def predict_and_plot(self):
print("PREDICTING")
self.df_new.head()
# this model will apply repeatedly until the positiveness value reaches to 100.
# this while loop will do that and calculate the number of iteration required
# then it will decide the number of days(decide according to the iteration count) needed to job fit
while (self.value <= 100):
self.days = self.days + 7
self.future = self.m.make_future_dataframe(periods=self.days)
self.future.tail()
forecast = self.m.predict(self.future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()
f_cast = forecast['yhat'].tail(1)
f_cast = f_cast.tolist()
self.value = f_cast[0]
print("Days: {}--- Positive value:{}".format(
self.days, self.value))
self.m.plot(forecast) #plotting the data
self.m.plot_components(forecast) #plotting the data
py.init_notebook_mode() # initialize the notebook mode
fig = plot_plotly(self.m, forecast) # This returns a plotly Figure
py.iplot(fig)
print("Number of days to fit the job: {} Days".format(self.days))
self.remove_temp_files()
def predict():
pytrend = TrendReq(timeout=(10,25), hl='en-US', tz=360)
pytrend.build_payload(kw_list=keywords, timeframe='today 5-y', geo=c_code)
five_years = pytrend.interest_over_time().reset_index()
if five_years[keywords].sum().values[0] < 5:
return 'not enough data', 'not enough data', 'not enough data', 'not enough data', 'not enough data'
else:
#five_years = five_years[five_years.isPartial == 'False']
five_years = five_years[:-1]
five_years.drop('isPartial', axis=1, inplace=True)
five_years.columns = ['ds', 'y']
temp = five_years.copy()
temp['cap'] = temp['y'].max()
temp['floor'] = temp['y'].min()
m = Prophet()
fit = m.fit(temp)
periods = 90
future = m.make_future_dataframe(periods=periods)
future["floor"] = temp['y'].min()
future["cap"] = temp["y"].max()
forecast = m.predict(future)
fig = plot_plotly(m, forecast)
fig.update_layout(template='plotly_white',title='{} "{}" Interest Prediction'.format(country,keywords[0]), xaxis={'title':'Date'}, yaxis={'title':'Intrest & Prediction'})
fig.update_traces(line_color="#F66162",
selector=dict(line_color='#0072B2'))
fig.update_traces(marker_color="royalblue",
selector=dict(marker_color='black'))
comp = plot_components_plotly(m, forecast)
comp.update_layout(template='plotly_white',title='{} "{}" Seasonality Prediction'.format(country,keywords[0]),yaxis={'title':'Prediction'})
comp.update_traces(line_color="#F66162",
selector=dict(line_color='#0072B2'))
comp.update_traces(marker_color="royalblue",
selector=dict(marker_color='black'))
return fig, comp, forecast, periods, pytrend
def do_prophet():
n_days = 7
# Predict forecast with Prophet.
df_train = df[['Date','Close']]
df_train = df_train.rename(columns={"Date": "ds", "Close": "y"})
m = Prophet()
m.fit(df_train)
future = m.make_future_dataframe(periods=n_days)
forecast = m.predict(future)
# Show and plot forecast
st.subheader('Forecast data')
st.write(forecast.tail())
st.subheader(f'시세 예측 그래프: {n_days} 일')
test = m.plot(forecast)
# st.write(test)
fig1 = plot_plotly(m, forecast)
# st.write('그렸어야해..')
# st.plotly_chart(forecast)
st.plotly_chart(fig1)
# st.write("Forecast components")
# fig2 = m.plot_components(forecast)
# st.write(fig2)
st.success("시세 예측 완료!")
