I am currently deploying a Flask app to AWS Beanstalk, and I am trying to log out some stuff in the application (print), but I am not sure how to view it within Beanstalk, do guide me along, thank you!
The print should be in /var/log/web.output.log. However, they can show up with a delay. Thus, I found that its easier to hook up into gunicorn logger from flask which update web.output.log in real time.
Below is sample application.py that you can test it out and see how to set it up:
import os
from flask import Flask, Blueprint
from flask_restful import Api
from datetime import datetime
import logging
application = Flask(__name__)
#application.route("/")
def hello():
current_time = datetime.now().strftime("%H:%M:%S")
print(f"print from the app {current_time}")
application.logger.info(f"info from hello {current_time}")
application.logger.error(f"error from hello {current_time}")
return "<h1 style='color:blue'>Hello There!</h1>"
if __name__ == '__main__':
application.run()
else:
gunicorn_logger = logging.getLogger('gunicorn.error')
application.logger.handlers = gunicorn_logger.handlers
application.logger.setLevel(logging.INFO)
Related
I am not able to get http://127.0.0.1:5000/movie Url from the browser.
Every time it gives 404. The only time it worked was with URL from hello world.
I am trying to run a recommender system deployment solution with flask, based on https://medium.com/analytics-vidhya/build-a-movie-recommendation-flask-based-deployment-8e2970f1f5f1.
I have tried uninstall flask and install again but nothing seems to work.
Thank you!
from flask import Flask,request,jsonify
from flask_cors import CORS
import recommendation
app = Flask(__name__)
CORS(app)
#app.route('/movie', methods=['GET'])
def recommend_movies():
res = recommendation.results(request.args.get('title'))
return jsonify(res)
if __name__=='__main__':
app.run(port = 5000, debug = True)
http://127.0.0.1:5000/movie
be careful not writing '/' after movie like http://127.0.0.1:5000/movie/
I have created a Flask prediction app (within Google Colab) and when I am trying to run it post adding all the dependencies within the colab environment I am getting the url but when I click on it it show site cannot be reached.
I have the Procfile, the pickled model and the requirements text file but for some reason its not working. Also, I tried deploying this app using Heroku and it met the same fate where I got the app error.
For more context please visit my github repo.
Any help or guidance will be highly appreciated.
from flask import Flask, url_for, redirect, render_template, jsonify
from pycaret.classification import*
import pandas as pd
import numpy as np
import pickle
app = Flask(__name__)
model = load_model('Final RF Model 23JUL2021')
cols = ['AHT','NTT','Sentiment','Complaints','Repeats']
#app.route('/')
def home():
return render_template("home.html")
#app.route('/predict',methods=['POST'])
def predict():
int_features = [x for x in request.form.values()]
final = np.array(int_features)
data_unseen = pd.DataFrame([finak], columns = cols)
prediction = predict_model(model, data=data_unseen, round=0)
prediction = int(prediction.Label[0])
return render_template('home.html',pred='Predicted Maturiy Level is{}'.format(prediction))
#app.route('/predict_api',methods=['POST'])
def predict_api():
data = request.get_json(force=True)
data_unseen = pd.DataFrame([data])
prediction = predict_model(model, data=data_unseen)
output = prediction.Label[0]
return jsonify(output)
if __name__ == '__main__':
app.run(debug=True)
You cannot run flask app same as in your machine. You need to use flask-ngrok.
!pip install flask-ngrok
from flask_ngrok import run_with_ngrok
[...]
app = Flask(__name__)
run_with_ngrok(app)
[...]
app.run()
You can't use debug=True parameter in ngrok.
I am trying to deploy a machine learning model on AWS EC2 instance using Flask. These are sklearn's fitted Random Forest models that are pickled using joblib. When I host Flask on localhost and load them into memory everything runs smoothly. However, when I deploy it on the apache2 server using mod_wsgi, joblib works sometimes(i.e. the models are loaded using joblib sometimes) and the other times the server just hangs. There is no error in logs. Any ideas would be appreciated.
