https://qiita.com/maitake9116/items/df733df5e608f8616d14 Let's incorporate the Sheltie judgment AI created in step into Flask.
It was implemented on ubuntu built with Vagrant. The IP is 192.168.33.31.
~/python/api
├─app.py
├─ai.py
├─model :Pre-made model
│ ├─model_predict.json
│ └─model_predict.hdf5
├─static
│ ├─favicon.ico :Favicon (optional)
│ └─style.css :Style sheet
├─templates :Screen template
│ └─index.html
├─images :Directory for storing uploaded images
└─uwsgi.ini :uWSGI configuration file
$ mkdir -p ~/python/api
$ cd ~/python/api
$ python3 -m venv api
$ source api/bin/activate
#Install the libraries needed to run ai
$ pip install werkzeug
$ pip install numpy==1.16.2
$ pip install sklearn
$ pip install tensorflow
$ pip install keras
$ pip install matplotlib
$ pip install pillow
When the file is uploaded, the uploaded image and the judgment result are displayed. I loaded bootstrap and decorated it a little.
{% extends "layout.html" %}
{% block content %}
<div class="container">
<div class="m-5">
<H1>Sheltie judgment AI</H1>
</div>
<div class="m-5">
<p>Determine if it is a sheltie. Please upload the image.</p>
<form action="/upload" method="post" enctype="multipart/form-data">
<input type="file" id=image name="image" accept="image/png, image/jpeg">
<button class="button" type="submit">submit</button>
</form>
</div>
{% if img_url %}
<div class="m-5 box">
<h3>image</h3>
<img src="{{ img_url }}">
<h3>judgment result</h3>
<p>{{ judgment }}</p>
</div>
{% endif %}
</div>
{% endblock %}
<!DOCTYPE html>
<html lang="ja">
<head>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" href="https://stackpath.bootstrapcdn.com/bootstrap/4.3.1/css/bootstrap.min.css" integrity="sha384-ggOyR0iXCbMQv3Xipma34MD+dH/1fQ784/j6cY/iJTQUOhcWr7x9JvoRxT2MZw1T" crossorigin="anonymous">
<script src="https://code.jquery.com/jquery-3.3.1.slim.min.js" integrity="sha384-q8i/X+965DzO0rT7abK41JStQIAqVgRVzpbzo5smXKp4YfRvH+8abtTE1Pi6jizo" crossorigin="anonymous"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/popper.js/1.14.7/umd/popper.min.js" integrity="sha384-UO2eT0CpHqdSJQ6hJty5KVphtPhzWj9WO1clHTMGa3JDZwrnQq4sF86dIHNDz0W1" crossorigin="anonymous"></script>
<script src="https://stackpath.bootstrapcdn.com/bootstrap/4.3.1/js/bootstrap.min.js" integrity="sha384-JjSmVgyd0p3pXB1rRibZUAYoIIy6OrQ6VrjIEaFf/nJGzIxFDsf4x0xIM+B07jRM" crossorigin="anonymous"></script>
<link rel=stylesheet type=text/css href="{{ url_for('static', filename='style.css') }}">
<link rel="shortcut icon" href="{{ url_for('.static',filename='favicon.ico') }}" />
<title>Sheltie judgment AI</title>
</head>
<body>
{% block content %}{% endblock %}
</body>
</html>
h1 {
background: #dfefff;
box-shadow: 0px 0px 0px 5px #dfefff;
border: dashed 1px #96c2fe;
padding: 0.2em 0.5em;
color: #454545;
}
.box {
padding: 0.5em 1em;
margin: 2em 0;
border: double 5px #4ec4d3;
}
When the image is uploaded by app.py, save the image and pass the file path to ai.py to get the judgment result.
# -*- coding: utf-8 -*-
from flask import Flask, render_template, request, redirect, url_for, send_from_directory
from werkzeug.utils import secure_filename
import numpy as np
import os
import ai
app = Flask(__name__)
SAVE_DIR = "./images"
app = Flask(__name__, static_url_path="")
@app.route('/')
def index():
return render_template('index.html')
@app.route('/images/<path:path>')
def send_js(path):
return send_from_directory(SAVE_DIR, path)
@app.route('/upload', methods=['POST'])
def upload():
img_file = request.files['image']
if img_file:
file_name = secure_filename(img_file.filename)
file_path = os.path.join(SAVE_DIR, file_name)
img_file.save(file_path)
judgment = ai.predict(file_path)
return render_template('index.html', img_url=file_path, judgment=judgment)
# -*- coding: utf-8 -*-
from keras import models
from keras.models import model_from_json
from keras.preprocessing import image
import numpy as np
import sys
import os
from keras.preprocessing.image import ImageDataGenerator, load_img, img_to_array, array_to_img
def predict(img_path: str):
#Model save destination
MODEL_ROOT_DIR = './model/'
MODEL_PATH = os.path.join(MODEL_ROOT_DIR, 'model_predict.json')
WEIGHT_PATH = os.path.join(MODEL_ROOT_DIR, 'model_predict.hdf5')
#category
CATEGORIES = [u'Sheltie', u'Corgi', u'Border collie']
#Image size
IMG_SIZE = 150
INPUT_SHAPE = (IMG_SIZE, IMG_SIZE,3)
#Load the model
model = model_from_json(open(MODEL_PATH).read())
model.load_weights(WEIGHT_PATH)
#Read image from input argument
img = image.load_img(img_path, target_size=INPUT_SHAPE)
x = image.img_to_array(img)
x = np.expand_dims(x, axis=0)
#Predict with a model
features = model.predict(x)
if features[0, 0] == 1:
return u'Sheltie.'
else:
for i in range(0, len(CATEGORIES)):
if features[0, i] == 1:
return u'It doesn't seem to be sheltie.{}is.'.format(CATEGORIES[i])
#Start Nginx
$ sudo systemctl start nginx
$ vi uwsgi.ini
#Start uWSGI
$ uwsgi uwsgi.ini
[uwsgi]
wsgi-file=app.py
callable=app
http=0.0.0.0:8000
socket=/tmp/uwsgi.sock
chmod-socket=666
I accessed http://192.168.33.31 and confirmed the operation.
It was easy to integrate AI into your web app.