Browse Source

announcement-functional block hosting

main
Michael Pilosov 2 years ago
parent
commit
c55ce23d76
  1. 1
      .gitignore
  2. 14
      Dockerfile
  3. 2
      app/__init__.py
  4. 50
      app/routes.py
  5. 5
      docker-compose.yml
  6. 57
      example_block/c2.py
  7. 41
      example_block/client.py
  8. 47
      example_block/docker-compose.yml
  9. 42
      example_block/eden-server/Dockerfile
  10. 55
      example_block/eden-server/announce.py
  11. 515
      example_block/eden-server/hosting.py
  12. 75
      example_block/eden-server/image_utils.py
  13. 69
      example_block/eden-server/server copy.py
  14. 123
      example_block/eden-server/server.py
  15. 24
      example_block/nginx.conf
  16. 12
      example_block/redis/Dockerfile
  17. 66
      example_block/s2.py
  18. 66
      example_block/server.py
  19. 75
      image_utils.py
  20. 1
      index.html
  21. 77
      registry/host.py
  22. 2
      registry/requirements.txt
  23. 39
      registry/s1.py
  24. 39
      registry/s2.py
  25. 1
      requirements.txt

1
.gitignore

@ -140,3 +140,4 @@ cython_debug/
# Project-specific files # Project-specific files
cookies.txt cookies.txt
*.dump.rdb

14
Dockerfile

@ -1,6 +1,16 @@
FROM python:3.9 FROM python:3.10
RUN apt-get update -yqq && apt-get install -y \
libgl1-mesa-glx \
&& rm -rf /var/lib/apt/lists/*
WORKDIR /app WORKDIR /app
RUN useradd -ms /bin/bash eden
# make them own /app
RUN chown eden:eden /app
USER eden
# add /home/eden/.local/bin to PATH
ENV PATH="/home/eden/.local/bin:${PATH}"
COPY requirements.txt . COPY requirements.txt .
RUN pip install -r requirements.txt RUN pip install -r requirements.txt
@ -8,5 +18,7 @@ RUN pip install -r requirements.txt
COPY app app COPY app app
EXPOSE 5000 EXPOSE 5000
# attempted bugfix
COPY image_utils.py /home/eden/.local/lib/python3.10/site-packages/eden/image_utils.py
CMD ["flask", "run", "--debug", "--host=0.0.0.0"] CMD ["flask", "run", "--debug", "--host=0.0.0.0"]

2
app/__init__.py

@ -3,7 +3,7 @@ from flask_sqlalchemy import SQLAlchemy
from flask_cors import CORS from flask_cors import CORS
app = Flask(__name__) app = Flask(__name__)
CORS(app, supports_credentials=True) # CORS(app, supports_credentials=True)
app.config.from_pyfile('config.py') app.config.from_pyfile('config.py')
db = SQLAlchemy(app) db = SQLAlchemy(app)

50
app/routes.py

@ -1,6 +1,10 @@
from flask import request, jsonify, redirect, url_for, render_template from flask import request, jsonify, redirect, url_for, render_template
from app import app, db from app import app, db
# from app.models import User # from app.models import User
import PIL
from eden.client import Client
from eden.datatypes import Image
import time
@app.route('/api', methods=['POST']) @app.route('/api', methods=['POST'])
def process_request(): def process_request():
@ -13,4 +17,48 @@ def process_request():
# log to flask log the blocks list. # log to flask log the blocks list.
app.logger.info(blocks) app.logger.info(blocks)
app.logger.info(text) app.logger.info(text)
return jsonify(success=True)
results = communicate_with_eden(app, image)
return jsonify(results)
# return jsonify(success=True)
def communicate_with_eden(app, image, ip_address="172.18.0.3", port="5656"):
url = f"http://{ip_address}:{port}"
## set up a client
c = Client(url=url, username="abraham")
# get server's identity
generator_id = c.get_generator_identity()
print(generator_id)
app.logger.info("setting config")
## define input args to be sent
config = {
"width": 2000, ## width
"height": 1000, ## height
"input_image": Image(
PIL.Image.open(image.stream)
), ## images require eden.datatypes.Image()
}
app.logger.info("set config, running now")
# start the task
run_response = c.run(config)
results = c.fetch(token=run_response["token"])
print("Intitial response")
# check status of the task, returns the output too if the task is complete
# results = c.await_results(token=run_response["token"], interval=1, show_progress=False)
i = 0
while results["status"].get("status") != "complete":
results = c.fetch(token=run_response["token"])
print(results)
time.sleep(0.1)
i += 1
if i > 50:
break
return results

5
docker-compose.yml

@ -23,3 +23,8 @@ services:
volumes: volumes:
db-data: db-data:
networks:
default:
name: eden-network
external: true

