Cleanup

apps/streamlit_ds_chat/README_04_vertex_ai_codegen_streamlit.md

@@ -1,6 +1,6 @@
 
 
-For testing manager the following questions are used:
+For testing manager, the following questions are used:
 
 <q>How many unique fruits are there in the dataset?</q>
 <q>How many fruits likes each person?</q>

apps/streamlit_ds_chat/README_05_putting_all_togather.md

@@ -1,4 +1,4 @@
-# Step 5. Put all components togather
+# Step 5. Put all components together
 
 Now I have code for:
 - running Streamlit.app
@@ -19,7 +19,7 @@ conda create -p ./env12 -c conda-forge coiled python=3.12 "numpy<2.0.0" streamli
 
 
 Known issues:
-- bokeh don't needed for workers but still is provided
+- Bokeh is not needed for workers but it is still provided.
 - StreamlitAPIException: Streamlit only supports Bokeh version 2.4.3, but you have version 3.5.0 installed. Please run `pip install --force-reinstall --no-deps bokeh==2.4.3` to install the correct version.
 - start of coiled cluster takes 1-2 minutes, but this will be the same for any cluster start on any tool
 - hard to acquire good code from code generation; it usually works, but the return value is not exactly what was requested

apps/streamlit_ds_chat/cloud_coiled_io/README.md

@@ -14,14 +14,14 @@ Your cloud credentials (securely)
 ```
 
 Works with AWS, GCloud, Azure, etc.
-Runs dusk and pure python.
+Runs Dask and pure Python.
 
 
 # How to install Coiled
 
 Just register on `coiled.io` and follow instructions. Everything is very simple and works great.
 
-The only pitfall: all you installs probably should go from single distribution system.
+The only pitfall: all you installs should probably come from single distribution system.
 If you use conda, you should install everything with `conda install`.
 If you use `pip`, you should install all packages with `pip install`.
 
@@ -34,7 +34,7 @@ coiled login
 > Authentication successful 🎉
 > Credentials have been saved at /home/s-nechuiviter/.config/dask/coiled.yaml
 
-Use `coiled setup gcp` once to setup connection to GCP.
+Use `coiled setup gcp` once to set up connection to GCP.
 
 Coiled requires that all modules should be `conda` compatible, at least if start from conda environment.
 Or use pip/venv everywhere.

apps/streamlit_ds_chat/ds_chat_streamlit_app.py

@@ -1,4 +1,3 @@
-import coiled
 import streamlit as st
 import importlib
 
@@ -34,6 +33,8 @@
 # Create or connect to a Coiled cluster
 with st.spinner("Creating or connecting to a Coiled cluster"):
     if "cluster" not in st.session_state:
+        import coiled
+
         cluster = coiled.Cluster(
             name="my-cluster", n_workers=1, idle_timeout="20 minutes"
         )
@@ -67,6 +68,7 @@
 
 if not uploaded_file:
     st.error("Upload dataset to work with")
+    st.stop()
 else:
     # if uploaded_file:  # and question:  # and anthropic_api_key:
     dataframe = pd.read_csv(uploaded_file)

apps/streamlit_ds_chat/readme.md

@@ -1,6 +1,16 @@
+Title
+feat: add Streamlit app for DS chat with Google Vertex AI and Coiled integration
+
+Overview
+Introduced a new Streamlit application for data science chat functionality using Google Vertex AI.
+Added multiple supporting scripts and modules for handling various functionalities such as code generation, cloud execution, and data visualization.
+Integrated Coiled for distributed computing and remote code execution.
+Provided detailed documentation and examples for setting up the environment, using the application, and testing different components.
+
+
 # Setting up a local development environment
 
-Before we can actually start building Streamlit apps, we will first have to setup a development environment.
+Before we can actually start building Streamlit apps, we will first have to set up a development environment.
 
 Let's start by installing and setting up a conda environment.
 

projects/ds_chat/20_in_progress/2024-07-02_use_streamlit_as_frontend.md

@@ -1,7 +1,7 @@
 Use https://streamlit.io/ as chat frontend.
 
 Requirements:
-- simple Stream.lit app - empty app - done
+- simple Streamlit app - empty app - done
 - connect to backend
 - get user questions
 - print answers