st.balloons()
def plot_forecast(
self,
title="Forecast",
template="plotly_white",
yaxis_title="Number of individuals",
output_image=False,
width=800,
height=600,
scale=2,
output_figure=False,
):
fig = plot_plotly(self.m, self.forecast)
fig.update_layout(
title=title,
yaxis_title=yaxis_title,
template=template,
title_x=0.5,
)
if output_image:
return Image(
fig.to_image(format="png",
width=width,
height=height,
scale=scale))
if output_figure:
return fig
return fig.show()
def run_forecast(category, state, seasonality, periods, uncertainty_samples,
years):
t0 = time.time()
col = f"('{category}', '{state}')"
df = walmart[["date", col]].rename(columns={"date": "ds", col: "y"})
m, forecast = run_prophet(
df,
weekly_seasonality=seasonality,
periods=periods,
uncertainty_samples=uncertainty_samples,
start=years[0],
end=years[1],
)
# Plot figures
margin = dict(l=30, r=30, t=80, b=30)
fig_forecast = plot_plotly(m, forecast).update_layout(
title="Forecasting with Prophet",
width=None,
height=None,
margin=margin,
)
fig_components = plot_components_plotly(m, forecast).update_layout(
title="Seasonal Components", width=None, height=None, margin=margin)
t1 = time.time()
print(f"Training and Inference time: {t1-t0:.2f}s.")
return fig_forecast, fig_components
def indiadeaths():
df = pd.read_csv("Indiajuly21.csv", parse_dates=['Date'], dayfirst=True)
df = df[['Date', 'State/UnionTerritory', 'Cured', 'Deaths', 'Confirmed']]
df.columns = ['date', 'state', 'cured', 'deaths', 'confirmed']
today = df[df.date == '2020-07-24']
df2 = df.groupby(['date'])['confirmed', 'deaths',
'cured', ].sum().reset_index()
#making columns for daily new cases
df2['new_confirmed'] = df2.confirmed.diff()
df2['new_deaths'] = df2.deaths.diff()
df2['new_cured'] = df2.cured.diff()
#taking dates from 15th March
df2 = df2.iloc[64:]
#Statewise new confirmed
df['new_confirmed'] = df.groupby(['state'])['confirmed'].diff()
df['new_deaths'] = df.groupby(['state'])['deaths'].diff()
df['new_cured'] = df.groupby(['state'])['cured'].diff()
df3 = df2[['date', 'deaths']]
#Renaming column names according to fb prophet
df3.columns = ['ds', 'y']
m = Prophet()
m.fit(df3)
future = m.make_future_dataframe(periods=10)
forecast = m.predict(future)
fig = plot_plotly(m, forecast) # This returns a plotly Figure
fig.update_layout(
title_text='<b>Covid-19 Total deaths cases Forecast<b>',
title_x=0.5,
paper_bgcolor='LightCoral',
plot_bgcolor="LightCoral",
)
fig.show()
return render_template("indiastatistics.html")
def update_predictions(ticker):
dict_timelines = portfoliomanager.download_stock_timelines([ticker])
df_timeline_with_future, m = portfoliomanager.predict_future(dict_timelines[ticker])
fig_future = plot_plotly(m, df_timeline_with_future)
fig_components = plot_components_plotly(m, df_timeline_with_future)
return fig_future, fig_components
def show_graph(self, model, forecast):
'''
model = Prophet object
forecast = Dataframe. Result of the model.predict(future)
'''
fig = plot_plotly(model, forecast)
py.iplot(fig)
def plotter_prophet(model, frcst):
fig = plot_plotly(model,
frcst,
figsize=(900, 600),
xlabel='Time period',
ylabel='Price')
fig.update_layout(hovermode='x')
fig.show()
def prediction(self, period):
self.model.fit(self.prophet_df)
future = self.model.make_future_dataframe(periods=period)
# Eliminate weekend from future dataframe
future['day'] = future['ds'].dt.weekday
future = future[future['day'] <= 4]
forecast = self.model.predict(future)
fig = plot_plotly(self.model, forecast)
fig2 = self.model.plot_components(forecast)
