Update README.md

README.md

@@ -12,23 +12,22 @@ A framework for human-informed reinforcement learning by subjective logic
     - `model.py` - Model classes
   - Hint/SL modules
     - `parser.py` - Human hint parser. Parses hints from `files/opinions.txt`
-    - `files/opinions.txt` - Human input. Format:
+    - `files/opinions(...).txt` - Human input. Naming convention: `opinions-[SIZE]x[SIZE]-seed[SEED].txt` Format:
       ```
-      grid size[1]
-      uncertainty[1]
-      hints[*]
+      grid size [1]
+      uncertainty [1]
+      hints [*]
       ```
     - `sl.py` - Subjective logic utilities
-  - Map modules
-    - `map_generator.py` - Generates map for human inspection. Saves maps under `/files` as `.xslx` files.
-    - `map_parser.py` - Parses map for experiments
+  - Map module
+    - `map_tools.py` - Generator, renderer, and parser for maps. Saves maps under `/files` as `.xslx` files.
 - [/tests](https://github.com/dagenaik/Uncertainty-in-Reinforcement-Learning/tree/main/tests) - Unit tests
 
 # Setup guide
 - Clone this repository.
 - Install requirements via ```pip install -r requirements.txt```.
 
 # How to use
-- Set `SEED` and `SIZE` in `src/map_parser.py` and run `python .\src\map_parser.py`
+- Generate a map by running `python .\src\map_tools.py -generate -render -size [SIZE] -seed [SEED]` -- Replace `[SIZE]` and `[SEED]` with the values (int) you need. The `-render` flag is optional.
 - Create an opinion file with the following name: `opinions-[SIZE]x[SIZE]-seed[SEED].txt` (e.g., `opinions-6x6-seed10.txt`)
-- Set the same `SEED` and `SIZE` in `src/runner.py` and run `python .\src\runner.py`
+- Run the experiment using `python .\src\runner.py -size [SIZE] -seed [SEED] -log [DEBUG_LEVEL]` -- Replace `[SIZE]` and `[SEED]` with the values (int) you need. The `[DEBUG_LEVEL]` value is one of the following: `critical`, `error`, `warn`, `warning`, `info`, `debug`.