CODEWITHSUNDEEP
pythongoogle-bigquery
西浦 直志
西浦 直志

Reputation: 9

GCP can't write Biq query using by to_gbq

Can't write Biq Query by following Error.

Python 3.5.6 
pandas-gbq 0.13.1 
google-cloud-bigquery 1.24.0

ImportError: pandas-gbq requires google-cloud-bigquery: cannot import name 'TimeoutGuard'

code:

sample_dataframe = pd.DataFrame(data_rows,columns['shop_name','category','nearest_station','telephone_number','address','DL_time'])
print(sample_dataframe)

sample_dataframe.to_gbq('NTT.aikidou2025', 'robotic-column-270803',if_exists = 'replace')

Upvotes: 0

Views: 3658

Answers (2)

Alexandre Moraes
Alexandre Moraes

Reputation: 4032

While you did not post the packages you imported and installed in your environment, this error is generally related to missing required packages.

I was able to reproduce your case using Pandas and to_gbq method successfully without any error. For my attempt I used a Jupyter Notebook in a Cloud AI instance running Python 3.7.

First, I installed the following packages in my environment:

!pip install --upgrade google-bigquery[pandas] --quiet
!pip install --upgrade pandas_gbq

The second module (pandas_gbq) is necessary because it is not included in the google-bigquery[pandas] package, you can check the documentation here.

Subsequently, within the python script it's necessary to import pandas and bigquery. I also created a dummy dataframe in order to reproduce the case. As following: 

import pandas as pd
from google.cloud import bigquery

records =[
    {
        "Name": "Alex",
        "Age": 25,
        "City":"New York"
    },
    {
        "Name": "Bryan",
        "Age": 27,
        "City":"San Francisco"

    }
]

dataframe = pd.DataFrame(
    records,columns=["Name","Age","City"])

print(dataframe)

And the output:

    Name  Age           City
0   Alex   25       New York
1  Bryan   27  San Francisco

Finally, I used the to_gbq method: 

#to_gbq
dataframe.to_gbq('sample.pandas_bq_test',project_id="test-proj-261014",if_exists='append')

It was well executed and the data frame was in BigQuery. Therefore, I encourage you to check above if you properly installed all the packages I used. In addition, you can use pip show <name_of_the_package> to check whether it is installed and its version.

Update

To use BigQuery and Pandas with DataLab, it is possible to use a virtual environment, read more about it here. So, it is assured all the necessary python dependencies are installed and no incompatibility is encountered.

I followed the following steps to run the above code in DataLab.

  1. Create a DataLab instance and connection via http://localhost:8081/, following the documentation.
  2. Open a new notebook and select Kernel Python 3

Run the below commands, where is the name of your virtual environment.

!pip install virtualenv
!virtualenv <your-env>
!source <your-env>/bin/activate
!<your-env>/bin/pip install google-cloud-bigquery

Now you will be able to use import pandas as pd and from google.cloud import bigquery. I have tested with the code I provided above and it worked. Let me know if you have any issues.

Upvotes: 1

西浦 直志
西浦 直志

Reputation: 9

I tried !pip install google-cloud-bigquery==1.10.1 and this solved the error, it seems like proper for my Python version.

Upvotes: 0

Related Questions