fig3 = self.model.plot(forecast)
a = add_changepoints_to_plot(fig3.gca(), self.model, forecast)
return (fig, fig2, fig3)
def prophet_prediction(row, zip_code, retirement_date='2029'):
# if os.path.exists('pickles/{}_forecast.pkl'.format(zip_code)):
# with open("pickles/{}_model.pkl".format(zip_code), 'rb') as f:
# unpickler = pickle.Unpickler(f)
# m = unpickler.load()
m = Prophet(seasonality_mode='multiplicative')
row = row[(row['ds'] > '2009')]
m.fit(row)
forecast = pd.read_pickle('pickles/{}_forecast.pkl'.format(zip_code))
print(forecast.dtypes)
forecast = forecast[(forecast['ds'] > '2009')
& (forecast['ds'] < str(retirement_date))]
return plot_plotly(m, forecast)
def model_from_coin(coin, base, growth, changepoints, changepoint_range,
seasonality_mode, seasonality_prior_scale,
holidays_prior_scale, changepoint_prior_scale,
interval_width):
df = pd.DataFrame()
df["ds"] = base["DATE_INFO"]
df["y"] = base["CLOSE"]
if growth == 'logistic':
df["cap"] = float(max(base["CLOSE"]))
m = Prophet(growth=growth,
changepoints=changepoints,
changepoint_range=changepoint_range,
yearly_seasonality=False,
weekly_seasonality=False,
daily_seasonality=False,
seasonality_mode=seasonality_mode,
seasonality_prior_scale=seasonality_prior_scale,
holidays_prior_scale=holidays_prior_scale,
changepoint_prior_scale=changepoint_prior_scale,
interval_width=interval_width).add_country_holidays(
country_name='US').add_seasonality(
name='bimonthly', period=365.25 / 6,
fourier_order=5).add_seasonality(name='yearly',
period=365.25,
fourier_order=3)
m.fit(df)
future = m.make_future_dataframe(periods=int(365 / 2))
if growth == 'logistic':
future["cap"] = float(max(base["CLOSE"]))
forecast = m.predict(future)
logging.info("Going to detect_anomalies")
forecasted = detect_anomalies(forecast, df)
logging.info("Saving graph for {}".format(coin))
fig = plot_plotly(m, forecast) # This returns a plotly Figure
#plio.write_html(fig, '/appdata/graph_{}.html'.format(coin), include_plotlyjs=True)
plot_anomalies(forecasted).save('/appdata/graph_{}.html'.format(coin))
logging.info("Graph for {} saved".format(coin))
now = datetime.datetime.now()
resultdf = forecast[forecast['ds'] == pd.Timestamp(
now.year, now.month, now.day)][[
'ds', 'yhat', 'yhat_lower', 'yhat_upper'
]]
return resultdf
def predict(df, days):
prophet = Prophet()
prophet.fit(df)
future = prophet.make_future_dataframe(periods=days)
forecast = prophet.predict(future)
df_forecast = forecast[["ds", "yhat", "yhat_lower",
"yhat_upper"]].set_index("ds")
df_forecast = df_forecast.round(0)
# fig_forecast = prophet.plot(forecast)
fig_forecast = plot_plotly(prophet, forecast)
return df_forecast
def main_gui(self):
st.image('cl.jpg')
st.title('COVID-19 LUXEMBOURG')
st.subheader('Last 5 days')
st.table(self.data.tail())
self.plot_data('Cases', 'Total Cases')
self.plot_data('Deaths', 'Total Deaths')
self.plot_data('Recovered', 'Total Recoveries')
st.subheader('Future Cases (Predictions)')
fig = plot_plotly(self.model, self.prediction)
fig.update_layout(width = 700, height = 500)
st.plotly_chart(fig)
st.text('Source: https://github.com/CSSEGISandData/COVID-19')
def predict():
temp = p_df.copy()
temp['cap'] = temp['y'].max()
temp['floor'] = temp['y'].min()
m = Prophet()
fit = m.fit(temp)
periods = 90
future = m.make_future_dataframe(periods=periods)
future["floor"] = temp['y'].min()
future["cap"] = temp["y"].max()
forecast = m.predict(future)
fig = plot_plotly(m, forecast)
fig.update_layout(
title='{} CBD Oil Intrest Prediction'.format(country),
