grammar

Signed-off-by: Can-Zhao <[email protected]>

Author: Can-Zhao

generation/maisi/README.md

@@ -84,12 +84,9 @@ When `autoencoder_sliding_window_infer_size` is equal to or larger than the late
 
 
 ### Training GPU Memory Usage
-The VAE is trained on patches and can be trained using a 16G GPU if the patch size is set to a small value, such as [64, 64, 64].  
-Users can adjust the patch size to fit the available GPU memory.  
-For the released model, we initially trained the autoencoder on a 16G V100 GPU with a small patch size of [64, 64, 64], and then continued training on a 32G V100 GPU with a larger patch size of [128, 128, 128].
+The VAE is trained on patches and can be trained using a 16G GPU if the patch size is set to a small value, such as [64, 64, 64]. Users can adjust the patch size to fit the available GPU memory. For the released model, we initially trained the autoencoder on a 16G V100 GPU with a small patch size of [64, 64, 64], and then continued training on a 32G V100 GPU with a larger patch size of [128, 128, 128].
 
-The DM and ControlNet are trained on whole images rather than patches.  
-The GPU memory usage during training depends on the size of the input images.
+The DM and ControlNet are trained on whole images rather than patches. The GPU memory usage during training depends on the size of the input images.
 
 |  image size  |  latent size  | Peak Memory |
 |--------------|:------------- |:-----------:|