diff --git a/resolve.py b/resolve.py
index b760e4e..c1fc908 100644
--- a/resolve.py
+++ b/resolve.py
@@ -58,7 +58,7 @@ def super_resolve(img: str, halve: bool = False):
     sr_img_srresnet = tf.squeeze(sr_img_srresnet['output_0'])
     sr_img_srresnet = transform.convert_image(sr_img_srresnet, source='[-1, 1]', target='pil')
 
-    sr_img_srgan = resnet_inference(lr_img)
+    sr_img_srgan = generator_inference(lr_img)
     sr_img_srgan = tf.squeeze(sr_img_srgan['output_0'])
     sr_img_srgan = transform.convert_image(sr_img_srgan, source='[-1, 1]', target='pil')
 
diff --git a/train.py b/train.py
index d02f8e3..c7c7576 100644
--- a/train.py
+++ b/train.py
@@ -19,7 +19,7 @@
 large_kernel_size = 9  # kernel size of the first and last convolutions which transform the inputs and outputs
 small_kernel_size = 3  # kernel size of the first and last convolutions which transform the inputs and outputs
 n_channels = 64  # number of channels in-between, input and output channels for residual & subpixel conv blocks
-n_blocks = 16  # number of residual blocks
+n_blocks = 32  # number of residual blocks
 srresnet_checkpoint = "SuperResolutionResNet_9999"  # trained SRResNet checkpoint used for initialization
 
 # Discriminator parameters
@@ -94,4 +94,4 @@ def main(architecture_type: str = "resnet"):
 
 
 if __name__ == "__main__":
-    main(architecture_type="gan")
+    main(architecture_type="resnet")