Created
October 14, 2025 01:52
-
-
Save zaemyung/64ec963b6c664369da4846550b2cb464 to your computer and use it in GitHub Desktop.
Computing mf-idf
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def compute_mfidf_distribution(machines, humans, motif_size): | |
| # document here is either "machine" or "human" | |
| machine_mf = np.array([sample[f'motif_m{motif_size}_hist_norm_by_edges'] for sample in machines]) | |
| machine_mf = np.mean(machine_mf, axis=0) | |
| human_mf = np.array([sample[f'motif_m{motif_size}_hist_norm_by_edges'] for sample in humans]) | |
| human_mf = np.mean(human_mf, axis=0) | |
| machine_df = np.array([sample[f'motif_m{motif_size}_hist_raw'] for sample in machines]) | |
| machine_df = (machine_df > 0).astype(int) | |
| machine_df = np.mean(machine_df, axis=0) | |
| human_df = np.array([sample[f'motif_m{motif_size}_hist_raw'] for sample in humans]) | |
| human_df = (human_df > 0).astype(int) | |
| human_df = np.mean(human_df, axis=0) | |
| df = machine_df + human_df | |
| idf = np.log((2 + 1) / (df + 1)) | |
| machine_mfidf = np.multiply(machine_mf, idf) | |
| human_mfidf = np.multiply(human_mf, idf) | |
| machine_top_motif_indices = machine_mfidf.argsort()[::-1] | |
| human_top_motif_indices = human_mfidf.argsort()[::-1] | |
| return {'machine_top_motif_indices': machine_top_motif_indices, 'human_top_motif_indices': human_top_motif_indices, | |
| 'machine_mfidf': machine_mfidf, 'human_mfidf': human_mfidf, 'machine_mfidf-human_mfidf': machine_mfidf - human_mfidf} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment