Spaces:
Sleeping
Sleeping
File size: 4,014 Bytes
3a66575 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 | import json
import math
import os
import time
import torch
PRESETS = {
"Fast": {
"magnification": 1.0,
"confidence_threshold": 0.55,
"nms_radius": 10.0,
"tracker_max_distance": 65.0,
"inference_batch_size": 16,
"frame_skip": 5,
"patch_overlap": 0.0,
},
"Balanced": {
"magnification": 1.5,
"confidence_threshold": 0.5,
"nms_radius": 8.0,
"tracker_max_distance": 50.0,
"inference_batch_size": 8,
"frame_skip": 2,
"patch_overlap": 0.25,
},
"Accurate": {
"magnification": 2.0,
"confidence_threshold": 0.45,
"nms_radius": 6.0,
"tracker_max_distance": 40.0,
"inference_batch_size": 4,
"frame_skip": 1,
"patch_overlap": 0.5,
},
}
def load_config(config_path, defaults):
if os.path.exists(config_path):
with open(config_path, "r", encoding="utf-8") as f:
return {**defaults, **json.load(f)}
return defaults
def save_config(config_path, config):
with open(config_path, "w", encoding="utf-8") as f:
json.dump(config, f, indent=2)
def run_with_oom_recovery(process_fn, *args, batch_size=8, min_batch_size=1, **kwargs):
current_batch = max(min_batch_size, int(batch_size))
while current_batch >= min_batch_size:
try:
return process_fn(*args, batch_size=current_batch, **kwargs), current_batch
except RuntimeError as exc:
if "out of memory" not in str(exc).lower():
raise
if torch.cuda.is_available():
torch.cuda.empty_cache()
if current_batch == min_batch_size:
raise
current_batch = max(min_batch_size, current_batch // 2)
raise RuntimeError("CUDA OOM recovery failed.")
def confidence_interval(values, z=1.96):
if not values:
return 0.0, 0.0
if len(values) == 1:
return float(values[0]), float(values[0])
mean = sum(values) / len(values)
variance = sum((x - mean) ** 2 for x in values) / (len(values) - 1)
margin = z * math.sqrt(variance) / math.sqrt(len(values))
return mean - margin, mean + margin
def result_summary(start_time, frames_read, frames_analyzed, counts):
elapsed = max(time.perf_counter() - start_time, 1e-9)
ci_low, ci_high = confidence_interval(counts)
return {
"elapsed_sec": round(elapsed, 3),
"frames_read": frames_read,
"frames_analyzed": frames_analyzed,
"effective_fps": round(frames_read / elapsed, 3),
"analysis_fps": round(frames_analyzed / elapsed, 3),
"avg_count": round(sum(counts) / len(counts), 3) if counts else 0,
"peak_count": max(counts) if counts else 0,
"count_95ci_low": round(ci_low, 3),
"count_95ci_high": round(ci_high, 3),
}
APP_PASTE_SECTION = r'''
CONFIG_PATH = os.path.join(BASE_DIR, "civic_pulse_config.json")
PRESETS = {
"Fast": {"magnification": 1.0, "confidence_threshold": 0.55, "nms_radius": 10.0, "tracker_max_distance": 65.0, "inference_batch_size": 16, "frame_skip": 5, "patch_overlap": 0.0},
"Balanced": {"magnification": 1.5, "confidence_threshold": 0.5, "nms_radius": 8.0, "tracker_max_distance": 50.0, "inference_batch_size": 8, "frame_skip": 2, "patch_overlap": 0.25},
"Accurate": {"magnification": 2.0, "confidence_threshold": 0.45, "nms_radius": 6.0, "tracker_max_distance": 40.0, "inference_batch_size": 4, "frame_skip": 1, "patch_overlap": 0.5},
}
def process_with_oom_recovery(*args, batch_size, **kwargs):
while batch_size >= 1:
try:
return process_frame(*args, batch_size=batch_size, **kwargs), batch_size
except RuntimeError as exc:
if "out of memory" not in str(exc).lower():
raise
if torch.cuda.is_available():
torch.cuda.empty_cache()
if batch_size == 1:
raise
batch_size = max(1, batch_size // 2)
'''
|