load_parquet.py

# import pandas as pd
#
# pd.set_option("display.max_colwidth", None)
# pd.set_option("display.max_columns", None)
# pd.set_option("display.width", None)
#
# # Load parquet file
# df = pd.read_parquet("/export/home/becky/verl-tool-mm-deepsearch/data/mm_deepsearch/fvqa_qw3vl/val_new.parquet")
#
# # Show first 5 rows
# print(df["images"].head(100))
#
# # Show basic info
# print(df.info())

import pandas as pd
import numpy as np

path = "/export/home/becky/verl-tool-mm-deepsearch/data/mm_deepsearch/fvqa_qw3vl/train_filtered.parquet"
prefix = "/export/home/becky/verl-tool-mm-deepsearch/"

# df = pd.read_parquet(path)

# def update_image_paths(image_array):
#     if isinstance(image_array, np.ndarray):
#         image_array = image_array.tolist()  # convert to list
#
#     if isinstance(image_array, list):
#         for item in image_array:
#             if isinstance(item, dict) and "image" in item:
#                 # avoid double prefix
#                 if not item["image"].startswith(prefix):
#                     item["image"] = prefix + item["image"]
#
#     return image_array
#
# df["images"] = df["images"].apply(update_image_paths)
# # save
# df.to_parquet(path, index=False)


pd.set_option("display.max_colwidth", None)
pd.set_option("display.max_columns", None)
pd.set_option("display.width", None)

# Load parquet file
df = pd.read_parquet(path)

# Show first 5 rows
print(df["images"].head(100))

# Show basic info
print(df.info())