fix.py

import json
from collections import defaultdict
import numpy as np

TRAIN_JSON_PATH = "./data/miko/train.json"
ACTION_STATS_PATH = "./data/action_statistics.json"

# Constants to prevent instability
MIN_TOTAL_LEN = 1.0     # Floor duration to prevent infinite weights
MAX_WEIGHT = 0.05       # Cap max weight to avoid domination

# Load train.json
with open(TRAIN_JSON_PATH, "r") as f:
    dataset = json.load(f)

# Load existing stats if available
try:
    with open(ACTION_STATS_PATH, "r") as f:
        existing_stats = json.load(f)
except FileNotFoundError:
    existing_stats = {}

# Initialize merged stats
merged_stats = defaultdict(lambda: {"total_len": 0.0, "total_weight": 1})

# Copy existing values
for act, stats in existing_stats.items():
    merged_stats[act]["total_len"] = stats.get("total_len", 0.0)
    merged_stats[act]["total_weight"] = 1

# Aggregate durations from dataset
for entry in dataset:
    labels = entry.get("frame_ann", {}).get("labels", [])
    for label in labels:
        start = label.get("start_t")
        end = label.get("end_t")
        act_cats = label.get("act_cat", [])
        if start is None or end is None:
            continue
        duration = max(0.01, end - start)
        for act in act_cats:
            merged_stats[act]["total_len"] += duration

# Recompute weights
cleaned_stats = {}
for act, stats in merged_stats.items():
    total_len = max(MIN_TOTAL_LEN, stats["total_len"])  # Floor to avoid tiny durations
    if np.isfinite(total_len):
        raw_weight = 1.0 / total_len
        weight = min(raw_weight, MAX_WEIGHT)  # Cap max weight
        cleaned_stats[act] = {
            "total_len": total_len,
            "total_weight": 1,
            "weight": weight
        }
    else:
        print(f"⚠️ Skipping act_cat '{act}' due to invalid total_len={stats['total_len']}")

# Normalize weights
total_weight_sum = sum(stats["weight"] for stats in cleaned_stats.values())
if total_weight_sum == 0:
    raise ValueError("❌ All weights are zero. Check act_cat labels or durations.")

for stats in cleaned_stats.values():
    stats["weight"] /= total_weight_sum

# Save updated statistics
with open(ACTION_STATS_PATH, "w") as f:
    json.dump(cleaned_stats, f, indent=2)

print(f"✅ Saved {len(cleaned_stats)} act_cat entries to {ACTION_STATS_PATH}")
print(f"🎯 Final normalized weight sum: {sum(stats['weight'] for stats in cleaned_stats.values()):.4f}")