def predict():
log = Log()
msg = __name__ + '.' + utils.get_function_caller() + ' -> enter'
log.print(msg)
# get data
json_data = request.get_json(force=True)
msg = 'json_data: ', json_data
log.print(msg)
user_id = json_data['user_id']
random_state = 42
input_df = pd.json_normalize(json_data)
# save json_data and input_df for debugging purpose, save using unique name
json_data_unique_filename = config.PATH_TO_DATASET + utils.get_unique_filename(
'json_data.json')
input_df_unique_filename = config.PATH_TO_DATASET + utils.get_unique_filename(
'input_df.csv')
with open(json_data_unique_filename, 'w') as outfile:
json.dump(json_data, outfile)
input_df = input_df[list(config.ACCEPTABLE_COLUMNS)]
input_df.to_csv(input_df_unique_filename, index=False)
prediction = Prediction(user_id, input_df_unique_filename, random_state)
predictions, labels = prediction.predict()
result = {'prediction': int(predictions[0]), 'label': str(labels[0])}
#dummy
# result = {'prediction': 1, 'label': 'Good Loan'}
output = {'result': result}
output = result
msg = __name__ + '.' + utils.get_function_caller() + ' -> exit'
log.print(msg)
return jsonify(results=output)
def test_get_unique_filename(self, time_):
# Setup
time_.return_value = 11111122222
# Execute
filename = get_unique_filename('abc.txt')
origin_filename = base64.b64decode(filename)\
.decode('utf-8')
# Verify
assert_equal(len(origin_filename), 7)
assert_equal(origin_filename, 'abc.txt')
def test(self,
model_path,
sample_path,
input_noise=None,
input_labels=None,
count=0):
sess, last_saved_step = self.restore_model(model_path)
with sess:
print("A sample is saved in {:s}".format(
utils.save_image(
utils.get_unique_filename(sample_path),
self.generate_images(sess, input_noise, input_labels,
count))))
def upload():
# check if the post request has the file part
if 'file' not in request.files or not request.files['file']:
file_not_supported()
file = request.files['file']
if file and allowed_file(file.filename):
# disable temp folder auto deletion
with tempfile.NamedTemporaryFile(delete=False) as t:
file.save(t)
return jsonify({
'success': True,
'file': get_unique_filename(t.name)
})
else:
file_not_supported()
def upload():
results = []
files = []
for file in request.files.getlist(config['upload_field']):
if file and file.filename and is_allowed_file(
file.filename, config['allowed_extensions']):
# filename = secure_filename(file.filename)
basename, ext = os.path.splitext(file.filename)
filename = get_unique_filename(ext)
filename = os.path.join(config['upload_folder'], filename)
file.save(filename)
files.append((basename, filename))
for basename, path in files:
result = check(basename, path, config['epubcheck_path'],
stdout_encoding)
results.append(result)
session['results'] = json.dumps(results, ensure_ascii=False)
# return redirect(url_for('validate', results=results))
return redirect(url_for('validate'))
def handleUpload():
if 'x_ray_image' not in request.files:
print('no file found in upload request')
return jsonify("{'message': 'no file in upload request'}")
file = request.files['x_ray_image']
if file.filename == '':
return jsonify({'message': 'No file name provided'})
file.filename = get_unique_filename() + '.' + file.filename.split('.')[-1]
if file and allowed_file_extensions(file.filename):
if not os.path.exists(os.getenv('UPLOAD_FOLDER')):
os.makedirs(os.getenv('UPLOAD_FOLDER'))
uploaded_file_path = os.path.join(os.getenv('UPLOAD_FOLDER'), file.filename)
file.save(uploaded_file_path)
prediction, all_predictions = perform_inference(uploaded_file_path)
s3_url = upload_to_s3(uploaded_file_path, str(prediction) + '__' + str(file.filename))
print('*****************')
print('*****************')
print('*****************')
print(prediction)
print('*****************')
print('*****************')
print('*****************')
return app.response_class(
response=json.dumps({"prediction": str(prediction), "file_url": str(s3_url), "all": all_predictions}),
status=200,
mimetype='application/json'
)
else:
return app.response_class(
response=json.dumps({"message": "invalid file provided"}),
status=500,
mimetype='application/json'
)
if dataset_path.exists():
pipeline = Pipeline(user_id=user_id, path_to_dataset=dataset_path, random_state=random_state, test_size=0.2)
pipeline.train()
# data dummy for prediction demo
# should be in json format as it will be fed into API, will be managed by controlled later on
json_data = pd.read_csv(dataset_path)[0:1].to_json(orient='records')[1:-1].replace('},{', '} {')
json_data = json.loads(json_data)
print('Predict:',json_data)
# controller will convert json data to dataframe
input_df = pd.json_normalize(json_data)
# save json_data and input_df for debugging purpose, save using unique name
json_data_unique_filename = config.PATH_TO_DATASET+utils.get_unique_filename('json_data.json')
input_df_unique_filename = config.PATH_TO_DATASET+utils.get_unique_filename('input_df.csv')
with open(json_data_unique_filename, 'w') as outfile:
json.dump(json_data, outfile)
input_df.to_csv(input_df_unique_filename, index=False)
prediction = Prediction(user_id, input_df_unique_filename, random_state)
predictions, labels = prediction.predict()
print('predictions:',predictions)
print('prediction labels:',labels)
print('ground truth:',json_data[config.TARGET_COLUMN])