def upload_file():
if flask.request.method == 'GET':
url = flask.request.args.get("url")
img = load_image_url(url)
tensor = transform_image(img)
length_digit, output_number = get_prediction(tensor)
else:
img_bytes = flask.request.files['file'].read()
tensor = transform_image(img_bytes)
length_digit, output_number = get_prediction(tensor)
# print(length_digit, output_number)
data = {'prediction': output_number, 'Length': str(length_digit)}
return flask.jsonify(data)
def testpredict():
if request.method == "POST":
file = request.files.get('file')
if file is None or file.filename == "":
return jsonify({"error": "no file"})
if not allowed_file(file.filename):
return jsonify({"error": "format not supported"})
img_bytes = file.read()
tensor = transform_image(img_bytes)
prediction = get_predictions(tensor)
data = {
'prediction': prediction.item(),
'class_name': str(prediction.item())
}
return jsonify(data)
# try:
# img_bytes = file.read()
# tensor = transform_image(img_bytes)
# prediction = get_predictions(tensor)
# data = {'prediction': prediction.item(), 'class_name' :str(prediction.item())}
# return jsonify(data)
# except:
# return jsonify({"error":"error during prediction"})
return jsonify({'result': 1})
def fileUpload():
target = os.path.join(UPLOAD_FOLDER, 'test_docs')
if not os.path.isdir(target):
os.mkdir(target)
logger.info("welcome to upload`")
file = request.files['file']
filename = secure_filename(file.filename)
destination = "/".join([target, filename])
file.save(destination)
#session['uploadFilePath']=destination
img = Image.open('./uploads/test_docs/img1.jpeg')
tensor = transform_image(img)
prediction = get_prediction(tensor)
predicted_idx = str(prediction.item())
print(predicted_idx)
data = {
'prediction': predicted_idx,
'class_name': class_index_name[predicted_idx]
}
response = make_response(jsonify(data))
# Add Access-Control-Allow-Origin header to allow cross-site request
response.headers[
'Access-Control-Allow-Origin'] = 'http://ec2-18-217-62-181.us-east-2.compute.amazonaws.com:3000'
return response
def predict():
if request.method == "POST":
file = request.files.get('file')
if file is None or file.filename == "":
return jsonify({'error': 'no file'})
if not allowed_file(file.filename):
return jsonify({'error': 'format not supported'})
try:
img_bytes = file.read()
tensor = transform_image(img_bytes)
prediction = get_prediction(tensor)
data = {
'prediction': prediction.item(),
'class_name': str(prediction.item())
}
return jsonify(data)
except:
return jsonify({'error': 'error during prediction'})
# 1. load image
# 2. image -> tensor
# 3. prediction
# 4. return json
return jsonify({'result': 1})
def predict():
if request.method == 'POST':
file = request.files.get('file')
if file is None or file.filename == "":
return jsonify({'error': 'no hay archivo'})
if not allowed_file(file.filename):
return jsonify({'error': 'formato no soportado'})
try:
img_bytes = file.read()
tensor = transform_image(img_bytes)
prediction = get_prediction(tensor)
data = {'prediction': prediction.item(), 'class_name': str(prediction.item())}
return jsonify(data)
except:
return jsonify({'error': 'error durante la predicción'})
def upload_pred():
if request.method == "POST":
image_file = request.values['canvas']
image = base64.b64decode(image_file)
im_file = io.BytesIO(image) # convert image to file-like object
img = PIL.Image.open(im_file)
filename = str(uuid.uuid4())
img.save(UPLOAD_FOLDER + filename + ".png", "PNG")
file = open(UPLOAD_FOLDER + filename + '.png', 'rb')
#file = open("eight.png", 'rb')
img_bytes = file.read()
tensor = transform_image(img_bytes)
pred = get_prediction(tensor)
return render_template('index.html', prediction=str(pred.item()))
return render_template('index.html')
def predict():
if request.method == 'POST':
file = request.files.get('image')
if file is None or file.filename == "":
return jsonify({'error': 'no file'})
if not allowed_file(file.filename):
return jsonify({'error': 'format not supported'})
try:
img_bytes = file.read()
image = io.BytesIO(img_bytes)
image = Image.open(image)
tensor = transform_image(image)
prediction = get_prediction(tensor)
return render_template('rezultat.html', rez=prediction)
except:
return jsonify({'error': 'error during prediction'})
def predict():
if request.method == 'POST':
print(request.files)
file = request.files.read()
if file is None or file.filename == "":
return jsonify({'error': 'no file'})
if not allowed_file(file.filename):
return jsonify({'error': 'format not supported'})
# try:
# print(file.filename)
print("reachd")
img_bytes = file.read()
tensor = transform_image(img_bytes)
length_digit, output_number = get_prediction(tensor)
print(length_digit, output_number)
data = {'prediction': output_number, 'Length': str(length_digit)}
return flask.render_template("index.html", data=data)
def upload_files():
uploaded_file = request.files['file']
filename = uploaded_file.filename
if filename != '':
uploaded_file.save(os.path.join(app.config['UPLOAD_PATH'], filename))
#img_bytes = uploaded_file.read()
img = Image.open('./uploads/img1.jpeg')
tensor = transform_image(img)
prediction = get_prediction(tensor)
predicted_idx = str(prediction.item())
print(predicted_idx)
data = {
'prediction': predicted_idx,
'class_name': class_index_name[predicted_idx]
}
return render_template('index.html',
file=uploaded_file.filename,
category=class_index_name[predicted_idx])
def predict():
if request.method == 'POST':
if 'file' not in request.files:
return redirect(request.url)
file = request.files.get('file')
if file is None or file.filename == "":
error = 'No file uploaded'
return render_template('index.html', messages=error)
if not allowed_file(file.filename):
error = 'Format Not Supported'
return render_template('index.html', messages=error)
try:
img_bytes = file.read()
tensor = transform_image(img_bytes)
labels = Predict(tensor)
return render_template('result.html', labels=labels)
except:
error = 'Prediction Error'
return render_template('index.html', messages=error)
return render_template('index.html')
def predict(image):
image = Image.fromarray((image[:, :, 0]).astype(np.uint8))
image = image.resize((28, 28))
tensor = transform_image(image)
prediction = get_prediction(tensor)
return prediction