explain model selection in the readme, fix missing model selection argument in generate_embeddings_batch

README.md

@@ -123,6 +123,26 @@ quizgen <command> [options]
     quizgen anki --input-dir path/to/csv/files --root-deck-name "Root Deck Name" 
     ```
 
+## Model Selection
+
+QuizGen allows you to select specific AI models for different stages of the question generation process. This can be useful for experimenting with different models or fine-tuning the output to your needs.
+
+The following model selection arguments are available:
+
+- `--concept-model`: Specifies the model used for concept extraction. Default is `gpt-4o-mini-2024-07-18`.
+- `--questions-model`: Specifies the model used for question generation. Default is `gpt-4o-2024-08-06`.
+- `--embedding-model`: Specifies the model used for generating embeddings. Default is `text-embedding-3-small`.
+
+These arguments can be used with both single chapter and batch commands. For example:
+
+```bash
+# Generate questions using a specific model:
+quizgen generate --chapter-path path/to/chapter.md --title "Course Title" --output path/to/output.json --questions-model gpt-4
+
+# Generate embeddings using a specific model:
+quizgen embeddings --batch --input-dir path/to/json/files --output-dir path/to/embeddings/output --embedding-model text-embedding-ada-002
+```
+
 ## Example Anki Decks
 
 Check out these example Anki decks generated using QuizGen:

src/quizgen/scripts/generate_embeddings_batch.py

@@ -7,7 +7,7 @@
 from tqdm import tqdm
 
 
-def generate_embeddings(input_dir, output_dir):
+def generate_embeddings(input_dir, output_dir, embedding_model):
     input_dir = pathlib.Path(input_dir).resolve()
     output_dir = pathlib.Path(output_dir).resolve()
     json_files = list(input_dir.rglob("*.json"))
@@ -34,6 +34,8 @@ def generate_embeddings(input_dir, output_dir):
                 str(json_file),
                 "--output-path",
                 str(output_path),
+                "--embedding-model",
+                embedding_model,
             ],
             check=True,
         )
@@ -54,9 +56,15 @@ def main():
     parser.add_argument(
         "--output-dir", required=True, help="Output directory for embedding files"
     )
+    parser.add_argument(
+        "--embedding-model",
+        type=str,
+        default="text-embedding-3-small",
+        help="Model to use for generating embeddings",
+    )
     args = parser.parse_args()
 
-    generate_embeddings(args.input_dir, args.output_dir)
+    generate_embeddings(args.input_dir, args.output_dir, args.embedding_model)
 
 
 if __name__ == "__main__":