57
example_block/c2.py

@ -0,0 +1,57 @@
import time
from eden.client import Client
from eden.datatypes import Image
import subprocess
import socket
# Get IP address of eden-server service
hostname = 'eden-server'
port = 5656
network_name = 'eden-network'
import docker
client = docker.from_env()
project_name = 'not_so_minimal'
container_name = f'{project_name}_{hostname}_1'
container = client.containers.get(container_name)
ip_address = container.attrs['NetworkSettings']['Networks'][network_name]['IPAddress']
print(ip_address)
url = f"http://{ip_address}:{port}"
## set up a client
c = Client(url=url, username="abraham")
# get server's identity
generator_id = c.get_generator_identity()
print(generator_id)
## define input args to be sent
config = {
"width": 2000, ## width
"height": 1000, ## height
"input_image": Image(
"/home/mm/Downloads/FF06F0EC-1B54-458A-BF12-FF7FC2A43C10.jpeg"
), ## images require eden.datatypes.Image()
}
# start the task
run_response = c.run(config)
print("Intitial response")
# check status of the task, returns the output too if the task is complete
results = c.fetch(token=run_response["token"])
print(results)
# one eternity later
# time.sleep(5)
print("Trying")
while results["status"].get("status") != "complete":
results = c.fetch(token=run_response["token"])
print(results)
time.sleep(0.1)
## check status again, hopefully the task is complete by now
# results = c.fetch(token=run_response["token"])
# print(results)
# results['output']['image'].show()

41
example_block/client.py

@ -0,0 +1,41 @@
import time
from eden.client import Client
from eden.datatypes import Image
## set up a client
c = Client(url="http://0.0.0.0:5656", username="abraham")
# get server's identity
generator_id = c.get_generator_identity()
print(generator_id)
## define input args to be sent
config = {
"width": 2000, ## width
"height": 1000, ## height
"input_image": Image(
"/home/mm/Downloads/FF06F0EC-1B54-458A-BF12-FF7FC2A43C10.jpeg"
), ## images require eden.datatypes.Image()
}
# start the task
run_response = c.run(config)
print("Intitial response")
# check status of the task, returns the output too if the task is complete
results = c.fetch(token=run_response["token"])
print(results)
# one eternity later
# time.sleep(5)
print("Trying")
while results["status"].get("status") != "complete":
results = c.fetch(token=run_response["token"])
print(results)
time.sleep(0.1)
## check status again, hopefully the task is complete by now
# results = c.fetch(token=run_response["token"])
# print(results)
# results['output']['image'].show()

47
example_block/docker-compose.yml

@ -0,0 +1,47 @@
# docker-compose for redis service defined in ./redis
version: '3.7'
services:
redis:
build: ./redis
image: redis
ports:
- "6379:6379"
volumes:
- ./data:/data
networks:
- default
# eden server, started with python server.py, based on Dockerfile in cwd.
eden-server:
build: ./eden-server
image: eden-server
# ports:
# - "5656:5656"
volumes:
- /home/mm/.cache/torch/hub/checkpoints/resnet50-11ad3fa6.pth:/root/.cache/torch/hub/checkpoints/resnet50-11ad3fa6.pth
networks:
- default
depends_on:
- redis
# pass nvidia gpu
runtime: nvidia
environment:
- CUDA_VISIBLE_DEVICES=0
- NVIDIA_VISIBLE_DEVICES=0
# load-balancer:
# image: nginx
# ports:
# - "5656:80"
# volumes:
# - ./nginx.conf:/etc/nginx/nginx.conf:ro
# networks:
# - default
# depends_on:
# - eden-server
networks:
default:
name: eden-network
external: true

42
example_block/eden-server/Dockerfile

@ -0,0 +1,42 @@
FROM pytorch/pytorch:1.13.1-cuda11.6-cudnn8-runtime
RUN apt-get update && apt-get install -y \
libgl1-mesa-glx \
libglib2.0-0 \
&& rm -rf /var/lib/apt/lists/*
# until we hack around gitpython, we need git
# RUN apt-get update && apt-get install -y \
# git \
# && rm -rf /var/lib/apt/lists/*
WORKDIR /app
# create a safe user
RUN useradd -ms /bin/bash eden
# make them own /app
RUN chown eden:eden /app
USER eden
# add /home/eden/.local/bin to PATH
ENV PATH="/home/eden/.local/bin:${PATH}"
RUN pip install eden-python
RUN pip install python-socketio[asyncio_server] aiohttp
COPY server.py .
# attempted bugfix
COPY image_utils.py /home/eden/.local/lib/python3.10/site-packages/eden/image_utils.py
# attempt git-python hackaround
COPY hosting.py /home/eden/.local/lib/python3.10/site-packages/eden/hosting.py
EXPOSE 5656
# ENV GIT_PYTHON_REFRESH=quiet
# hack around gitpython
# RUN git init .
# RUN git config --global user.email "none@site.com"
# RUN git config --global user.name "eden-service-user"
# # add fake remote upstream
# RUN git remote add origin https://git.clfx.cc/mm/eden-app.git
# RUN git add server.py
# RUN git commit -am "initial commit"
ENV GIT_PYTHON_REFRESH=quiet
CMD ["python", "server.py"]

