Created
August 28, 2018 19:10
-
-
Save paddy74/0900042929d5350be88e3db09ab53223 to your computer and use it in GitHub Desktop.
Create a memory-mapped array comprised of the concatenation of every .npy file in a directory
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
def concatenate_array_files(src_dir, file_name): | |
"""Create a memory-mapped array comprised of the concatenation of every | |
.npy file in a directory | |
Parameters | |
---------- | |
src_dir : str | |
Directory of 2D .npy arrays with matching feature counts | |
file_name : str | |
File in which to save the concatenated arrays | |
""" | |
total_samples = 0 | |
for f in os.listdir(src_dir): | |
# Load the next array | |
pat_arr = np.load(os.path.join(src_dir, f), mmap_mode='r') | |
n_samples, n_features = pat_arr.shape | |
total_samples += n_samples | |
#print("**************") | |
#print("File : " + f) | |
#print("n_samples : " + str(n_samples)) | |
#print("total_samples: " + str(total_samples)) | |
# Load the concatenated file | |
big_arr = np.memmap(target_file, dtype = np.float32, mode='w+', shape=(total_samples, n_features)) | |
big_arr[:n_samples, :] = pat_arr |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment