update documentation

README.md

@@ -77,6 +77,7 @@ Fork does differ in few things:
 - Uses simplified folder structure  
   e.g. `/train`, `/outputs/*`  
 - Modified training templates  
+- End-to-end `LoRA` training support
 
 Only Python library which is not auto-updated is `PyTorch` itself as that is very system specific  
 For some Torch optimizations notes, see Wiki

cli/README.md

@@ -37,10 +37,10 @@ LoRA training
 > python train-lora.py
 
 Combined pipeline:
-1. Creates embedding  
-2. Extracts images if input is movie  
-3. Preprocesses images  
-4. Runs training  
+1. Preprocesses images  
+2. Runs training  
+
+[Detailed documentation](https://github.com/vladmandic/automatic/wiki/Process.md)
 
 LoRA extract from model
 > python moidules/lora-extract.py
@@ -112,6 +112,8 @@ Note: Offline except for interrogate to generate caption files which requires SD
 
 > python modules/process.py
 
+[Detailed documentation](https://github.com/vladmandic/automatic/wiki/Process.md)
+
 ### Prompt Ideas
 
 Generate complex prompt ideas

cli/modules/process.py

@@ -432,7 +432,7 @@ def unload_models():
 
 if __name__ == '__main__':
     # log.setLevel(logging.DEBUG)
-    parser = argparse.ArgumentParser(description = 'image watermarking')
+    parser = argparse.ArgumentParser(description = 'dataset processor')
     parser.add_argument('--output', type=str, required=True, help='folder to store images')
     parser.add_argument('--preview', default=False, action='store_true', help = "run processing but do not store results")
     parser.add_argument('--offline', default=False, action='store_true', help = "run only processing steps that do not require running server")