Skip to content

armgit / burr

Public Imported from GitHub
0

Build applications that make decisions (chatbots, agents, simulations, etc...). Monitor, trace, persist, and execute on your own infrastructure.

Changes the gpt app example to use inputs

elijahbenizzy5a4d72265bbc
diff --git a/burr/core/action.py b/burr/core/action.py
index d10cd024..f806dbae 100644
--- a/burr/core/action.py
+++ b/burr/core/action.py
@@ -56,7 +56,7 @@ class Function(abc.ABC):
if missing_inputs or additional_inputs:
raise ValueError(
f"Inputs to function {self} are invalid. "
- f"Missing the following inputs: {', '.join(missing_inputs)}."
+ + f"Missing the following inputs: {', '.join(missing_inputs)}."
if missing_inputs
else "" f"Additional inputs: {','.join(additional_inputs)}."
if additional_inputs
diff --git a/examples/gpt/application.py b/examples/gpt/application.py
index 47a8c810..734115d2 100644
--- a/examples/gpt/application.py
+++ b/examples/gpt/application.py
@@ -18,12 +18,12 @@ MODES = {
}
-@action(reads=["prompt"], writes=["chat_history"])
-def process_prompt(state: State) -> Tuple[dict, State]:
- result = {"processed_prompt": {"role": "user", "content": state["prompt"], "type": "text"}}
+@action(reads=[], writes=["chat_history", "prompt"])
+def process_prompt(state: State, prompt: str) -> Tuple[dict, State]:
+ result = {"processed_prompt": {"role": "user", "content": prompt, "type": "text"}}
return result, state.wipe(keep=["prompt", "chat_history"]).append(
chat_history=result["processed_prompt"]
- )
+ ).update(prompt=prompt)
@action(reads=["prompt"], writes=["safe"])
diff --git a/examples/gpt/streamlit_app.py b/examples/gpt/streamlit_app.py
index 13cb8eff..ed114678 100644
--- a/examples/gpt/streamlit_app.py
+++ b/examples/gpt/streamlit_app.py
@@ -12,6 +12,13 @@ from burr.integrations.streamlit import (
update_state,
)
+st.set_page_config(layout="wide")
+
+"""This file is a work in progress. Note that this will be replaced shortly
+with a simple chatbot + a reference to the Burr UI (to be run in a separate page),
+which is more powerful and much simpler to understand. For now,
+this is purely for demos."""
+
def render_chat_message(record: Record):
if record.action in ["prompt", "response"]:
@@ -32,6 +39,8 @@ def render_chat_message(record: Record):
def retrieve_state():
if "burr_state" not in st.session_state:
+ # TODO --enable usage of hamilton. Currently its not wiring in inputs
+ # But it should be easy enough
state = AppState.from_empty(
app=chatbot_application.application(use_hamilton=False),
)
@@ -47,15 +56,17 @@ def chatbot_step(app_state: AppState, prompt: Optional[str]) -> bool:
:param prompt: Prompt to set the chatbot to. If this is None it means it should continue and not be reset.
:return:
"""
+ inputs = None
if prompt is not None:
# We need to update
- app_state.app.update_state(app_state.app.state.update(prompt=prompt))
+ inputs = {"prompt": prompt}
+ # app_state.app.update_state(app_state.app.state.update(prompt=prompt))
st.session_state.running = True # set to running
# if its not running this is a no-op
if not st.session_state.get("running", False):
return False
application = app_state.app
- step_output = application.step()
+ step_output = application.step(inputs=inputs)
if step_output is None:
st.session_state.running = False
return False
@@ -70,7 +81,6 @@ def chatbot_step(app_state: AppState, prompt: Optional[str]) -> bool:
def main():
- st.set_page_config(layout="wide")
st.title("Chatbot example with Burr")
app_state = retrieve_state() # retrieve first so we can use for the ret of the step
columns = st.columns(2)