modified: Figure_1.png

modified:   checkpoints/forward/best_tcn_model.pt
	modified:   checkpoints/forward/training_history.csv
	new file:   checkpoints_mlp/task1_feature_mlp/best_feature_mlp.pt
	new file:   checkpoints_mlp/task1_feature_mlp/training_history.csv
	new file:   checkpoints_rms/forward_rms/best_rms_model.pt
	new file:   checkpoints_rms/forward_rms/training_history.csv
	modified:   evaluation_outputs/forward/evaluation_forward_test_b0_s0.png
	new file:   evaluation_outputs/forward/evaluation_forward_val_b0_s0.png
	new file:   evaluation_outputs/forward/evaluation_forward_val_b3_s0.png
	new file:   evaluation_outputs/forward_rms/evaluation_test_all_samples.csv
	new file:   evaluation_outputs/forward_rms/evaluation_test_s0.png
	new file:   evaluation_outputs/forward_rms/evaluation_train_all_samples.csv
	new file:   evaluation_outputs/forward_rms/evaluation_val_all_samples.csv
	new file:   evaluation_outputs/forward_rms/evaluation_val_s0.png
	new file:   evaluation_outputs/forward_rms/evaluation_val_s0_waveform.png
	new file:   evaluation_outputs/task1_feature_mlp/evaluation_train_all_samples.csv
	new file:   evaluation_outputs/task1_feature_mlp/evaluation_train_curve.png
	new file:   evaluation_outputs/task1_feature_mlp/evaluation_val_all_samples.csv
	new file:   evaluation_outputs/task1_feature_mlp/evaluation_val_curve.png
	new file:   evaluation_outputs/task1_feature_mlp/evaluation_val_s0.png
	new file:   evaluation_outputs/task1_feature_mlp/harmonic_5mm_0.75Hz_prediction.png
	new file:   evaluation_outputs/task1_feature_mlp/harmonic_5mm_1.55Hz_prediction.png
	new file:   scripts/__pycache__/config.cpython-310.pyc
	new file:   scripts/__pycache__/dataset.cpython-310.pyc
	new file:   scripts/__pycache__/model.cpython-310.pyc
	new file:   scripts/config.py
	new file:   scripts/dataset.py
	new file:   scripts/evaluate.py
	new file:   scripts/model.py
	new file:   scripts/predict_single.py
	new file:   scripts/train.py
	modified:   src/__pycache__/config.cpython-310.pyc
	modified:   src/__pycache__/dataset.cpython-310.pyc
	modified:   src/__pycache__/model.cpython-310.pyc
	modified:   src/config.py
	modified:   src/dataset.py
	modified:   src/model.py
	new file:   src_new/__pycache__/config.cpython-310.pyc
	new file:   src_new/__pycache__/dataset.cpython-310.pyc
	new file:   src_new/__pycache__/evaluate.cpython-310.pyc
	new file:   src_new/__pycache__/model.cpython-310.pyc
	new file:   src_new/__pycache__/train.cpython-310.pyc
	new file:   src_new/config.py
	new file:   src_new/dataset.py
	new file:   src_new/evaluate.py
	new file:   src_new/model.py
	new file:   src_new/train.py
This commit is contained in:
2026-05-06 12:19:55 +08:00
parent dcc023cc04
commit 484643409d
48 changed files with 2706 additions and 18 deletions

224
scripts/evaluate.py Normal file
View File

@@ -0,0 +1,224 @@
from __future__ import annotations
import argparse
from pathlib import Path
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import torch
try:
from .config import CORE_FEATURE_NAMES, ExperimentConfig, make_experiment_config
from .dataset import build_dataloaders, normalization_from_payload, report_to_text
from .model import build_model
except ImportError:
from config import CORE_FEATURE_NAMES, ExperimentConfig, make_experiment_config
from dataset import build_dataloaders, normalization_from_payload, report_to_text
from model import build_model
def parse_args() -> argparse.Namespace:
parser = argparse.ArgumentParser(description="Evaluate feature MLP on train/val/test splits.")
