new field in the transitions is next_section_and_step which is optional
and defaults to None but also can be "section_3:step_1"
modified: app.py
new file: research/activity0.yaml
new file: research/activity13-choose-adventure.yaml
modified: research/guarded_ai.py
modified: app.py
new file: migrations/versions/d04950c5a624_add_activitystate_table2.py
new file: migrations/versions/d3631b8bb652_add_activitystate_table.py
new file: research/activity.yaml
new file: research/activity10.yaml
new file: research/activity11.yaml
new file: research/activity12.yaml
new file: research/activity2.yaml
new file: research/activity3.yaml
new file: research/activity4.yaml
new file: research/activity5.yaml
new file: research/activity6.yaml
new file: research/activity7.yaml
new file: research/activity8.yaml
new file: research/activity9.yaml
new file: research/guarded_ai.py
This commit addresses the issue where the Claude model was throwing an error due to multiple consecutive "user" roles in the chat history. The following changes have been made:
- Implemented a new function `group_consecutive_roles` to group consecutive messages of the same role into one.
- Updated the `chat_claude`, function to use the `group_consecutive_roles` function before sending the chat history to the respective models.
By grouping consecutive messages of the same role, the chat history now alternates between "user" and "assistant" roles, resolving the "roles must alternate between 'user' and 'assistant'" error from the Claude model.
I replaced gpt-3.5-turbo workloads with openchat a local GPU powered inference server
The openchat inference server supports using the latest and official openai python client.
This means you can replace both standard and streaming workloads with an "offline" LLM.
modified: README.rst
modified: app.py
