Created
July 31, 2020 09:22
-
-
Save agn-7/571e84efe34512e9380de4e900cf23b5 to your computer and use it in GitHub Desktop.
Apply PCA on a vector.
This file contains 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
| import numpy as np | |
| import pandas as pd | |
| import csv | |
| import sys | |
| from sklearn.decomposition import PCA | |
| from sklearn.preprocessing import scale | |
| __author__ = 'aGn' | |
| def apply_pca( | |
| file: 'word vec', | |
| ratio_similarity: "ratio_similarity to real data", | |
| scale_type: 'Type of Scale method'=1 | |
| )->'applied pca': | |
| """ | |
| Applying PCA and Normalization. | |
| :param file: Input Vector | |
| :param ratio_similarity: | |
| :param scale_type: There is two types of scale, whether using .scale() method which annotated | |
| by 1 or using PCA default scaling method which annotated by 2. | |
| :return: Applied PCA vector. | |
| """ | |
| df = pd.read_csv(file, header=None, delimiter=" ", quoting=csv.QUOTE_NONE, encoding='utf-8') | |
| limited_df = df.iloc[:, 1:] # Remove first column | |
| whiten = False | |
| if scale_type == 1: | |
| '''Set scale using .scale() method.''' | |
| scaled_df = scale(limited_df) # Normalizing data. | |
| elif scale_type == 2: | |
| '''Init inner PCA scaling.''' | |
| scaled_df = limited_df | |
| whiten = True | |
| else: | |
| '''Without any scaling.''' | |
| scaled_df = limited_df | |
| pca = PCA(whiten=whiten) | |
| pca.fit(scaled_df) | |
| w = pca.components_.T | |
| y = pca.fit_transform(scaled_df) | |
| reached_ratio = 0 | |
| for best_indices, ratio_ in enumerate(pca.explained_variance_ratio_): | |
| reached_ratio += ratio_ | |
| if reached_ratio >= ratio_similarity: | |
| print(f"Reached ratio percentage is: {reached_ratio:.0%}") | |
| break | |
| print(f"Number of PC is first {best_indices + 1} column(s)") | |
| file_name = 'pca_applied_on_' + file | |
| pca_applied = np.zeros((len(df), best_indices + 1)) | |
| pca_applied = pd.DataFrame(pca_applied) | |
| pca_applied.iloc[:, 1:] = y[:, :best_indices] | |
| pca_applied.iloc[:, 0] = df.iloc[:, 0] | |
| np.savetxt(file_name, pca_applied.values, fmt='%s') | |
| return file_name | |
| def usage(): | |
| return "exec type: python %s word_vec desired_ratio" % sys.argv[0] | |
| if __name__ == '__main__': | |
| print(apply_pca.__annotations__) | |
| if len(sys.argv) == 3: | |
| word_vec = sys.argv[1] | |
| desired_ratio = sys.argv[2] | |
| else: | |
| print(usage()) | |
| sys.exit(1) | |
| print('Please wait ...') | |
| applied_pca = apply_pca(word_vec, ratio_similarity=desired_ratio) | |
| print(f"Your destination file is: {applied_pca}") | |
| # print('Press any key to continue ...') | |
| # sys.stdin.read(1) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment