Ever wanted an easier way to collect your entire dataset captions into one file for use as a wildcard to test your LoRA/Checkpoint? Look no further!

1. Download txtCombine.py script to your desired location.

2. Right click and edit with your favorite text editor(Notepad++)

3. Lines 4 and 6 are all you need to change, input and output respectively. Change input to the folder with your entire captioned dataset containing the .txt files. Change output where you want the script to save the combined wildcard file(be sure to use proper \\ formatting i.e C\\path\\to\\captions or else the script will not work).

4. Run by simply double clicking the txtCombine.py file.

5. Now the output file is a perfectly organized single Wildcard file. Place this where your other wildcards are and start genning!