def prediction():
if request.method == 'POST':
data = [int(x) for x in request.form.values()]
data = np.array([data])
print(data)
data_df = pd.DataFrame(
data=data, columns=["YearBuilt", "BedroomAbvGr", "KitchenAbvGr"])
# data_df = pd.DataFrame()
# data_df["YearBuilt"] = data[0]
# data_df["BedroomAbvGr"] = data[1]
# data_df["KitchenAbvGr"] = data[2]
print("data frame after assigning values", data_df.head())
# data preprocessing
data_df = data_preprocessing(data_df)
data_df["SalePrice"] = 0
# feature extractor
features, label = feature_extractor(data_df)
model = read_model("output/model.pkl")
label = predict_price(features, model)
print(label)
return render_template(
'houseprice.html',
predicted_house_price='House Price should be $ {}'.format(
label[0][0]))
else:
return render_template('houseprice.html')
message = "defining stubs"
print(message)
# read data
data = pd.read_csv(dataset_path)
print(data.shape)
print(data.head(5))
# data selection
data = data_selection(data, selected_variables)
print(data.shape)
print(data.head(5))
# data preprocessing
data = data_preprocessing(data)
print(data.shape)
print(data.head(5))
data[target_varibale[0]] = 0
print(data.shape)
print(data.head(5))
# feature extractor
featueres, label = feature_extractor(data)
print(featueres.shape)
print(featueres.head(5))
print(label.shape)
print(label.head(5))
model = read_model(model_path)
predicted_labels = predict_price(featueres, model)
print(predicted_labels)