55
example_block/eden-server/announce.py

@ -0,0 +1,55 @@
import asyncio
from functools import wraps
import socketio
import socket
def get_ip_address():
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
try:
# This IP address doesn't need to be reachable, as we're only using it to find the local IP address
s.connect(("10.255.255.255", 1))
ip = s.getsockname()[0]
except Exception:
ip = "127.0.0.1"
finally:
s.close()
return ip
# Update these with the correct values for your host and server
HOST_SERVER_IP = "192.168.1.113"
HOST_SERVER_PORT = 4999
SERVER_NAME = "server_1"
SERVER_IP = get_ip_address()
SERVER_PORT = 8000
sio = socketio.AsyncClient()
async def announce_server():
await sio.connect(f'http://{HOST_SERVER_IP}:{HOST_SERVER_PORT}')
await sio.emit('register', {'name': SERVER_NAME, 'ip': SERVER_IP, 'port': SERVER_PORT})
@sio.on("heartbeat")
async def on_heartbeat():
print("Received heartbeat from host")
@sio.event
async def disconnect():
print("Disconnected from host")
def announce_server_decorator(host_block_function):
@wraps(host_block_function)
def wrapper(*args, **kwargs):
loop = asyncio.get_event_loop()
# Start the server announcement task
announce_task = loop.create_task(announce_server())
# Run the original host_block function
result = host_block_function(*args, **kwargs)
# Cancel the announcement task after the host_block function is done
announce_task.cancel()
return result
return wrapper

515
example_block/eden-server/hosting.py

@ -0,0 +1,515 @@
import os
import git
import warnings
import uvicorn
import logging
from fastapi import FastAPI
from prometheus_client import Gauge
from starlette_exporter import PrometheusMiddleware, handle_metrics
from fastapi.middleware.cors import CORSMiddleware
from .datatypes import Image
from .queue import QueueData
from .log_utils import Colors
from .models import Credentials, WaitFor
from .result_storage import ResultStorage
from .config_wrapper import ConfigWrapper
from .data_handlers import Encoder, Decoder
from .threaded_server import ThreadedServer
from .progress_tracker import fetch_progress_from_token
from .log_utils import log_levels, celery_log_levels, PREFIX
from .prometheus_utils import PrometheusMetrics
from .utils import stop_everything_gracefully, generate_random_string
from uvicorn.config import LOGGING_CONFIG
"""
Celery+redis is needed to be able to queue tasks
"""
from celery import Celery
from .celery_utils import run_celery_app
"""
tool to allocate gpus on queued tasks
"""
from .gpu_allocator import GPUAllocator
def host_block(
block,
port=8080,
host="0.0.0.0",
max_num_workers=4,
redis_port=6379,
redis_host="localhost",
requires_gpu=True,
log_level="warning",
logfile="logs.log",
exclude_gpu_ids: list = [],
remove_result_on_fetch = False
):
"""
Use this to host your eden.Block on a server. Supports multiple GPUs and queues tasks automatically with celery.
Args:
block (eden.block.Block): The eden block you'd want to host.
port (int, optional): Localhost port where the block would be hosted. Defaults to 8080.
host (str): specifies where the endpoint would be hosted. Defaults to '0.0.0.0'.
max_num_workers (int, optional): Maximum number of tasks to run in parallel. Defaults to 4.
redis_port (int, optional): Port number for celery's redis server. Defaults to 6379.
redis_host (str, optional): Place to host redis for `eden.queue.QueueData`. Defaults to localhost.
requires_gpu (bool, optional): Set this to False if your tasks dont necessarily need GPUs.
log_level (str, optional): Can be 'debug', 'info', or 'warning'. Defaults to 'warning'
logfile(str, optional): Name of the file where the logs would be stored. If set to None, it will show all logs on stdout. Defaults to 'logs.log'
exclude_gpu_ids (list, optional): List of gpu ids to not use for hosting. Example: [2,3]
"""
"""
Response templates:
/run:
{
'token': some_long_token,
}
/fetch:
if task is queued:
{
'status': {
'status': queued,
'queue_position': int
},
config: current_config
}
elif task is running:
{
'status': {
'status': 'running',
'progress': float between 0 and 1,
},
config: current_config,
'output': {} ## optionally the user should be able to write outputs here
}
elif task failed:
{
'status': {
'status': 'failed',
}
'config': current_config,
'output': {} ## will still include the outputs if any so that it gets returned even though the task failed
}
elif task succeeded:
{
'status': {
'status': 'complete'
},
'output': user_output,
'config': config
}
"""
"""
Initiating celery app
"""
celery_app = Celery(__name__, broker=f"redis://{redis_host}:{str(redis_port)}")
celery_app.conf.broker_url = os.environ.get(
"CELERY_BROKER_URL", f"redis://{redis_host}:{str(redis_port)}"
)
celery_app.conf.result_backend = os.environ.get(
"CELERY_RESULT_BACKEND", f"redis://{redis_host}:{str(redis_port)}"
)
celery_app.conf.task_track_started = os.environ.get(
"CELERY_TRACK_STARTED", default=True
)
celery_app.conf.worker_send_task_events = True
celery_app.conf.task_send_sent_event = True
"""
each block gets its wown queue
"""
celery_app.conf.task_default_queue = block.name
"""
set prefetch mult to 1 so that tasks dont get pre-fetched by workers
"""
celery_app.conf.worker_prefetch_multiplier = 1
"""
task messages will be acknowledged after the task has been executed
"""
celery_app.conf.task_acks_late = True
"""
Initiating GPUAllocator only if requires_gpu is True
"""
if requires_gpu == True:
gpu_allocator = GPUAllocator(exclude_gpu_ids=exclude_gpu_ids)
else:
print(PREFIX + " Initiating server with no GPUs since requires_gpu = False")
if requires_gpu == True:
if gpu_allocator.num_gpus < max_num_workers:
"""
if a task requires a gpu, and the number of workers is > the number of available gpus,
then max_num_workers is automatically set to the number of gpus available
this is because eden assumes that each task requires one gpu (all of it)
"""
warnings.warn(
"max_num_workers is greater than the number of GPUs found, overriding max_num_workers to be: "
+ str(gpu_allocator.num_gpus)
)
max_num_workers = gpu_allocator.num_gpus
"""
Initiating queue data to keep track of the queue
"""
queue_data = QueueData(
redis_port=redis_port, redis_host=redis_host, queue_name=block.name
)
"""
Initiate encoder and decoder
"""
data_encoder = Encoder()
data_decoder = Decoder()
"""
Initiate fastAPI app
"""
app = FastAPI()
origins = ["*"]
app.add_middleware(
CORSMiddleware,
allow_origins=origins,
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
app.add_middleware(PrometheusMiddleware)
app.add_route("/metrics", handle_metrics)
"""
Initiate result storage on redis
"""
result_storage = ResultStorage(
redis_host=redis_host,
redis_port=redis_port,
)
## set up result storage and data encoder for block
block.result_storage = result_storage
block.data_encoder = data_encoder
"""
initiate a wrapper which handles 4 metrics for prometheus:
* number of queued jobs
* number of running jobs
* number of failed jobs
* number of succeeded jobs
"""
prometheus_metrics = PrometheusMetrics()
"""
define celery task
"""
@celery_app.task(name="run")
def run(args, token: str):
## job moves from queue to running
prometheus_metrics.queued.dec(1)
prometheus_metrics.running.inc(1)
args = data_decoder.decode(args)
"""
allocating a GPU ID to the tast based on usage
for now let's settle for max 1 GPU per task :(
"""
if requires_gpu == True:
# returns None if there are no gpus available
gpu_name = gpu_allocator.get_gpu()
else:
gpu_name = None ## default value either if there are no gpus available or requires_gpu = False
"""
If there are no GPUs available, then it returns a sad message.
But if there ARE GPUs available, then it starts run()
"""
if (
gpu_name == None and requires_gpu == True
): ## making sure there are no gpus available
status = {
"status": "No GPUs are available at the moment, please try again later",
}
else:
"""
refer:
https://github.com/abraham-ai/eden/issues/14
"""
args = ConfigWrapper(
data=args,
token=token,
result_storage=result_storage,
gpu=None, ## will be provided later on in the run
progress=None, ## will be provided later on in the run
)
if requires_gpu == True:
args.gpu = gpu_name
if block.progress == True:
"""
if progress was set to True on @eden.Block.run() decorator, then add a progress tracker into the config
"""
args.progress = block.get_progress_bar(
token=token, result_storage=result_storage
)
try:
output = block.__run__(args)
# job moves from running to succeeded
prometheus_metrics.running.dec(1)
prometheus_metrics.succeeded.inc(1)
# prevent further jobs from hitting a busy gpu after a caught exception
except Exception as e:
# job moves from running to failed
prometheus_metrics.running.dec(1)
prometheus_metrics.failed.inc(1)
if requires_gpu == True:
gpu_allocator.set_as_free(name=gpu_name)
raise Exception(str(e))
if requires_gpu == True:
gpu_allocator.set_as_free(name=gpu_name)
success = block.write_results(output=output, token=token)
return success ## return None because results go to result_storage instead
@app.post("/run")
def start_run(config: block.data_model):
## job moves into queue
prometheus_metrics.queued.inc(1)
"""
refer:
https://github.com/celery/celery/issues/1813#issuecomment-33142648
"""
token = generate_random_string(len=10)
kwargs = dict(args=dict(config), token=token)
res = run.apply_async(kwargs=kwargs, task_id=token, queue_name=block.name)
initial_dict = {"config": dict(config), "output": {}, "progress": "__none__"}
success = result_storage.add(token=token, encoded_results=initial_dict)
response = {"token": token}
return response
@app.post("/update")
def update(credentials: Credentials, config: block.data_model):
token = credentials.token
config = dict(config)
status = queue_data.get_status(token=token)
if status["status"] != "invalid token":
if (
status["status"] == "queued"
or status["status"] == "running"
or status["status"] == "starting"
):
output_from_storage = result_storage.get(token=token)
output_from_storage["config"] = config
success = result_storage.add(
encoded_results=output_from_storage, token=token
)
response = {
"status": {
"status": "successfully updated config",
}
}
return response
elif status["status"] == "failed":
return {
"status": {
"status": "could not update config because job failed",
}
}
elif status["status"] == "complete":
return {
"status": {
"status": "could not update config because job is already complete",
}
}
else:
response = {"status": {"status": "invalid token"}}
return response
@app.post("/fetch")
def fetch(credentials: Credentials):
"""
Returns either the status of the task or the result depending on whether it's queued, running, complete or failed.
Args:
credentials (Credentials): should contain a token that points to a task
"""
token = credentials.token
status = queue_data.get_status(token=token)
if status["status"] != "invalid token":
if status["status"] == "running":
results = result_storage.get(token=token)
response = {
"status": status,
"config": results["config"],
"output": results["output"],
}
if block.progress == True:
progress_value = fetch_progress_from_token(
result_storage=result_storage, token=token
)
response["status"]["progress"] = progress_value
elif status["status"] == "complete":
results = result_storage.get(token=token)
## if results are deleted, it still returns the same schema
if results == None and remove_result_on_fetch == True:
response = {
"status": {
"status": "removed"
},
}
else:
response = {
"status": status,
"config": results["config"],
"output": results["output"],
}
if remove_result_on_fetch == True:
result_storage.delete(token=token)
elif (
status["status"] == "queued"
or status["status"] == "starting"
or status["status"] == "failed"
or status["status"] == "revoked"
):
results = result_storage.get(token=token)
response = {"status": status, "config": results["config"]}
else:
response = {"status": status} ## invalid token
return response
@app.post("/stop")
async def stop(wait_for: WaitFor):
"""
Stops the eden block, and exits the script
Args:
config (dict, optional): Amount of time in seconds before the server shuts down. Defaults to {'time': 0}.
"""
logging.info(f"Stopping gracefully in {wait_for.seconds} seconds")
stop_everything_gracefully(t=wait_for.seconds)
@app.post("/get_identity")
def get_identity():
"""
Returns name and active commit hash of the generator
"""
try:
repo = git.Repo(search_parent_directories=True)
name = repo.remotes.origin.url.split('.git')[0].split('/')[-1]
sha = repo.head.object.hexsha
except git.exc.InvalidGitRepositoryError:
name = "repo-less-eden"
sha = "none"
response = {
"name": name,
"commit": sha
}
return response
## overriding the boring old [INFO] thingy
LOGGING_CONFIG["formatters"]["default"]["fmt"] = (
"[" + Colors.CYAN + "EDEN" + Colors.END + "] %(asctime)s %(message)s"
)
LOGGING_CONFIG["formatters"]["access"]["fmt"] = (
"["
+ Colors.CYAN
+ "EDEN"
+ Colors.END
+ "] %(levelprefix)s %(client_addr)s - '%(request_line)s' %(status_code)s"
)
config = uvicorn.config.Config(app=app, host=host, port=port, log_level=log_level)
server = ThreadedServer(config=config)
# context starts fastAPI stuff and run_celery_app starts celery
with server.run_in_thread():
message = (
PREFIX
+ " Initializing celery worker on: "
+ f"redis://localhost:{str(redis_port)}"
)
print(message)
## starts celery app
run_celery_app(
celery_app,
max_num_workers=max_num_workers,
loglevel=celery_log_levels[log_level],
logfile=logfile,
queue_name=block.name,
)
message = PREFIX + " Stopped"
print(message)

