File size: 13,290 Bytes
e4194f4 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 | import os
import json
import tqdm
import torch
import numpy as np
import click
from datetime import datetime
import lightning.pytorch as pl
import sklearn.metrics as skm
from torch.utils.data import DataLoader
from torchvision.transforms import transforms as T
from torchvision.transforms._transforms_video import ToTensorVideo
from pytorchvideo.transforms import Normalize
# Импорты из соседних папок (относительные пути)
from full_model.rnn_dataset import SyntaxDataset
from full_model.rnn_model import SyntaxLightningModule
from metrics_visualization import visualize_final_syntax_plotly_multi
DEVICE = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(f"DEVICE: {DEVICE}")
def safe_sample_std(values):
"""Sample std (ddof=1). Если значение одно/пусто — 0.0."""
arr = np.array(values, dtype=float)
if arr.size <= 1:
return 0.0
return float(arr.std(ddof=1))
def compute_metrics(y_true, y_pred, thr=22.0):
"""R2, MAE, Pearson, MAPE, Mean_Recall."""
y_true_arr = np.array(y_true, dtype=float)
y_pred_arr = np.array(y_pred, dtype=float)
r2 = float(skm.r2_score(y_true_arr, y_pred_arr))
mae = float(skm.mean_absolute_error(y_true_arr, y_pred_arr))
pearson = float(np.corrcoef(y_true_arr, y_pred_arr)[0, 1]) if len(y_true_arr) > 1 else 0.0
mape = float(skm.mean_absolute_percentage_error(y_true_arr, y_pred_arr))
y_true_bin = (y_true_arr >= thr).astype(int)
y_pred_bin = (y_pred_arr >= thr).astype(int)
unique_classes = np.unique(np.concatenate([y_true_bin, y_pred_bin]))
mean_recall = float(np.mean(skm.recall_score(y_true_bin, y_pred_bin, average=None, labels=[0, 1]))) \
if len(unique_classes) > 1 else 0.0
return r2, mae, pearson, mape, mean_recall
@click.command()
@click.option("-d", "--dataset-paths", multiple=True,
help="JSON с метаданными датасетов (относительно dataset_root).")
@click.option("-n", "--dataset-names", multiple=True,
help="Имена датасетов для метрик/графиков.")
@click.option("-p", "--postfixes", multiple=True,
help="Суффиксы для файлов предсказаний.")
@click.option("-r", "--dataset-root", type=click.Path(exists=True),
help="Корень датасета (где лежат JSON и DICOM).")
@click.option("-v", "--video-size", type=click.Tuple([int, int]),
help="Размер видео (H, W).")
@click.option("--frames-per-clip",
help="Количество кадров в клипе.")
@click.option("--num-workers",
help="Число DataLoader workers.")
@click.option("--seed",
help="Random seed.")
@click.option("--pt-weights-format", is_flag=True,
help="True → модели в .pt (torch.save), False → .ckpt (Lightning).")
@click.option("--use-scaling", is_flag=True,
help="Применить a*x+b scaling из JSON.")
@click.option("--scaling-file",
help="JSON с коэффициентами scaling (относительно dataset_root).")
@click.option("-e", "--ensemble-name",
help="Имя ансамбля в metrics.json.")
@click.option("-m", "--metrics-file",
help="JSON с метриками экспериментов.")
