Delete combine_raters.py

combine_raters.py

@@ -1,91 +0,0 @@
-import pandas as pd
-import json
-import re
-
-# -----------------------------------------------------
-# CONFIG: Rater files mapping
-# -----------------------------------------------------
-RATER_FILES = {
-    "Vidhi": "data/Vidhi_50.csv",
-    "Grant": "data/Grant_50.csv",
-    "Amanuel": "data/Amanuel_50.csv"
-}
-
-OUTPUT_CSV = "combined_rater_with_embed.csv"
-
-
-def extract_video_id(filename):
-    """Extract numeric ID from filename '0111.mp4' → '0111'."""
-    m = re.match(r"(\d+)", filename)
-    return m.group(1) if m else None
-
-
-def parse_video_labels(field):
-    """Parse JSON-like videoLabels field."""
-    try:
-        items = json.loads(field)
-    except:
-        return []
-
-    out = []
-    for entry in items:
-        labels = entry.get("timelinelabels") or entry.get("timelineLabels") or []
-        ranges = entry.get("ranges") or []
-
-        if not labels or not ranges:
-            continue
-
-        label = labels[0]
-        for r in ranges:
-            out.append({
-                "label": label,
-                "start": r.get("start"),
-                "end": r.get("end")
-            })
-    return out
-
-
-records = []
-
-for rater_name, csv_path in RATER_FILES.items():
-    df = pd.read_csv(csv_path)
-
-    for _, row in df.iterrows():
-        filename = row["filename"]
-        vid_id = extract_video_id(filename)
-
-        # Local video file path
-        video_path = f"videos/{filename}"
-
-        labels = parse_video_labels(row["videoLabels"])
-
-        if not labels:
-            records.append({
-                "filename": filename,
-                "video_id": vid_id,
-                "video_path": video_path,
-                "label": None,
-                "rater": rater_name,
-                "start": None,
-                "end": None
-            })
-        else:
-            for lab in labels:
-                records.append({
-                    "filename": filename,
-                    "video_id": vid_id,
-                    "video_path": video_path,
-                    "label": lab["label"],
-                    "rater": rater_name,
-                    "start": lab["start"],
-                    "end": lab["end"]
-                })
-
-# Build DataFrame with explicit column order so the web app can rely on it
-columns = ["filename", "video_id", "video_path", "label", "rater", "start", "end"]
-combined = pd.DataFrame(records)
-# Ensure columns exist in the desired order (missing keys will produce NaN)
-combined = combined.reindex(columns=columns)
-combined.to_csv(OUTPUT_CSV, index=False)
-
-print(f"✅ CSV created → {OUTPUT_CSV} (columns: {', '.join(columns)})")