75
example_block/eden-server/image_utils.py

@ -0,0 +1,75 @@
import PIL
import cv2
import base64
import numpy as np
from PIL.Image import Image as ImageFile
from PIL.JpegImagePlugin import JpegImageFile
from PIL.PngImagePlugin import PngImageFile
from PIL import Image
from io import BytesIO
def _encode_numpy_array_image(image):
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if image.shape[-1] == 3:
_, buffer = cv2.imencode(".jpg", image)
elif image.shape[-1] == 4:
_, buffer = cv2.imencode(".png", image)
image_as_text = base64.b64encode(buffer)
return image_as_text
def _encode_pil_image(image):
opencv_image = np.array(image)
image_as_text = _encode_numpy_array_image(image=opencv_image)
return image_as_text
def _encode_image_file(image):
pil_image = Image.open(image)
return _encode_pil_image(pil_image)
def encode(image):
if (
type(image) == np.ndarray
or type(image) == str
or isinstance(
image,
(
JpegImageFile,
PngImageFile,
ImageFile,
),
)
):
if type(image) == np.ndarray:
image_as_text = _encode_numpy_array_image(image)
elif type(image) == str:
image_as_text = _encode_image_file(image)
else:
image_as_text = _encode_pil_image(image)
return image_as_text.decode("ascii")
else:
raise Exception(
"expected numpy.array, PIL.Image or str, not: ", str(type(image))
)
def decode(jpg_as_text):
if jpg_as_text is None:
return None
pil_image = Image.open(BytesIO(base64.b64decode(jpg_as_text)))
return pil_image

69
example_block/eden-server/server copy.py

@ -0,0 +1,69 @@
from eden.block import Block
from eden.datatypes import Image
from eden.hosting import host_block
## eden <3 pytorch
from torchvision import models, transforms
import torch
model = models.resnet50(weights=models.ResNet50_Weights.DEFAULT)
model = model.eval() ## no dont move it to the gpu just yet :)
my_transforms = transforms.Compose(
[
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]), # this normalizes the image to the same format as the pretrained model
]
)
eden_block = Block()
my_args = {
"width": 224, ## width
"height": 224, ## height
"input_image": Image(), ## images require eden.datatypes.Image()
}
import requests
labels = requests.get(
"https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt"
).text.split("\n")
@eden_block.run(args=my_args, progress=False)
def do_something(config):
global model, labels
pil_image = config["input_image"]
pil_image = pil_image.resize((config["width"], config["height"]))
device = config.gpu
input_tensor = my_transforms(pil_image).to(device).unsqueeze(0)
model = model.to(device)
with torch.no_grad():
pred = model(input_tensor)[0].cpu()
index = torch.argmax(pred).item()
value = pred[index].item()
# the index is the classification label for the pretrained resnet18 model.
# the human-readable labels associated with this index are pulled and returned as "label"
# we need to get them from imagenet labels, which we need to get online.
label = labels[index]
# serialize the image
pil_image = Image(pil_image)
return {"value": value, "index": index, "label": label, 'image': pil_image}
if __name__ == "__main__":
host_block(
block=eden_block,
port=5656,
host="0.0.0.0",
redis_host="redis",
# logfile="log.log",
logfile=None,
log_level="debug",
max_num_workers=1,
requires_gpu=True,
)

123
example_block/eden-server/server.py

@ -0,0 +1,123 @@
from eden.block import Block
from eden.datatypes import Image
from eden.hosting import host_block
## eden <3 pytorch
from torchvision import models, transforms
import torch
model = models.resnet50(weights=models.ResNet50_Weights.DEFAULT)
model = model.eval() ## no dont move it to the gpu just yet :)
my_transforms = transforms.Compose(
[
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]), # this normalizes the image to the same format as the pretrained model
]
)
eden_block = Block()
my_args = {
"width": 224, ## width
"height": 224, ## height
"input_image": Image(), ## images require eden.datatypes.Image()
}
import requests
labels = requests.get(
"https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt"
).text.split("\n")
@eden_block.run(args=my_args, progress=False)
def do_something(config):
global model, labels
pil_image = config["input_image"]
pil_image = pil_image.resize((config["width"], config["height"]))
device = config.gpu
input_tensor = my_transforms(pil_image).to(device).unsqueeze(0)
model = model.to(device)
with torch.no_grad():
pred = model(input_tensor)[0].cpu()
index = torch.argmax(pred).item()
value = pred[index].item()
# the index is the classification label for the pretrained resnet18 model.
# the human-readable labels associated with this index are pulled and returned as "label"
# we need to get them from imagenet labels, which we need to get online.
label = labels[index]
# serialize the image
pil_image = Image(pil_image)
return {"value": value, "index": index, "label": label, 'image': pil_image}
def run_host_block():
host_block(
block=eden_block,
port=5656,
host="0.0.0.0",
redis_host="redis",
# logfile="log.log",
logfile=None,
log_level="debug",
max_num_workers=1,
requires_gpu=True,
)
import asyncio
import socketio
import socket
def get_ip_address():
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
try:
# This IP address doesn't need to be reachable, as we're only using it to find the local IP address
s.connect(("10.255.255.255", 1))
ip = s.getsockname()[0]
except Exception:
ip = "127.0.0.1"
finally:
s.close()
return ip
# Update these with the correct values for your host and server
HOST_SERVER_IP = "0.0.0.0"
HOST_SERVER_PORT = 4999
SERVER_NAME = "server_1"
SERVER_IP = get_ip_address()
SERVER_PORT = 8000
sio = socketio.AsyncClient()
async def announce_server():
await sio.connect(f'http://{HOST_SERVER_IP}:{HOST_SERVER_PORT}')
await sio.emit('register', {'name': SERVER_NAME, 'ip': SERVER_IP, 'port': SERVER_PORT})
@sio.on("heartbeat")
async def on_heartbeat():
print("Received heartbeat from host")
@sio.event
async def disconnect():
print("Disconnected from host")
async def main():
# Run host_block in a separate thread
loop = asyncio.get_event_loop()
host_block_thread = loop.run_in_executor(None, run_host_block)
# Announce the server to the host
await announce_server()
# Wait for host_block to finish
await host_block_thread
if __name__ == "__main__":
asyncio.run(main())

24
example_block/nginx.conf

@ -0,0 +1,24 @@
worker_processes 1;
events {
worker_connections 1024;
}
http {
upstream eden-servers {
server eden-server:5656;
}
server {
listen 80;
server_name _;
location / {
proxy_pass http://eden-servers;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
}
}
}

