It seems likely that you can export data from a Phoenix dashboard using the `get_spans_dataframe` method. This method allows you to pull a dataframe of the traces and optionally filter them based on specific criteria. You can then manipulate this dataframe to include only the columns you are interested in, such as test name, start time, total tokens, status, and latency.
Here's a basic example of how you might achieve this:
```
import phoenix as px
# Get the active session
session = px.active_session()
# Export all traces to a dataframe
df = session.get_spans_dataframe()
# Select only the columns you are interested in
filtered_df = df[['test_name', 'start_time', 'total_tokens', 'status', 'latency']]
```
Would you like more detail on this process?
Sources:
- <https://github.com/Arize-ai/phoenix/issues/2354|https://github.com/Arize-ai/phoenix/issues/2354>
- <https://docs.arize.com/phoenix/~/changes/v6Zhm276x8LlKmwqElIA/integrations/bring-production-data-to-notebook-for-eda-or-retraining|https://docs.arize.com/phoenix/~/changes/v6Zhm276x8LlKmwqElIA/integrations/bring-production-data-to-notebook-for-eda-or-retraining>
