Skip to content

Minor fixes for local (non-Docker) evals #5604

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cte merged 2 commits into from
Jul 11, 2025
Merged

Minor fixes for local (non-Docker) evals #5604

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3a3f7d5
Minor fixes for local (non-Docker) evals
cte Jul 11, 2025
5cacfa3
Update packages/evals/README.md
cte Jul 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions packages/evals/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,26 +26,26 @@ echo "OPENROUTER_API_KEY=sk-or-v1-[...]" > packages/evals/.env.local
Start the evals service:

```sh
docker compose -f packages/evals/docker-compose.yml --profile server --profile runner up --build --scale runner=0
pnpm evals
```

The initial build process can take a minute or two. Upon success you should see output indicating that a web service is running on [localhost:3000](http://localhost:3000/):
<img width="1182" alt="Screenshot 2025-06-05 at 12 05 38 PM" src="https://github.com/user-attachments/assets/34f25a59-1362-458c-aafa-25e13cdb2a7a" />
The initial build process can take a minute or two. Upon success you should see ouput indicating that a web service is running on localhost:3000:
<img width="1182" src="https://github.com/user-attachments/assets/34f25a59-1362-458c-aafa-25e13cdb2a7a" />

Additionally, you'll find in Docker Desktop that database and redis services are running:
<img width="1283" alt="Screenshot 2025-06-05 at 12 07 09 PM" src="https://github.com/user-attachments/assets/ad75d791-9cc7-41e3-8168-df7b21b49da2" />
<img width="1283" src="https://github.com/user-attachments/assets/ad75d791-9cc7-41e3-8168-df7b21b49da2" />

Navigate to [localhost:3446](http://localhost:3446/) in your browser and click the 🚀 button.

By default a evals run will run all programming exercises in [Roo Code Evals](https://github.com/RooCodeInc/Roo-Code-Evals) repository with the Claude Sonnet 4 model and default settings. For basic configuration you can specify the LLM to use and any subset of the exercises you'd like. For advanced configuration you can import a Roo Code settings file which will allow you to run the evals with Roo Code configured any way you'd like (this includes custom modes, a footgun prompt, etc).

<img width="1053" alt="Screenshot 2025-06-05 at 12 08 06 PM" src="https://github.com/user-attachments/assets/2367eef4-6ae9-4ac2-8ee4-80f981046486" />
<img width="1053" src="https://github.com/user-attachments/assets/2367eef4-6ae9-4ac2-8ee4-80f981046486" />

After clicking "Launch" you should find that a "controller" container has spawned as well as `N` "task" containers where `N` is the value you chose for concurrency:
<img width="1283" alt="Screenshot 2025-06-05 at 12 13 29 PM" src="https://github.com/user-attachments/assets/024413e2-c886-4272-ab59-909b4b114e7c" />
<img width="1283" src="https://github.com/user-attachments/assets/024413e2-c886-4272-ab59-909b4b114e7c" />

The web app's UI should update in realtime with the results of the eval run:
<img width="1053" alt="Screenshot 2025-06-05 at 12 14 52 PM" src="https://github.com/user-attachments/assets/6fe3b651-0898-4f14-a231-3cc8d66f0e1f" />
<img width="1053" src="https://github.com/user-attachments/assets/6fe3b651-0898-4f14-a231-3cc8d66f0e1f" />

## Resource Usage

Expand All @@ -60,7 +60,7 @@ CPU Limit = 2 * concurrency

The memory and CPU limits can be set from the "Resources" section of the Docker Desktop settings:

<img width="996" alt="Screenshot 2025-06-06 at 8 54 24 AM" src="https://github.com/user-attachments/assets/a1cbb27d-b09c-450c-9fa8-b662c0537d48" />
<img width="996" src="https://github.com/user-attachments/assets/a1cbb27d-b09c-450c-9fa8-b662c0537d48" />

## Stopping

Expand Down
4 changes: 0 additions & 4 deletions packages/evals/docker-compose.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,8 +17,6 @@ services:
db:
container_name: evals-db
image: postgres:15.4
# expose:
# - 5432
ports:
- "${EVALS_DB_PORT:-5432}:5432"
volumes:
Expand All @@ -40,8 +38,6 @@ services:
redis:
container_name: evals-redis
image: redis:7-alpine
# expose:
# - 6379
ports:
- "${EVALS_REDIS_PORT:-6379}:6379"
volumes:
Expand Down
6 changes: 3 additions & 3 deletions packages/evals/src/cli/runEvals.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,16 +20,16 @@ export const runEvals = async (runId: number) => {
throw new Error(`Run ${run.id} has no tasks.`)
}

const containerized = isDockerContainer()

const logger = new Logger({
logDir: `/var/log/evals/runs/${run.id}`,
logDir: containerized ? `/var/log/evals/runs/${run.id}` : `/tmp/evals/runs/${run.id}`,
filename: `controller.log`,
tag: getTag("runEvals", { run }),
})

logger.info(`running ${tasks.length} task(s)`)

const containerized = isDockerContainer()

if (!containerized) {
await resetEvalsRepo({ run, cwd: EVALS_REPO_PATH })
}
Expand Down
5 changes: 3 additions & 2 deletions packages/evals/src/cli/runTask.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,10 +44,12 @@ export const processTask = async ({ taskId, logger }: { taskId: number; logger?:
const run = await findRun(task.runId)
await registerRunner({ runId: run.id, taskId })

const containerized = isDockerContainer()

logger =
logger ||
new Logger({
logDir: `/var/log/evals/runs/${run.id}`,
logDir: containerized ? `/var/log/evals/runs/${run.id}` : `/tmp/evals/runs/${run.id}`,
filename: `${language}-${exercise}.log`,
tag: getTag("runTask", { run, task }),
})
Expand Down Expand Up @@ -298,7 +300,6 @@ export const runTask = async ({ run, task, publish, logger }: RunTaskOptions) =>
...run.settings, // Allow the provided settings to override `openRouterApiKey`.
},
text: prompt,
newTab: true,
},
})

Expand Down
2 changes: 1 addition & 1 deletion packages/types/src/global-settings.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,7 +177,7 @@ export const EVALS_SETTINGS: RooCodeSettings = {
apiProvider: "openrouter",
openRouterUseMiddleOutTransform: false,

lastShownAnnouncementId: "may-29-2025-3-19",
lastShownAnnouncementId: "jul-09-2025-3-23-0",

pinnedApiConfigs: {},

Expand Down
Loading