12
example_block/redis/Dockerfile

@ -0,0 +1,12 @@
# Use an official Redis image as a parent image
FROM redis:latest
# Set the working directory to /data
WORKDIR /data
# Expose Redis port
EXPOSE 6379
# Run Redis server as daemon
#CMD ["redis-server", "--daemonize", "yes"]
CMD ["redis-server", "--daemonize", "no"]

66
example_block/s2.py

@ -0,0 +1,66 @@
from eden.block import Block
from eden.datatypes import Image
from eden.hosting import host_block
## eden <3 pytorch
from torchvision import models, transforms
import torch
model = models.resnet50(weights=models.ResNet50_Weights.DEFAULT)
model = model.eval() ## no dont move it to the gpu just yet :)
my_transforms = transforms.Compose(
[
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]), # this normalizes the image to the same format as the pretrained model
]
)
eden_block = Block()
my_args = {
"width": 224, ## width
"height": 224, ## height
"input_image": Image(), ## images require eden.datatypes.Image()
}
import requests
labels = requests.get(
"https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt"
).text.split("\n")
@eden_block.run(args=my_args, progress=False)
def do_something(config):
global model, labels
pil_image = config["input_image"]
pil_image = pil_image.resize((config["width"], config["height"]))
device = config.gpu
input_tensor = my_transforms(pil_image).to(device).unsqueeze(0)
model = model.to(device)
with torch.no_grad():
pred = model(input_tensor)[0].cpu()
index = torch.argmax(pred).item()
value = pred[index].item()
# the index is the classification label for the pretrained resnet18 model.
# the human-readable labels associated with this index are pulled and returned as "label"
# we need to get them from imagenet labels, which we need to get online.
label = labels[index]
# serialize the image
pil_image = Image(pil_image)
return {"value": value, "index": index, "label": label, 'image': pil_image}
if __name__ == "__main__":
host_block(
block=eden_block,
port=5655,
logfile="log2.log",
log_level="debug",
max_num_workers=1,
requires_gpu=True,
)

66
example_block/server.py

@ -0,0 +1,66 @@
from eden.block import Block
from eden.datatypes import Image
from eden.hosting import host_block
## eden <3 pytorch
from torchvision import models, transforms
import torch
model = models.resnet50(weights=models.ResNet50_Weights.DEFAULT)
model = model.eval() ## no dont move it to the gpu just yet :)
my_transforms = transforms.Compose(
[
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]), # this normalizes the image to the same format as the pretrained model
]
)
eden_block = Block()
my_args = {
"width": 224, ## width
"height": 224, ## height
"input_image": Image(), ## images require eden.datatypes.Image()
}
import requests
labels = requests.get(
"https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt"
).text.split("\n")
@eden_block.run(args=my_args, progress=False)
def do_something(config):
global model, labels
pil_image = config["input_image"]
pil_image = pil_image.resize((config["width"], config["height"]))
device = config.gpu
input_tensor = my_transforms(pil_image).to(device).unsqueeze(0)
model = model.to(device)
with torch.no_grad():
pred = model(input_tensor)[0].cpu()
index = torch.argmax(pred).item()
value = pred[index].item()
# the index is the classification label for the pretrained resnet18 model.
# the human-readable labels associated with this index are pulled and returned as "label"
# we need to get them from imagenet labels, which we need to get online.
label = labels[index]
# serialize the image
pil_image = Image(pil_image)
return {"value": value, "index": index, "label": label, 'image': pil_image}
if __name__ == "__main__":
host_block(
block=eden_block,
port=5656,
logfile="logs.log",
log_level="debug",
max_num_workers=1,
requires_gpu=True,
)

75
image_utils.py

@ -0,0 +1,75 @@
import PIL
import cv2
import base64
import numpy as np
from PIL.Image import Image as ImageFile
from PIL.JpegImagePlugin import JpegImageFile
from PIL.PngImagePlugin import PngImageFile
from PIL import Image
from io import BytesIO
def _encode_numpy_array_image(image):
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if image.shape[-1] == 3:
_, buffer = cv2.imencode(".jpg", image)
elif image.shape[-1] == 4:
_, buffer = cv2.imencode(".png", image)
image_as_text = base64.b64encode(buffer)
return image_as_text
def _encode_pil_image(image):
opencv_image = np.array(image)
image_as_text = _encode_numpy_array_image(image=opencv_image)
return image_as_text
def _encode_image_file(image):
pil_image = Image.open(image)
return _encode_pil_image(pil_image)
def encode(image):
if (
type(image) == np.ndarray
or type(image) == str
or isinstance(
image,
(
JpegImageFile,
PngImageFile,
ImageFile,
),
)
):
if type(image) == np.ndarray:
image_as_text = _encode_numpy_array_image(image)
elif type(image) == str:
image_as_text = _encode_image_file(image)
else:
image_as_text = _encode_pil_image(image)
return image_as_text.decode("ascii")
else:
raise Exception(
"expected numpy.array, PIL.Image or str, not: ", str(type(image))
)
def decode(jpg_as_text):
if jpg_as_text is None:
return None
pil_image = Image.open(BytesIO(base64.b64decode(jpg_as_text)))
return pil_image