xaxis={'title': 'Date'},
yaxis={'title': 'Intrest Prediction'})
return fig, forecast, periods
def fit_predict_prophet(df1, city):
"""Subset for city column and rename Passengers columns"""
df1 = df1[df1['city'] == city]
df1 = df1.rename(columns={'PASSENGERS': 'y'})
size = int(len(df1) * 0.7)
train, test = df1[['ds','y']][0:size], df1[['ds','y']][size:len(df1)]
# train_exog, test_exog = df1[0:size], df1[size:len(df1)]
"""Reset Index for both training and testing set"""
train.reset_index(inplace=True)
test.reset_index(inplace=True)
start=len(train)
end=len(train)+len(test)-1
# endog = train_exog['y']
# exog = train_exog[['temperature_mean']]
"""Prophet Modeling and Fit"""
Model = pr(daily_seasonality = False, yearly_seasonality = True, weekly_seasonality = False,
seasonality_mode = 'multiplicative',
interval_width = 0.90,
changepoint_range = 0.8)
Model.fit(train)
"""Forecast and Plot Model"""
forecast = Model.predict(pd.DataFrame(df1.reset_index()['ds'][start-1:end]))
forecast.head()
Model.plot(forecast, uncertainty=True)
"""Plot Fbprophet Using plotly"""
from fbprophet.plot import plot_plotly
import plotly.offline as py
py.init_notebook_mode()
"""This returns a plotly Figure"""
fig = plot_plotly(Model, forecast)
"""Calculate Root Mean Squared"""
rms = sqrt(mean_squared_error(test['y'], forecast.loc[:, 'yhat']))
print(rms)
return fig
def create_forecasts(value):
#dataframe for fbprophet prediction/visualization
fb_df = df[df['Geographic_Area'] == 'Total'][['Date', value]]
fb_df.columns = ['ds', 'y']
#removing any non-numerical characters
fb_df['y'] = fb_df['y'].apply(lambda x: int(re.sub('[^0-9]', '', str(x))))
#removing weekends
fb_df_weekdays = fb_df[fb_df['ds'].dt.dayofweek < 5]
#fitting the fbprophet model
prophet = Prophet()
prophet.fit(fb_df_weekdays)
#making dataframe of future dates, 30 days into future
future = prophet.make_future_dataframe(periods=30)
future_weekdays = future[future['ds'].dt.dayofweek < 5]
#generating forecast for future dates
forecast = prophet.predict(future_weekdays)
#dictionary of title options based on radio items feature to be used in app callbacks for fb prophet section
title_options = {
'Daily Cases': 'New Cases Per Day',
'Total\xa0 Confirmed Cases': 'Total Confirmed Cases',
'Recovered by Region': 'Recovered Patients',
'Still infectious by Region': 'Active Cases',
'Number of Deaths': 'Number of Deaths'
}
#plotting forecast
fig1 = plot_plotly(prophet, forecast)
fig1.update_layout(paper_bgcolor=colors['background'],
plot_bgcolor=colors['background'],
font={'color': colors['text']},
xaxis={'title': 'Date'},
yaxis={'title': title_options[value]},
width=1000)
fig1.update_traces(marker={'color': 'LightSkyBlue'})
return fig1
def plot_fbprophet_forecasting(df, dates, days_in_future, feature):
k = df[df['Country/Region'] == 'India'].loc[:, '1/22/20':]
india_values = k.values.tolist()[0]
data = pd.DataFrame(columns=['ds', 'y'])
data['ds'] = dates
data['y'] = india_values
pd.set_option('float_format', '{:f}'.format)
model_prophet = fbprophet.Prophet()
model_prophet.fit(data)
future = model_prophet.make_future_dataframe(periods=days_in_future)
prop_forecast = model_prophet.predict(future)
forecast = prop_forecast[['ds', 'yhat', 'yhat_lower',
'yhat_upper']].tail(days_in_future)
print(forecast)
fig = plot_plotly(model_prophet, prop_forecast)
fig = model_prophet.plot(prop_forecast,
xlabel='Date',
ylabel='{}'.format(feature))
fig.show()
def make_forecast(selection):
"""Takes a name from the selection and makes a forecast plot."""
