Spaces:
Running
Running
| import numpy as np | |
| import os | |
| base = r"c:\Users\ASUS\lung_ai_project\data" | |
| dir_orig = os.path.join(base, "hear_embeddings") | |
| dir_aug = os.path.join(base, "hear_embeddings_augmented") | |
| x1 = np.load(os.path.join(dir_orig, "X_hear.npy")) | |
| y1 = np.load(os.path.join(dir_orig, "y_hear.npy")) | |
| x2 = np.load(os.path.join(dir_aug, "X_hear_aug.npy")) | |
| y2 = np.load(os.path.join(dir_aug, "y_hear_aug.npy")) | |
| # Detailed check | |
| print(f"Original: {x1.shape}") | |
| print(f"Augmented: {x2.shape}") | |
| # Check first sick sample in Orig | |
| sick_indices_1 = np.where(y1 == 'sick')[0] | |
| sick_sample_1 = x1[sick_indices_1[0]] | |
| # Check if this sample exists in x2 | |
| matches = np.all(np.isclose(x2, sick_sample_1, atol=1e-5), axis=1) | |
| if np.any(matches): | |
| print("Found exact match of original sick sample in augmented data.") | |
| print(f"Count of matches: {np.sum(matches)}") | |
| else: | |
| print("Original sick sample NOT found in augmented data (implies transformation or different subset).") | |
| # Check if x2 contains duplicates within itself | |
| _, counts = np.unique(x2, axis=0, return_counts=True) | |
| if np.any(counts > 1): | |
| print("Augmented data contains exact duplicates!") | |
| else: | |
| print("Augmented data has unique samples.") | |