def main(dataset_paths, dataset_names, postfixes, dataset_root, video_size,
frames_per_clip, num_workers, seed, pt_weights_format, use_scaling,
scaling_file, ensemble_name, metrics_file):
pl.seed_everything(seed)
postfix_plotly = "Ensemble"
# Пути к моделям (backbone + RNN-head)
model_paths = {
"left": [
"full_model/checkpoints/leftBinSyntax_R3D_fold00_lstm_mean_post_best.pt",
"full_model/checkpoints/leftBinSyntax_R3D_fold01_lstm_mean_post_best.pt",
"full_model/checkpoints/leftBinSyntax_R3D_fold02_lstm_mean_post_best.pt",
"full_model/checkpoints/leftBinSyntax_R3D_fold03_lstm_mean_post_best.pt",
"full_model/checkpoints/leftBinSyntax_R3D_fold04_lstm_mean_post_best.pt",
],
"right": [
"full_model/checkpoints/rightBinSyntax_R3D_fold00_lstm_mean_post_best.pt",
"full_model/checkpoints/rightBinSyntax_R3D_fold01_lstm_mean_post_best.pt",
"full_model/checkpoints/rightBinSyntax_R3D_fold02_lstm_mean_post_best.pt",
"full_model/checkpoints/rightBinSyntax_R3D_fold03_lstm_mean_post_best.pt",
"full_model/checkpoints/rightBinSyntax_R3D_fold04_lstm_mean_post_best.pt",
]
}
# Scaling параметры
scaling_params_dict = {}
if use_scaling:
postfix_plotly += "_scaled"
ensemble_name += "_scaled"
scaling_path = os.path.join(dataset_root, scaling_file)
if os.path.exists(scaling_path):
with open(scaling_path, "r") as f:
scaling_params_dict = json.load(f)
print(f"Loaded scaling from {scaling_path}")
else:
print(f"⚠️ Scaling file not found: {scaling_path}")
# Результаты ансамбля
ensemble_results = {
"timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
"use_scaling": use_scaling,
"pt_weights_format": pt_weights_format,
"datasets": {}
}
all_datasets, all_r2, all_recalls = {}, {}, {}
for dataset_path, dataset_name, postfix in zip(dataset_paths, dataset_names, postfixes):
# Относительные пути
abs_dataset_path = os.path.join(dataset_root, dataset_path)
results_file = os.path.join(dataset_root, "coeffs", f"{postfix}.json")
# Загрузка/вычисление предсказаний
if os.path.exists(results_file):
print(f"[{postfix}] Loading from {results_file}")
with open(results_file, "r") as f:
data = json.load(f)
syntax_true = data["syntax_true"]
left_preds_all = data["left_preds"]
right_preds_all = data["right_preds"]
else:
print(f"[{postfix}] Computing predictions...")
left_preds_all, left_sids = run_artery(
abs_dataset_path, "left", model_paths["left"],
video_size, frames_per_clip, num_workers, pt_weights_format
)
right_preds_all, right_sids = run_artery(
abs_dataset_path, "right", model_paths["right"],
video_size, frames_per_clip, num_workers, pt_weights_format
)
assert left_sids == right_sids
with open(abs_dataset_path, "r") as f:
dataset = json.load(f)
syntax_true = [rec.get("mean_syntax", rec.get("syntax")) for rec in dataset]
os.makedirs(os.path.dirname(results_file), exist_ok=True)
save_data = {
"syntax_true": syntax_true,
"left_preds": left_preds_all,
"right_preds": right_preds_all
}
with open(results_file, "w") as f:
json.dump(save_data, f)
print(f"[{postfix}] Saved to {results_file}")
# Scaling (fold-wise для left/right)
if use_scaling:
left_scaled_all, right_scaled_all = [], []
for pred_list in left_preds_all:
scaled = [scaling_params_dict.get(f"fold{i}", (1.0, 0.0))[0] * val +
scaling_params_dict.get(f"fold{i}", (1.0, 0.0))[1]
for i, val in enumerate(pred_list)]
left_scaled_all.append(scaled)
for pred_list in right_preds_all:
scaled = [scaling_params_dict.get(f"fold{i}", (1.0, 0.0))[0] * val +
scaling_params_dict.get(f"fold{i}", (1.0, 0.0))[1]
for i, val in enumerate(pred_list)]
right_scaled_all.append(scaled)
else:
left_scaled_all, right_scaled_all = left_preds_all, right_preds_all
# Ансамбль: mean по фолдам + left+right
syntax_pred = [max(0.0, float(np.mean([l + r for l, r in zip(l_list, r_list)])))
for l_list, r_list in zip(left_scaled_all, right_scaled_all)]