parser.add_argument("--split", choices=("train", "val", "test"), default="val")
parser.add_argument("--sample-index", type=int, default=0)
parser.add_argument("--all-samples", action="store_true")
parser.add_argument("--device", type=str, default=None)
parser.add_argument("--checkpoint", type=str, default=None)
return parser.parse_args()
def resolve_device(device_name: str | None, config: ExperimentConfig) -> torch.device:
requested = device_name or config.train.device
if requested.startswith("cuda") and not torch.cuda.is_available():
return torch.device("cpu")
return torch.device(requested)
def resolve_checkpoint_path(config: ExperimentConfig, checkpoint_arg: str | None) -> Path:
if checkpoint_arg:
return Path(checkpoint_arg).resolve()
return config.data.project_root / config.train.checkpoint_dir / "task1_feature_mlp" / config.train.best_model_name
def relative_percent_error(true_value: float, pred_value: float) -> float:
denominator = max(abs(true_value), 1e-12)
return abs(pred_value - true_value) / denominator * 100.0
def predict_rms(
model: torch.nn.Module,
feature_norm: torch.Tensor,
x_rms_raw: torch.Tensor,
target_mean: torch.Tensor,
target_std: torch.Tensor,
) -> torch.Tensor:
pred_norm = model(feature_norm)
pred_tr = torch.clamp(pred_norm * target_std + target_mean, min=1e-6)
return pred_tr * x_rms_raw
def evaluate_sample(
model: torch.nn.Module,
dataset,
raw_record,
sample_index: int,
device: torch.device,
) -> dict[str, float | str]:
sample = dataset[sample_index]
normalization = dataset.normalization
feature_norm = sample["x"].unsqueeze(0).to(device)
x_rms_raw = sample["x_rms_raw"].unsqueeze(0).to(device)
target_mean = normalization.target_mean.to(device).view(1, 1)
target_std = normalization.target_std.to(device).view(1, 1)
with torch.no_grad():
pred_rms = predict_rms(model, feature_norm, x_rms_raw, target_mean, target_std)
true_rms = float(sample["y_rms_raw"].item())
pred_rms_value = float(pred_rms.detach().cpu().numpy().reshape(-1)[0])
row = {
"file_name": str(sample["file_name"]),
"frequency_hz": float(sample["frequency_hz"].item()),
"true_rms": true_rms,
"pred_rms": pred_rms_value,
"relative_error_percent": relative_percent_error(true_rms, pred_rms_value),
}
features_raw = sample["features_raw"].numpy().reshape(-1)
for name, value in zip(CORE_FEATURE_NAMES, features_raw):
row[name] = float(value)
row["sampling_rate"] = float(raw_record.sampling_rate)
return row
def save_all_samples_plot(result_df: pd.DataFrame, split: str, save_dir: Path) -> Path | None:
if result_df["frequency_hz"].isna().any():
return None
plot_df = result_df.sort_values("frequency_hz").reset_index(drop=True)
fig, axes = plt.subplots(2, 1, figsize=(12, 8))
fig.suptitle(f"Task1 Feature-MLP Evaluation | {split}")
axes[0].plot(plot_df["frequency_hz"], plot_df["true_rms"], marker="o", label="True RMS")
axes[0].plot(plot_df["frequency_hz"], plot_df["pred_rms"], marker="o", label="Pred RMS")
axes[0].set_xlabel("Frequency (Hz)")
axes[0].set_ylabel("RMS")
axes[0].grid(True, alpha=0.3)
axes[0].legend()
axes[1].bar(plot_df["frequency_hz"].astype(str), plot_df["relative_error_percent"], color="tab:orange")
axes[1].set_xlabel("Frequency (Hz)")
axes[1].set_ylabel("Relative Error (%)")
axes[1].grid(True, axis="y", alpha=0.3)
axes[1].tick_params(axis="x", labelrotation=45)
plt.tight_layout()
save_dir.mkdir(parents=True, exist_ok=True)
figure_path = save_dir / f"evaluation_{split}_curve.png"
plt.savefig(figure_path, dpi=180, bbox_inches="tight")
plt.close(fig)
return figure_path
def save_single_sample_plot(raw_record, result: dict[str, float | str], split: str, save_dir: Path, sample_index: int) -> Path:
time_middle = raw_record.time_middle.detach().cpu().numpy().reshape(-1)
x_middle = raw_record.x_middle.detach().cpu().numpy().reshape(-1)
