Feat/add max debug retry parameter

vizro-ai/changelog.d/20240110_093842_anna_xiong_add_max_retry_parameter.md

@@ -0,0 +1,48 @@
+<!--
+A new scriv changelog fragment.
+
+Uncomment the section that is right (remove the HTML comment wrapper).
+-->
+
+<!--
+### Highlights ✨
+
+- A bullet item for the Highlights ✨ category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Removed
+
+- A bullet item for the Removed category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Added
+
+- A bullet item for the Added category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Changed
+
+- A bullet item for the Changed category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Deprecated
+
+- A bullet item for the Deprecated category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Fixed
+
+- A bullet item for the Fixed category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->
+<!--
+### Security
+
+- A bullet item for the Security category with a link to the relevant PR at the end of your entry, e.g. Enable feature XXX ([#1](https://github.com/mckinsey/vizro/pull/1))
+
+-->

vizro-ai/docs/pages/user_guides/run_vizro_ai.md

@@ -80,3 +80,19 @@ Vizro-AI's `_get_chart_code` method returns the Python code string that can be u
 
 The returned `code_string` can be used to dynamically render charts within your application. You may have the option to encapsulate the chart within a `fig` object or convert the figure into a JSON string for further integration.
 In case you would like to use the insights or code explanation, you can use `vizro_ai._run_plot_tasks(df, ..., explain=True)`, which returns a dictionary containing the code explanation and chart insights alongside the code.
+
+### 4. Increase the maximum number of attempts to debug
+Sometimes there is error in code generated first round since the code is generated by Large Language Models (LLMs),
+
+The `max_debug_retry` parameter in the plot function specifies the maximum number of attempts the function will make to debug any errors that occur during its execution.
+When an error is encountered, the function will automatically invoke a debugging process. This process involves running the code, capturing any errors that occur, and then using the LLM to analyze and attempt to resolve these errors.
+
+By default, the max_debug_retry is set to 3, the function will attempt to debug errors up to three times.
+If the errors are not resolved after the maximum number of retries, the function will cease further debugging attempts.
+
+if users would like to increase or decrease the maximum number of retries, adding parameter `max_debug_retry` in plot call as below works:
+!!! example "Change max_debug_retry"
+    === "Code for the cell"
+        ```py
+        vizro_ai.plot(df = df, user_input = "your user input", max_debug_retry= 5)
+        ```

vizro-ai/src/vizro_ai/_vizro_ai.py

@@ -103,15 +103,18 @@ def _get_chart_code(self, df: pd.DataFrame, user_input: str) -> str:
         # TODO refine and update error handling
         return self._run_plot_tasks(df, user_input, explain=False).get("code_string")
 
-    def plot(self, df: pd.DataFrame, user_input: str, explain: bool = False) -> Union[None, Dict[str, Any]]:
+    def plot(
+        self, df: pd.DataFrame, user_input: str, explain: bool = False, max_debug_retry: int = 3
+    ) -> Union[None, Dict[str, Any]]:
         """Plot visuals using vizro via english descriptions, english to chart translation.
 
         Args:
             df: The dataframe to be analyzed
             user_input: User questions or descriptions of the desired visual
             explain: Flag to include explanation in response
+            max_debug_retry: Number of times LLM can retry for debug
         """
-        output_dict = self._run_plot_tasks(df, user_input, explain=explain)
+        output_dict = self._run_plot_tasks(df, user_input, explain=explain, max_debug_retry=max_debug_retry)
         code_string = output_dict.get("code_string")
         business_insights = output_dict.get("business_insights")
         code_explanation = output_dict.get("code_explanation")