How do I append a row to an empty Dataframe using Pandas

Question

I am iterating over rows in a dataframe to see if an image is not found on a server. If it's not found, then it removes the row from the dataframe and makes a new CSV file. I would like to take that row that contains the image not found on the server and append it to a new dataframe that will create a new CSV that we can investigate why an image is not found. I can successfully remove the row from the original dataframe, but appending the new row to the new dataframe results in an empty dataframe once run. How might I solve this

filename2 = 'images_not_found.csv'
orders_csv = pd.read_csv(path + filename)
not_found_images = pd.DataFrame()


for index, row in orders_csv.iterrows():
    if not os.path.isfile(row['Image']):
        orders_csv.drop(index, inplace=True)
        not_found_images.append(row, ignore_index=True)

orders_csv.to_csv(path + filename, index=False)
not_found_images.to_csv(path + filename2, index=False)

itaishz · Accepted Answer · 2020-12-08 16:50:38Z

2

Unlike a list, df.append(new_df) doesn't append inline, meaning doesn't add new_df to df. You need to define df again:

df = df.append(new_df)

In your code:

not_found_images = not_found_images.append(row, ignore_index=True)

answered Dec 8, 2020 at 16:50

itaishz

7111 gold badge4 silver badges10 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Whole Brain · Accepted Answer · 2020-12-08 16:50:30Z

1

It might sound confusing, but the append method of pandas DataFrame doesn't work in place. You have to re-assign the resulting Dataframe to your variable.

not_found_images = not_found_images.append(row, ignore_index=True)

answered Dec 8, 2020 at 16:50

Whole Brain

2,1772 gold badges13 silver badges20 bronze badges

1 Comment

Whole Brain Over a year ago

You're welcome. Consider accepting my solution.

Collectives™ on Stack Overflow

How do I append a row to an empty Dataframe using Pandas

2 Answers 2

Comments

1 Comment

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Related