ArrowInvalid Exception: Failed to parse string: '' as a scalar of type timestamp[s]

#2
by RaymondAISG - opened

Tried loading the dataset via the following:

from datasets import load_dataset
data = load_dataset("common-pile/foodista_filtered")

Ended up with the following exception:

ArrowInvalid: Failed to parse string: '' as a scalar of type timestamp[s]

The above exception was the direct cause of the following exception:

DatasetGenerationError                    Traceback (most recent call last)
Cell In[2], line 1
----> 1 data = load_dataset("common-pile/foodista_filtered")

Hi, I ran into the same error when calling load_dataset("common-pile/foodista_filtered").

After checking the raw JSON file (foodista-dolma-0000.json.gz), I found that the field created contains empty strings ("").
In my case, it was 4090 out of 65640 rows.

As a workaround, I replaced "" with null before constructing the dataset.
Using the code below, I was able to load the data successfully:

import gzip
import json
from datasets import Dataset

rows = []

with gzip.open("foodista-dolma-0000.json.gz", "rt", encoding="utf-8") as f:
    for line in f:
        row = json.loads(line)
        if "created" in row and row["created"] == "":
            row["created"] = None
        rows.append(row)

ds = Dataset.from_list(rows)

Sign up or log in to comment