# Метрики ансамбля
r2, mae, pearson, mape, mean_recall = compute_metrics(syntax_true, syntax_pred)
print(f"[{postfix}] ENSEMBLE: R2={r2:.4f}, Pearson={pearson:.4f}, "
f"MAE={mae:.4f}, MAPE={mape:.4f}, Recall={mean_recall:.4f}")
# STD по фолдам
n_folds = len(left_scaled_all[0]) if left_scaled_all else 0
fold_metrics = {metric: [] for metric in ["R2", "MAE", "Pearson", "MAPE", "Mean_Recall"]}
for k in range(n_folds):
pred_k = [max(0.0, l_list[k] + r_list[k])
for l_list, r_list in zip(left_scaled_all, right_scaled_all)]
fold_r2, fold_mae, fold_pearson, fold_mape, fold_recall = compute_metrics(syntax_true, pred_k)
for metric, value in zip(fold_metrics.keys(),
[fold_r2, fold_mae, fold_pearson, fold_mape, fold_recall]):
fold_metrics[metric].append(value)
fold_summary = {k: {"mean": float(np.mean(v)), "std": safe_sample_std(v), "values": v}
for k, v in fold_metrics.items()}
# Визуализация и сохранение
all_datasets[dataset_name] = (syntax_true, syntax_pred)
all_r2[dataset_name] = r2
all_recalls[dataset_name] = mean_recall
ensemble_results["datasets"][dataset_name] = {
# Ансамбль
"R2": round(r2, 4), "MAE": round(mae, 4),
"Pearson": round(pearson, 4), "MAPE": round(mape, 4),
"Mean_Recall": round(mean_recall, 4), "N_samples": len(syntax_true),
# По фолдам (mean±std)
**{f"{k}_mean": round(v["mean"], 4) for k, v in fold_summary.items()},
**{f"{k}_std": round(v["std"], 4) for k, v in fold_summary.items()},
**{f"{k}_folds": [round(x, 4) for x in v["values"]] for k, v in fold_summary.items()}
}
# Сохранение метрик
metrics_path = os.path.join(dataset_root, metrics_file)
full_history = {}
if os.path.exists(metrics_path):
try:
with open(metrics_path, "r") as f:
full_history = json.load(f)
except json.JSONDecodeError:
print("⚠️ Metrics file corrupted. Creating new.")
full_history[ensemble_name] = ensemble_results
with open(metrics_path, "w") as f:
json.dump(full_history, f, indent=4)
print(f"✅ Metrics saved: {metrics_path}")
# Визуализация
visualize_final_syntax_plotly_multi(
datasets=all_datasets, r2_values=all_r2, recall_values=all_recalls,
gt_row="ENSEMBLE", postfix=postfix_plotly
)
def run_artery(dataset_path, artery, model_paths, video_size, frames_per_clip,
num_workers, pt_weights_format=False):
"""Инференс для одной артерии (5 фолдов)."""
imagenet_mean = [0.485, 0.456, 0.406]
imagenet_std = [0.229, 0.224, 0.225]
test_transform = T.Compose([
ToTensorVideo(),
T.Resize(size=video_size, antialias=True),
Normalize(mean=imagenet_mean, std=imagenet_std),
])
val_set = SyntaxDataset(
root=os.path.dirname(dataset_path),
meta=dataset_path,
train=False,
length=frames_per_clip,
label="", # inference mode
artery=artery,
inference=True,
transform=test_transform
)
val_loader = DataLoader(val_set, batch_size=1, num_workers=num_workers,
shuffle=False, pin_memory=True)
print(f"{artery} artery: {len(val_loader)} samples")
models = []
for path in model_paths:
if not os.path.exists(path):
print(f"⚠️ Model not found: {path}")
continue
model = SyntaxLightningModule(
num_classes=2, lr=1e-5, variant="lstm_mean",
weight_decay=0.001, max_epochs=1,
pl_weight_path=path, pt_weights_format=pt_weights_format
)
model.to(DEVICE)
model.eval()
models.append(model)
if not models:
raise RuntimeError(f"No models loaded for {artery}")
preds_all, sids = [], []
with torch.no_grad():
for x, [y], [t], [sid] in tqdm.tqdm(val_loader, desc=f"{artery} infer"):
if len(x.shape) == 1: # пустое видео
val_syntax_list = [0.0] * len(models)
else:
x = x.to(DEVICE)
val_syntax_list = []
for model in models:
pred = model(x)
_, val_log = pred # регрессионный logit
val = float(torch.exp(val_log).cpu()) - 1
val_syntax_list.append(val)
preds_all.append(val_syntax_list)
sids.append(sid[0]) # study_uid
return preds_all, sids
if __name__ == "__main__":
main()
|