Here is the relevant code that I am using:
# In[49]:
from flask import Flask, jsonify, request, render_template
from datetime import datetime
from sklearn.externals import joblib
import pickle as pkl
import os
# In[50]:
app = Flask(__name__, template_folder="/home/ubuntu/flaskapp/")
# In[51]:
log = lambda msg: app.logger.info(msg, extra={'worker_id': "request.uuid" })
# Logger
import logging
handler = logging.FileHandler('/home/ubuntu/app.log')
handler.setLevel(logging.ERROR)
app.logger.addHandler(handler)
# In[52]:
#app.route('/')
def host_template():
return render_template('Static_GUI.html')
# In[53]:
def load_models(path):
model_arr = [0]*len(os.listdir(path))
for filename in os.listdir(path):
f = open(path+"/"+filename, 'rb')
model_arr[int(filename[2:])] = joblib.load(f)
print("Classifier ", filename[2:], " added.")
f.close()
return model_arr
# In[54]:
partition_limit = 30
# In[55]:
print("Dictionaries being loaded.")
dict_file_path = "/home/ubuntu/Dictionaries/VARR"
dictionaries = pkl.load(open(dict_file_path, "rb"))
print("Dictionaries Loaded.")
# In[56]:
print("Begin loading classifiers.")
model_path = "/home/ubuntu/RF_Models/"
classifier_arr = load_models(model_path)
print("Classifiers Loaded.")
if __name__ == '__main__':
log("/home/ubuntu/print.log")
print("Starting API")
app.run(debug=True)
I was stuck with this for quite sometime. Posting the answer in case someone runs into this problem. Using print statements and looking at logs I narrowed the problem down to joblib.load statement. I found this awesome blog: http://blog.rtwilson.com/how-to-fix-flask-wsgi-webapp-hanging-when-importing-a-module-such-as-numpy-or-matplotlib
The idea of using a global process group fixed the problem. That forced the use of main interpreter just as the top comment on that blog page mentions.
Trying to add flask-Ask to an existing flask website that uses the runserver pattern where app setup done in init but app.run is called in runserver
/myapp
/myapp
__init__.py
views.py
alexa_views.py
runserver.py
This pattern works fine for Flask ( its recommended for larger apps) but Flask-Ask is failing silently when app.run(debug=True) is called from runserver.py.
If I call app.run(debug=True) in _init__.py and run that then Flask-Ask works fine and Alexa responds.
Any ideas?
code:
alexa_views.py
from flask import blueprints
from flask_ask import Ask, statement
askblueprint = blueprints.Blueprint('alexa', __name__, url_prefix='/alexa')
ask = Ask(blueprint=askblueprint)
#ask.launch
def launch():
return statement (' it works')
init.py
from flask import Flask, blueprints
from myapp.alexa_views import askblueprint
app = Flask(__name__)
app.register_blueprint(askblueprint)
# lots of other unrelated configuration here - db etc
# running app here causes Flask-Ask to work!
# if __name__ == '__main__':
# app.run(debug=True)
# late import of views to break circular import
import myapp.views
runserver.py
# running this starts website normally but Flask-Ask does nothing
from myapp import app
if __name__ == '__main__':
app.run(debug=True)
I am going to close this.
The problem does exist in my real app but this simple example now works fine so I will have to dig deeper to find something I can demonstrate.
Bill
I have a simple flask application that works really well. I've been developing it separately from the main desktop application, I want to "plugin" the flask application into the main application. I also want to use cherrypy as the webserver as the default webserver that comes with flask is not production ready. I am not sure how to get both these to work together. My flask application code looks like this
from flask import Flask, render_template,send_from_directory
from scripts_data import test_data
from schedule_data import scheduledata
import os
app=Flask(__name__)
#app.route('/')
def index():
return render_template('index.html')
#app.route('/scripts')
def scripts():
test_data=t_data()
return render_template('scripts.html',data_scripts=test_data)
#app.route('/sch')
def schedules():
data_schedule=s_data()
return render_template('schedules.html',table_data=data_schedule)
if __name__=='__main__':
app.run(debug=True)
so obviously as I want to integrate into the main application I can't use app.run. Its not clear how to swap out the flask webserver for the Cherrypy Webserver
I have seen the following
from flask import Flask
import cherrypy
app = Flask(__name__)
app.debug = True
Class setup_webserver(object):
#app.route("/")
def hello():
return "Hello World!"
def run_server():
# Mount the WSGI callable object (app) on the root directory
cherrypy.tree.graft(app, '/')
# Set the configuration of the web server
cherrypy.config.update({
'engine.autoreload_on': True,
'log.screen': True,
'server.socket_port': 5000,
'server.socket_host': '0.0.0.0'
})
# Start the CherryPy WSGI web server
cherrypy.engine.start()
cherrypy.engine.block()
Class start_it_all(object)
import setup_webserver
setup_webserver.run_server()
But when I start the webserver and go to the site (0.0.0.0:5000) I get a 404?
I don't get the 404 when its just flask on its own. All I want to do is swap out the flask built-in webserver for the cherrpy webserver. I don't want to use cherrypy for anything else, as Flask will be the framework
Any suggestions? I'm on Windows and using python 2.7