def success():
DATE = request.args.get(
'DATE') # get the key/value pairs in the URL query string from /form
CRS_DEP_TIME = request.args.get('CRS_DEP_TIME')
UNIQUE_CARRIER = request.args.get('UNIQUE_CARRIER')
FL_NUM = request.args.get('FL_NUM')
ORIGIN = request.args.get('ORIGIN')
DEST = request.args.get('DEST')
df_input = preprocess_input(DATE, CRS_DEP_TIME, UNIQUE_CARRIER, FL_NUM,
ORIGIN, DEST, filename_metadata)
res = make_prediction(df_input, filename_pickle)
return res
def predictor(data,
saved_model,
labels=False,
interactive=False,
interactive_x='none'):
''' Function for making predictions on a saved model.
NOTE: 1) remember to use the same 'x' as with training
2) call the model by its name
'''
pred = make_prediction(data,
saved_model,
labels=labels,
interactive=interactive,
interactive_x=interactive_x)
return pred
def start():
to_predict = request.json
# print(to_predict)
pred = make_prediction(to_predict)
return jsonify({"Predicted Clusters": pred})
def result():
if request.method == 'POST':
form_dict = request.form
result = make_prediction(form_dict)
return render_template('result.html', result=form_dict, proba=result)
def predict(self):
self.predictions = prediction.make_prediction(self.emu_data.values, self.max_predictions)
self.mistakes = prediction.calculate_mistake(self.predictions,
self.all_squares.values[len(self.emu_data):len(self.emu_data) + self.max_predictions])
# Making dataset that consists of images of selected desks with plants
# copy_selected_plant_photos.copy_selected_plant('all_photos', 6)
# Importing images from dataset
# images = import_images('6_plant')
# Calculating areas of plants on the desk
# count_area.calculate_squares(images)
# Calculating areas of one selected plant on the desk
# count_area.calculate_single_plant(images, 2)
# Getting raw array with areas of one selected plant
# areas = count_area.read_areas_from_file(3, type='array')
# show_plots([areas])
# # Getting Pandas.Series with areas of one selected plant
# areas = count_area.read_areas_from_file(2)
# lag_plot(areas)
# plt.show()
series = Series.from_csv('2_plant_areas.csv', header=0)
# prediction.show_autocorrelation(series)
# prediction.show_baseline(series)
prediction.make_prediction(series)