Skip to content


1. Tutorial code stopped working after the latest update.

This might happen when there is an update to the ThanoSQL syntax. If you have any further questions about this issue, please contact us at and we will look into it as soon as possible. 

2. I got an error when using capital letters in ThanoSQL.

ThanoSQL handles model/column names in lowercase letters. Please use only lowercase letters for model names or column names to call the appropriate model or column.

3. When I used the COPY statement to read a csv file, I received an encoding error.

The COPY clause in ThanoSQL only supports the utf-8 encoding format. For csv files containing Korean, use the COPY clause after the appropriate utf-8 encoding.

4. What types of unstructured data does ThanoSQL support?

ThanoSQL supports the following extensions for image, audio, and video data:

  • Image:  "jpg", "png"
  • Audio: "wav", "flac", "mp3"
  • Video: "mp4"

ThanoSQL does not require any special extensions for text data.

For image, extensions other than the two mentioned may not work reliably.

5. How do I auto close brackets?

You can turn on the feature by clicking on Menu > Settings > Auto Close Brackets.

This also applies for quotation marks.

6. When using a LIKE statement, the “%” operator does not work.

Since ThanoSQL uses “%” as an escape character, you need to use “%%” instead.

7. How do I decode base64 encoded DataFrame columns?

When using ThanoSQL's REST API, columns created using "CONVERT" are encoded using base64. Therefore, to use it as a column containing bytes, it must be decoded using base64's b64decode.

# to decode back to array an of bytes
import base64
df['column name with encoded values'] = df['column name with encoded values'].apply(lambda x: base64.b64encode(x))

# to decode array of bytes to array of floats 
import numpy as np 
df['column name with array of bytes'] = df['column name with array of bytes'].apply(lambda x: list(np.frombuffer(x, dtype="float32")))

Last update: 2023-01-25