refactoring

app.py

@@ -6,7 +6,7 @@ from pathlib import Path
 import pandas as pd
 import os
 import json
-from utils.data import parse_json_files, preprocess_traces
+from utils.data import parse_json_files
 from utils.viz import create_scatter_plot, create_flow_chart
 from utils.processing import check_and_process_uploads
 from huggingface_hub import snapshot_download
@@ -36,9 +36,27 @@ def download_latest_results():
 
 abs_path = Path(__file__).parent
 
-
 # Global variable to store preprocessed data
 preprocessed_traces = {}
+def preprocess_traces():
+    global preprocessed_traces
+    processed_dir = Path("evals_live")
+    for file in processed_dir.glob('*.json'):
+        try:
+            with open(file, 'r') as f:
+                data = json.load(f)
+                agent_name = data['config']['agent_name']
+                benchmark_name = data['config']['benchmark_name']
+                if benchmark_name not in preprocessed_traces:
+                    preprocessed_traces[benchmark_name] = {}
+
+                assert type(data['raw_logging_results']) == dict, f"Invalid format for raw_logging_results: {type(data['raw_logging_results'])}"
+                preprocessed_traces[benchmark_name][agent_name] = data['raw_logging_results']
+        except AssertionError as e:
+            preprocessed_traces[benchmark_name][agent_name] = None
+        except Exception as e:
+            print(f"Error preprocessing {file}: {e}")
+            preprocessed_traces[benchmark_name][agent_name] = None
 
 def get_analyzed_traces(agent_name, benchmark_name):
     return preprocessed_traces.get(benchmark_name, {}).get(agent_name)

utils/data.py

@@ -6,26 +6,6 @@ from utils.pareto import Agent, compute_pareto_frontier
 import plotly.graph_objects as go
 import textwrap
 
-def preprocess_traces():
-    global preprocessed_traces
-    processed_dir = "evals_live"
-    for file in processed_dir.glob('*.json'):
-        try:
-            with open(file, 'r') as f:
-                data = json.load(f)
-                agent_name = data['config']['agent_name']
-                benchmark_name = data['config']['benchmark_name']
-                if benchmark_name not in preprocessed_traces:
-                    preprocessed_traces[benchmark_name] = {}
-
-                assert type(data['raw_logging_results']) == dict, f"Invalid format for raw_logging_results: {type(data['raw_logging_results'])}"
-                preprocessed_traces[benchmark_name][agent_name] = data['raw_logging_results']
-        except AssertionError as e:
-            preprocessed_traces[benchmark_name][agent_name] = None
-        except Exception as e:
-            print(f"Error preprocessing {file}: {e}")
-            preprocessed_traces[benchmark_name][agent_name] = None
-
 def parse_json_files(folder_path, benchmark_name):
     # Convert folder path to Path object
     folder = Path(folder_path)