y_middle = raw_record.y_middle.detach().cpu().numpy().reshape(-1)
frequency_hz = float(result["frequency_hz"])
pred_rms = float(result["pred_rms"])
true_rms = float(result["true_rms"])
fig, axes = plt.subplots(3, 1, figsize=(12, 10))
fig.suptitle(f"Task1 Feature-MLP | {split} | {result['file_name']}")
axes[0].plot(time_middle, x_middle, color="tab:blue")
axes[0].set_title("Input Base Excitation (Middle Segment)")
axes[0].set_xlabel("Time")
axes[0].set_ylabel("Acceleration")
axes[0].grid(True, alpha=0.3)
axes[1].plot(time_middle, y_middle, color="tab:green")
axes[1].set_title("True Top Response (Middle Segment)")
axes[1].set_xlabel("Time")
axes[1].set_ylabel("Acceleration")
axes[1].grid(True, alpha=0.3)
axes[2].bar(["True RMS", "Pred RMS"], [true_rms, pred_rms], color=["tab:green", "tab:orange"])
axes[2].set_title(
f"Freq: {frequency_hz:.4f} Hz | True RMS: {true_rms:.6f} | Pred RMS: {pred_rms:.6f} | "
f"Error: {float(result['relative_error_percent']):.2f}%"
)
axes[2].set_ylabel("RMS")
axes[2].grid(True, axis="y", alpha=0.3)
plt.tight_layout()
save_dir.mkdir(parents=True, exist_ok=True)
figure_path = save_dir / f"evaluation_{split}_s{sample_index}.png"
plt.savefig(figure_path, dpi=180, bbox_inches="tight")
plt.close(fig)
return figure_path
def main() -> None:
args = parse_args()
config = make_experiment_config()
device = resolve_device(args.device, config)
checkpoint_path = resolve_checkpoint_path(config, args.checkpoint)
if not checkpoint_path.exists():
raise FileNotFoundError(f"Checkpoint not found: {checkpoint_path}")
checkpoint = torch.load(checkpoint_path, map_location=device, weights_only=False)
if "normalization" in checkpoint:
normalization = normalization_from_payload(checkpoint["normalization"])
else:
normalization = None
loaders, datasets, raw_records, reports = build_dataloaders(config)
del loaders
print(report_to_text(reports))
dataset = datasets[args.split]
if normalization is not None:
dataset.normalization = normalization
model = build_model(config).to(device)
model.load_state_dict(checkpoint["model_state_dict"])
model.eval()
save_dir = config.data.project_root / "evaluation_outputs" / "task1_feature_mlp"
if args.all_samples:
rows = [
evaluate_sample(model, dataset, raw_records[args.split][index], index, device)
for index in range(len(dataset))
]
result_df = pd.DataFrame(rows).sort_values(["relative_error_percent", "file_name"]).reset_index(drop=True)
save_dir.mkdir(parents=True, exist_ok=True)
csv_path = save_dir / f"evaluation_{args.split}_all_samples.csv"
result_df.to_csv(csv_path, index=False)
figure_path = save_all_samples_plot(result_df, args.split, save_dir)
summary = {
"count": len(result_df),
"mean_error_percent": float(result_df["relative_error_percent"].mean()),
"median_error_percent": float(result_df["relative_error_percent"].median()),
"max_error_percent": float(result_df["relative_error_percent"].max()),
"min_error_percent": float(result_df["relative_error_percent"].min()),
}
print(f"Checkpoint: {checkpoint_path}")
print(f"Summary: {summary}")
print(f"CSV saved to: {csv_path}")
if figure_path is not None:
print(f"Figure saved to: {figure_path}")
print(result_df.to_string(index=False))
return
result = evaluate_sample(model, dataset, raw_records[args.split][args.sample_index], args.sample_index, device)
figure_path = save_single_sample_plot(raw_records[args.split][args.sample_index], result, args.split, save_dir, args.sample_index)
print(f"Checkpoint: {checkpoint_path}")
print(f"Sample file: {result['file_name']}")
print(f"True RMS: {float(result['true_rms']):.6f}")
print(f"Pred RMS: {float(result['pred_rms']):.6f}")
print(f"Relative RMS Error (%): {float(result['relative_error_percent']):.4f}")
print(f"Figure saved to: {figure_path}")
if __name__ == "__main__":
main()