Skip to content
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.

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

Add example for inference server #994

Closed
wants to merge 12 commits into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
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
18 changes: 18 additions & 0 deletions examples/inference_server/inference.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
import os
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we have a more realistic example in addition to (or instead of) this one? As pointed out in server.py, we need to send model weights and inputs to the cluster and get the prediction outputs from the cluster. Why don't we show such a complete example here?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That makes sense! I went with a simpler example because I don't have much experience with ML and inference - what do you think would be a good library and model to implement here?

import sys

INFERENCE_RESULT_MARKER = "INFERENCE RESULT:"


def run_inference(image_path):
# Perform some computation on the image located at image_path

# Instead of returning the result,
# print it to stdout so that the server can retrieve the result from the logs
print(
f"{INFERENCE_RESULT_MARKER}Ran inference on the image at '{image_path}' with size {os.path.getsize(image_path)}B."
)


if __name__ == "__main__":
run_inference(sys.argv[1])
1 change: 1 addition & 0 deletions examples/inference_server/requirements.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
flask
100 changes: 100 additions & 0 deletions examples/inference_server/server.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,100 @@
"""Flask inference server.

Implements a Flask server that handles inference requests for some input via a HTTP handler.
To run the server, run the following command from the root directory:
`FLASK_APP=examples/inference_server/server.py flask run`
"""

import os
import pathlib
import random
import re
import string
import subprocess

import flask
import sky
iojw marked this conversation as resolved.
Show resolved Hide resolved
from werkzeug import utils

import inference

LOCAL_UPLOAD_FOLDER = os.path.abspath('examples/inference_server/uploads/')
REMOTE_UPLOAD_FOLDER = '/remote/path/to/folder'
ALLOWED_EXTENSIONS = {'.png', '.jpg', '.jpeg'}

app = flask.Flask(__name__)
app.config['UPLOAD_FOLDER'] = LOCAL_UPLOAD_FOLDER


def run_output(cmd: str, **kwargs) -> str:
completed_process = subprocess.run(cmd,
stdout=subprocess.PIPE,
shell=True,
check=True,
**kwargs)
return completed_process.stdout.decode("ascii").strip()


def allowed_file(filename):
return os.path.splitext(filename)[1].lower() in ALLOWED_EXTENSIONS


@app.route("/", methods=["GET", "POST"])
def run_inference():
if flask.request.method == 'POST':
image = flask.request.files['file']
if not image or not allowed_file(image.filename):
flask.abort(400, "Invalid image upload")

filename = utils.secure_filename(image.filename)
local_image_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
remote_image_path = os.path.join(REMOTE_UPLOAD_FOLDER, filename)

pathlib.Path(app.config['UPLOAD_FOLDER']).mkdir(exist_ok=True)
image.save(local_image_path)

with sky.Dag() as dag:
workdir = os.path.dirname(os.path.abspath(__file__))
task_name = 'inference_task'
setup = 'pip3 install --upgrade pip'
run_fn = f'python inference.py {remote_image_path}'

task = sky.Task(name=task_name,
setup=setup,
workdir=workdir,
run=run_fn)

resources = sky.Resources(cloud=sky.Azure())
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there any reason for using Azure here?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I only had access to Azure at the time, so I used it for easier testing locally. I recently gained access to AWS as well - would you suggest to leave it as the default in this case?

task.set_resources(resources)
task.set_file_mounts({
# Copy model weights to the cluster
# Instead of local path, can also specify a cloud object store URI
'/remote/path/to/model-weights': 'local/path/to/model-weights',
# Copy image to the cluster
remote_image_path: local_image_path,
Comment on lines +70 to +74
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice point! Our storage APIs are indeed useful in sending model weights and inputs.

})

cluster_name = f'inference-cluster-{" ".join(random.choice(string.ascii_lowercase) for _ in range(10))}'
sky.launch(
dag,
cluster_name=cluster_name,
detach_run=True,
)

cmd_output = run_output(f'sky logs {cluster_name}')
inference_result = re.findall(
f'{inference.INFERENCE_RESULT_MARKER}((?:[^\n])+)', cmd_output)

# Down the cluster in the background
subprocess.Popen(f'sky down -y {cluster_name}', shell=True)

return {'result': inference_result}
elif flask.request.method == 'GET':
return '''
<title>Upload Image</title>
<h1>Upload Image</h1>
<form method=post enctype=multipart/form-data>
<input type=file name=file>
<input type=submit value=Upload>
</form>
'''