chore: add demo to add the docstring in Google format to a Python fun…

…ction

demo.md

@@ -18,20 +18,45 @@
 
 # Programming - 2
 
-In the CUDA code
-
-```shell
-template <typename T>
-__global__ void f(int* ids, int* atomic_counters, int n)
-{
-    int idx = blockDim.x * blockIdx.x + threadIdx.x;
-    if (idx >= n) return;
-    ids[idx] = idx;
-    int current_counter = atomicAdd(atomic_counters + idx, 1);
-}
+Add the docstring in Google format to the following Python function:
 ```
+    def generate_response(
+        self, retrieved_contents: List[Document], question: str, max_new_tokens: int = 512
+    ) -> Union[str, Any]:
+        cur_response = None
+        fmt_prompts = []
 
-is ensured that the atomicAdd operation is run after the execution of ids[idx] = idx?
+        if not retrieved_contents:
+            qa_prompt = self.llm.generate_qa_prompt(question=question)
+            logger.info("--- Generating a single response ... ---")
+            response = self.llm.start_answer_iterator_streamer(qa_prompt, max_new_tokens=max_new_tokens)
+            return response, qa_prompt
+
+        num_of_contents = len(retrieved_contents)
+
+        for idx, node in enumerate(retrieved_contents, start=1):
+            logger.info(f"--- Generating an answer for the chunk {idx} ... ---")
+            context = node.page_content
+            logger.debug(f"--- Context: '{context}' ... ---")
+            if idx == 0:
+                fmt_prompt = self.llm.generate_ctx_prompt(question=question, context=context)
+            else:
+                fmt_prompt = self.llm.generate_refined_ctx_prompt(
+                    context=context,
+                    question=question,
+                    existing_answer=str(cur_response),
+                )
+
+            if idx == num_of_contents:
+                cur_response = self.llm.start_answer_iterator_streamer(fmt_prompt, max_new_tokens=max_new_tokens)
+
+            else:
+                cur_response = self.llm.generate_answer(fmt_prompt, max_new_tokens=max_new_tokens)
+                logger.debug(f"--- Current response: '{cur_response}' ... ---")
+            fmt_prompts.append(fmt_prompt)
+
+        return cur_response, fmt_prompts
+```
 
 # Test if the model is uncensored - 1