Convert array to string with fixed decimal precision, multiprocessing with pandarallel based on pandas df

array_to_string.py

import json
import numpy as np
from pandarallel import pandarallel

pandarallel.initialize(progress_bar=True)

# Function to round array to 2 decimal places and serialize to JSON
def round_and_serialize(x):
    if isinstance(x, np.ndarray):
        # Round and format each number to 2 decimal places
        rounded_array = [round(float(num), 2) for num in x]
        return json.dumps(rounded_array)  # Serialize to JSON
    else:
        return json.dumps(x)

# Apply the function in parallel
sample["embeddings"] = sample["embeddings"].parallel_apply(round_and_serialize)