Skip to content

Commit e52d0f6

Browse files
committed
changes instructions for workshop
1 parent 4e87636 commit e52d0f6

1 file changed

Lines changed: 0 additions & 73 deletions

File tree

docs/workshop.md

Lines changed: 0 additions & 73 deletions
Original file line numberDiff line numberDiff line change
@@ -104,44 +104,6 @@ cp learning_observer/learning_observer/creds.yaml.workshop learning_observer/cre
104104

105105
If you have a file comparison tool like `meld`, it might be worth comparing our changes: `meld learning_observer/creds.yaml learning_observer/learning_observer/creds.yaml.example`
106106

107-
## Configuration: Create `creds.yaml`
108-
109-
Instead of copying `creds.yaml.workshop`, create a **minimal, local-friendly config** file at:
110-
111-
Paste this content:
112-
113-
```yaml
114-
# learning_observer/creds.yaml
115-
116-
server:
117-
host: 0.0.0.0
118-
port: 8888
119-
120-
db:
121-
backend: filesystem
122-
path: learning_observer/logs
123-
124-
auth:
125-
insecure_login: true # disable Google OAuth for workshop
126-
# Do not include any google_oauth block
127-
128-
modules:
129-
writing_observer:
130-
use_nlp: false
131-
use_languagetool: false
132-
133-
gpt_responders:
134-
# Option A: Ollama (local LLM, recommended for offline use)
135-
ollama:
136-
model: llama3.1:8b
137-
# host: http://localhost:11434 # uncomment if Ollama runs on a non-default host
138-
139-
# Option B: OpenAI (comment Ollama block and use this instead)
140-
# openai:
141-
# model: gpt-4o-mini
142-
# api_key_env: OPENAI_API_KEY
143-
144-
145107
## Test the system
146108

147109
To run the system, use the run command
@@ -150,41 +112,6 @@ To run the system, use the run command
150112
make run
151113
```
152114

153-
Notes:
154-
155-
If using OpenAI, make sure you export your key:
156-
157-
export OPENAI_API_KEY="your_api_key_here"
158-
159-
160-
If using Ollama, install and run it first (ollama run llama3.1:8b).
161-
162-
Run & Verify
163-
164-
Start the server (you may need to run this 1–3 times until setup completes):
165-
166-
make run
167-
168-
169-
You should see:
170-
171-
======== Running on http://0.0.0.0:8888 ========
172-
173-
174-
Then open in your browser:
175-
176-
http://localhost:8888/
177-
178-
or http://127.0.0.1:8888/
179-
180-
On Windows/WSL, if localhost:8888 doesn’t work, find your WSL IP:
181-
182-
hostname -I
183-
184-
185-
and open http://<that_ip>:8888/ in your Windows browser. Also check firewall settings if the page doesn’t load.
186-
187-
188115
*This does a lot of sanity checks on startup, and won't work the first time.* Rather, it will download required files, and create a file files (like `admins.yaml` and `teachers.yaml`, which are one way to define roles for teachers and admins on the system, but which we won't need for this workshop since we are using an insecure login). Once it is done, it will give you an opportunity to check whether it fixed issues correctly (we're working on having nice warnings, but we're not 100% of the way there). It did, so just run it again (perhaps 1-3 more times if it has more things to configure):
189116

190117
```bash

0 commit comments

Comments
 (0)