1
index.html

@ -126,6 +126,7 @@
method: "POST", method: "POST",
body: formData, body: formData,
credentials: "include", credentials: "include",
mode: "no-cors",
}); });
if (response.ok) { if (response.ok) {

77
registry/host.py

@ -0,0 +1,77 @@
import asyncio
import signal
import socketio
from aiohttp import web
SERVER_0_IP = "192.168.1.113"
FLASK_SERVER_PORT = 4999
HEARTBEAT_INTERVAL = 1
HEARTBEAT_TIMEOUT = 3
sio = socketio.AsyncServer(async_mode='aiohttp')
app = web.Application()
sio.attach(app)
servers = {}
async def available(request):
return web.json_response(servers)
app.router.add_get("/available", available)
@sio.event
async def connect(sid, environ):
print("I'm connected!", sid)
@sio.event
async def register(sid, data):
server_info = data
name = server_info["name"]
servers[name] = {"ip": server_info["ip"], "port": server_info["port"], "sid": sid}
print(servers)
@sio.event
async def disconnect(sid):
print("I'm disconnected!", sid)
for name, server in servers.items():
if server["sid"] == sid:
del servers[name]
break
async def heartbeat():
while True:
await asyncio.sleep(HEARTBEAT_INTERVAL)
server_values_copy = list(servers.values())
for server in server_values_copy:
sid = server["sid"]
try:
print(f"Sending heartbeat to {sid}...")
heartbeat_future = sio.emit("heartbeat", to=sid)
await asyncio.wait_for(heartbeat_future, timeout=HEARTBEAT_TIMEOUT)
except (asyncio.TimeoutError, socketio.exceptions.TimeoutError):
print(f"Server {sid} failed to respond to heartbeat.")
await sio.disconnect(sid)
def exit_handler(sig, frame):
print("Shutting down host...")
loop = asyncio.get_event_loop()
heartbeat_task.cancel()
loop.run_until_complete(loop.shutdown_asyncgens())
loop.stop()
if __name__ == "__main__":
signal.signal(signal.SIGINT, exit_handler)
signal.signal(signal.SIGTERM, exit_handler)
loop = asyncio.get_event_loop()
heartbeat_task = loop.create_task(heartbeat())
aiohttp_app = loop.create_task(web._run_app(app, host=SERVER_0_IP, port=FLASK_SERVER_PORT))
try:
loop.run_until_complete(asyncio.gather(heartbeat_task, aiohttp_app))
except asyncio.CancelledError:
pass
finally:
loop.run_until_complete(loop.shutdown_asyncgens())
loop.stop()

2
registry/requirements.txt

@ -0,0 +1,2 @@
python-socketio[asyncio_client]==6.1.1
aiohttp==3.8.1

39
registry/s1.py

@ -0,0 +1,39 @@
import signal
import socketio
SERVER_0_IP = "localhost"
SERVER_0_PORT = 4999
SERVER_1_PORT = 5001
SERVER_1_NAME = "server_1"
sio = socketio.Client()
@sio.event
def connect():
print("I'm connected!")
sio.emit("register", {"name": SERVER_1_NAME, "ip": SERVER_0_IP, "port": SERVER_1_PORT})
@sio.event
def connect_error(data):
print("The connection failed!")
@sio.event
def disconnect():
print("I'm disconnected!")
@sio.event
def heartbeat():
print("Received heartbeat")
def main():
sio.connect(f"http://{SERVER_0_IP}:{SERVER_0_PORT}")
sio.wait()
def exit_handler(sig, frame):
sio.disconnect()
exit(0)
if __name__ == "__main__":
signal.signal(signal.SIGINT, exit_handler)
signal.signal(signal.SIGTERM, exit_handler)
main()

39
registry/s2.py

@ -0,0 +1,39 @@
import signal
import socketio
SERVER_0_IP = "localhost"
SERVER_0_PORT = 4999
SERVER_1_PORT = 5002
SERVER_1_NAME = "server_2"
sio = socketio.Client()
@sio.event
def connect():
print("I'm connected!")
sio.emit("register", {"name": SERVER_1_NAME, "ip": SERVER_0_IP, "port": SERVER_1_PORT})
@sio.event
def connect_error(data):
print("The connection failed!")
@sio.event
def disconnect():
print("I'm disconnected!")
@sio.event
def heartbeat():
print("Received heartbeat")
def main():
sio.connect(f"http://{SERVER_0_IP}:{SERVER_0_PORT}")
sio.wait()
def exit_handler(sig, frame):
sio.disconnect()
exit(0)
if __name__ == "__main__":
signal.signal(signal.SIGINT, exit_handler)
signal.signal(signal.SIGTERM, exit_handler)
main()

1
requirements.txt

@ -4,3 +4,4 @@ psycopg2-binary
flask_bcrypt flask_bcrypt
# flask_login # flask_login
flask_cors flask_cors
eden-python
Loading…
Cancel
Save