, 512, n_mlp, channel_multiplier=args.channel_multiplier)
     state_dict = g.state_dict()
-    state_dict = fill_statedict(state_dict, g_ema.vars, size)
+    state_dict = fill_statedict(state_dict, g_ema.vars, size, n_mlp)
 
     g.load_state_dict(state_dict)
 
@@ -248,7 +254,7 @@ def fill_statedict(state_dict, vars, size):
     ckpt = {"g_ema": state_dict, "latent_avg": latent_avg}
 
     if args.gen:
-        g_train = Generator(size, 512, 8, channel_multiplier=args.channel_multiplier)
+        g_train = Generator(size, 512, n_mlp, channel_multiplier=args.channel_multiplier)
         g_train_state = g_train.state_dict()
         g_train_state = fill_statedict(g_train_state, generator.vars, size)
         ckpt["g"] = g_train_state
@@ -292,5 +298,4 @@ def fill_statedict(state_dict, vars, size):
 
     utils.save_image(
         img_concat, name + ".png", nrow=n_sample, normalize=True, range=(-1, 1)
-    )
-
+    )