def truncate_and_save()

in tbsm_data_pytorch.py [0:0]

• 30 lines of code
• 5 McCabe index (conditional complexity)

    def truncate_and_save(self, out_file, do_shuffle, t, users, items, cats, times, y):

        # truncate. If for some users we didn't generate had too short history
        # we truncate the unused portion of the pre-allocated matrix.
        if t < self.total_out:
            users = users[:t, :]
            items = items[:t, :]
            cats = cats[:t, :]
            times = times[:t, :]
            y = y[:t]

        # shuffle
        if do_shuffle:
            indices = np.arange(len(y))
            indices = np.random.permutation(indices)
            users = users[indices]
            items = items[indices]
            cats = cats[indices]
            times = times[indices]
            y = y[indices]

        N = len(y)
        X_cat = np.zeros((3, N, self.ts_length + 1), dtype="i4")  # 4 byte int
        X_int = np.zeros((1, N, self.ts_length + 1), dtype=np.float)
        X_cat[0, :, :] = users
        X_cat[1, :, :] = items
        X_cat[2, :, :] = cats
        X_int[0, :, :] = times

        # saving to compressed numpy file
        if not path.exists(out_file):
            np.savez_compressed(
                out_file,
                X_cat=X_cat,
                X_int=X_int,
                y=y,
            )
        return