from datasets import load_dataset import numpy as np import matplotlib.pyplot as plt # Load training split dataset = load_dataset("OCEAN-Lab/CTS-PADC_Dataset", split="train") # Extract sample 0 sample = dataset[0] x_distorted = np.array(sample["arcsin_results"]) # Length: 400 y_ground_truth = np.array(sample["test_data"]) # Length: 400 rf_freq = sample["Frequencies"] # Frequency in GHz rf_power = sample["Power"] # Power in dBm print(f"Sample 0 -> RF Frequency: {rf_freq} GHz | RF Power: {rf_power} dBm") print(f"Input shape: {x_distorted.shape} | Target shape: {y_ground_truth.shape}") # Plot time-domain waveform plt.figure(figsize=(9, 3.5)) plt.plot(y_ground_truth, label="Ground Truth", color="black", linestyle="--") plt.plot(x_distorted, label="Distorted Input", color="crimson", alpha=0.7) plt.title(f"PTS-ADC Capture: {rf_freq} GHz at {rf_power} dBm") plt.xlabel("Sample Index n") plt.ylabel("Amplitude (V)") plt.legend() plt.grid(True, linestyle=":") plt.tight_layout() plt.show()