From cd237292ea17bc0922168f0c45189102f9443f35 Mon Sep 17 00:00:00 2001 From: you-n-g Date: Sun, 29 Jun 2025 21:22:16 +0800 Subject: [PATCH] feat: add hide_base_name option and update data folder prompts (#1004) * feat: add hide_base_name option and update data folder prompts * rename folder --- .../components/coder/data_science/pipeline/prompts.yaml | 2 +- .../coder/data_science/raw_data_loader/prompts.yaml | 2 +- rdagent/scenarios/data_science/scen/utils.py | 7 ++++--- 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/rdagent/components/coder/data_science/pipeline/prompts.yaml b/rdagent/components/coder/data_science/pipeline/prompts.yaml index 302c7dca..a29f141b 100644 --- a/rdagent/components/coder/data_science/pipeline/prompts.yaml +++ b/rdagent/components/coder/data_science/pipeline/prompts.yaml @@ -80,7 +80,7 @@ pipeline_coder: --------- Competition Information --------- {{ competition_info }} - --------- Data Folder Description (All path are relative to the data folder) --------- + --------- Data Folder Description (All path are relative to the data folder, i.e. "{% include "scenarios.data_science.share:scen.input_path" %}") --------- {{ folder_spec }} {% if latest_code %} diff --git a/rdagent/components/coder/data_science/raw_data_loader/prompts.yaml b/rdagent/components/coder/data_science/raw_data_loader/prompts.yaml index 60af0e21..54261a1c 100644 --- a/rdagent/components/coder/data_science/raw_data_loader/prompts.yaml +++ b/rdagent/components/coder/data_science/raw_data_loader/prompts.yaml @@ -312,7 +312,7 @@ data_loader_coder: --------- Code Specification --------- {{ code_spec }} - --------- Data Folder Description (All path are relative to the data folder) --------- + --------- Data Folder Description (All path are relative to the data folder, i.e. "{% include "scenarios.data_science.share:scen.input_path" %}") --------- {{ folder_spec }} {% if latest_code %} diff --git a/rdagent/scenarios/data_science/scen/utils.py b/rdagent/scenarios/data_science/scen/utils.py index 3517ae8c..12cd97c2 100644 --- a/rdagent/scenarios/data_science/scen/utils.py +++ b/rdagent/scenarios/data_science/scen/utils.py @@ -3,7 +3,7 @@ An example of the generated data folder description: ## File tree: ``` -plant-pathology-2020-fgvc7/ +./ ├── images/ │ ├── Test_0.jpg (182.7 kB) │ ├── Test_1.jpg (362.4 kB) @@ -294,7 +294,7 @@ class FileTreeGenerator: Smart file tree generator with symlink handling and intelligent truncation. """ - def __init__(self, max_lines: int = 200, priority_files: Set[str] = None): + def __init__(self, max_lines: int = 200, priority_files: Set[str] = None, hide_base_name: bool = True): """ Initialize the file tree generator. @@ -306,6 +306,7 @@ class FileTreeGenerator: self.priority_files = priority_files or {".csv", ".json", ".parquet", ".md", ".txt"} self.lines = [] self.line_count = 0 + self.hide_base_name = hide_base_name def generate_tree(self, path: Union[str, Path]) -> str: """ @@ -325,7 +326,7 @@ class FileTreeGenerator: base_path = path.resolve() self.lines = [] self.line_count = 0 - self._add_line(f"{path.name}/") + self._add_line(f"{'.' if self.hide_base_name else path.name}/") self._process_directory(path, 0, "", base_path) except MaxLinesExceededError: pass # Expected when hitting line limit