def main():
if len(sys.argv) == 3:
database_filepath, model_filepath = sys.argv[1:]
print('Loading data...\n DATABASE: {}'.format(database_filepath))
X, Y, category_names = load_data(database_filepath)
X_train, X_test, Y_train, Y_test = train_test_split(X,
Y,
test_size=0.2)
print('Building model...')
model = model_pipeline_v2()
print('Training model...')
model.fit(X_train, Y_train)
print('Evaluating model...')
evaluate_model(model, X_test, Y_test, category_names)
print('Saving model...\n MODEL: {}'.format(model_filepath))
save_model(model, model_filepath)
from models.train_classifier import load_data
from models.train_classifier import load_model
from figures.figures import return_data_figures
from figures.figures import return_result_figures
# Handle user input
if len(sys.argv) == 3:
database_filepath, model_filepath = sys.argv[1:]
else:
database_filepath = '../data/disaster_response.db'
model_filepath = '../models/model.pickle'
app = Flask(__name__)
# Load training data
_, Y, category_names = load_data(database_filepath, TRAIN_TABLE_NAME)
# Load model
model = load_model(model_filepath)
# Load document-term matrix
X_term = sp.load_npz('../models/' + DOCUMENT_TERM_MATRIX_NAME)
# Index webpage displays cool visuals and receives user input text for model
@app.route('/')
@app.route('/index')
def index():
# Return plotly graphs for visualization
graphs = return_data_figures(X_term, Y, model)
from models.train_classifier import evaluate_model, load_data, tokenize
from sklearn.metrics import classification_report
from sklearn.model_selection import train_test_split
from pandas import DataFrame
import joblib
model = joblib.load("../models/classifier.pkl")
database_filepath = '../data/DisasterResponse.db'
X, Y, category_names = load_data(database_filepath)
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2)
Y_pred = DataFrame(data=model.predict(X_test.head(10)), columns=category_names)
evaluate_model(model, X_test.head(10), Y_test.head(10), category_names)
global _pipe
return _pipe.transform(np.array([text]))
def apply_model(x_transformed):
global _clf1, _clf2
labels_zuordnung_mlp = _clf1.classes_
y_pred_mlp_proba = _clf1.predict_proba(x_transformed)
y_pred_mlp_label = np.array(
[labels_zuordnung_mlp[np.argmax(t)] for t in y_pred_mlp_proba])
y_pred_moc_labels = np.array(_clf2.predict(x_transformed))
return y_pred_mlp_label, y_pred_moc_labels
# load data
DATA, Y, target = tc.load_data("./data/DisasterResponse.db")
Y = pd.DataFrame(Y)
Y.columns = target
# load model
load_model("./data/rodelbahn_model.pckl")
def show_example():
t = pd.Series(DATA).sample()
print(t.iloc[0])
print(Y.iloc[int(t.index[0])].to_dict())
# index webpage displays 2 visuals and receives user input text for model's prediction
@app.route('/')
from numpy import log
from numpy.random import random
from wordcloud import WordCloud, STOPWORDS
from models.train_classifier import load_data
####################################################################################################
# Initialization #
####################################################################################################
# Load External Dependencies #######################################################################
with open('models/model.pkl', 'rb') as f:
classifier = pickle.load(f)
X, y, cat_names = load_data('data/udacity.db')
cat_counts = {cat: y[:, inx].sum() for inx, cat in enumerate(cat_names)}
####################################################################################################
# Category Display #
####################################################################################################
def gen_cat_card(cat):
'''Generates an html element to display one of the categories which a message has
been tagged with'''
# Format for display
cat = cat.replace('_', ' ')
cat = cat.title()