if selection == CASES:
cumulative_series_name = "cumulative_cases"
title = "Daily Cases"
x_label = "Cases"
if selection == DEATHS:
cumulative_series_name = "cumulative_deaths"
title = "Daily Deaths"
x_label = "Deaths"
if selection == RECOVERIES:
cumulative_series_name = "cumulative_recoveries"
title = "Daily Recoveries"
x_label = "Recoveries"
prophet_df = (
df[cumulative_series_name]
.diff()
.dropna()
.to_frame()
.reset_index()
.rename(columns={"date": "ds", cumulative_series_name: "y"})
)
model = Prophet()
model.fit(prophet_df)
future = model.make_future_dataframe(periods=90)
forecast = model.predict(future)
fig = plot_plotly(model, forecast)
fig.update_layout(
title=title, yaxis_title=x_label, xaxis_title="Date",
)
return fig
def make_forecast(selection):
"""Takes a name from the selection and makes a forecast plot."""
if selection == PRODUCT_A:
cumulative_series_name = "cumulative_cases"
title = "Forecast Product A"
x_label = "Prices"
if selection == PRODUCT_B:
cumulative_series_name = "cumulative_deaths"
title = "Forecast Product B"
x_label = "Prices"
if selection == PRODUCT_C:
cumulative_series_name = "cumulative_recoveries"
title = "Forecast Product C"
x_label = "Prices"
prophet_df = (df[cumulative_series_name].diff().dropna().to_frame().
reset_index().rename(columns={
"date": "ds",
cumulative_series_name: "y"
}))
model = Prophet()
model.fit(prophet_df)
future = model.make_future_dataframe(periods=90)
forecast = model.predict(future)
fig = plot_plotly(model, forecast)
fig.update_layout(
title=title,
yaxis_title=x_label,
xaxis_title="Date",
)
return fig
def predict_n_days(required_data, start_date, end_date, company_symbol):
number_of_days = input(
'How many days in the future you want to predict? : ')
required_data[['ds', 'y']] = required_data[['Date', 'Close']]
df = required_data[['ds', 'y']]
m = Prophet(weekly_seasonality=False,
yearly_seasonality=True,
daily_seasonality=True,
n_changepoints=15)
m.fit(df)
future = m.make_future_dataframe(periods=int(number_of_days))
forecast = m.predict(future)
fig = plot_plotly(
m,
forecast,
xlabel=
f'Trained model from {start_date} to {end_date} and predicted {number_of_days} days in the future',
ylabel=f'Closing price of {company_symbol} ',
figsize=(1800, 800))
fig.update_layout(showlegend=True)
plot(fig,
filename=f'{company_symbol}_predict_{number_of_days}_days_graph.html')
def karcured():
df = pd.read_csv("onlykarnatakajune3.csv",
parse_dates=['Date'],
dayfirst=True)
df = df[['Date', 'State/UnionTerritory', 'Cured', 'Deaths', 'Confirmed']]
df.columns = ['date', 'state', 'cured', 'deaths', 'confirmed']
#df.isna().sum()
today = df[df.date == '2020-07-24']
df2 = df.groupby(['date'])['confirmed', 'deaths',
'cured', ].sum().reset_index()
#making columns for daily new cases
df2['new_confirmed'] = df2.confirmed.diff()
df2['new_deaths'] = df2.deaths.diff()
df2['new_cured'] = df2.cured.diff()
#taking dates from 15th March
df2 = df2.iloc[122:]
#Statewise new confirmed
df['new_confirmed'] = df.groupby(['state'])['confirmed'].diff()
df['new_deaths'] = df.groupby(['state'])['deaths'].diff()
df['new_cured'] = df.groupby(['state'])['cured'].diff()
df3 = df2[['date', 'cured']]
#Renaming column names according to fb prophet
df3.columns = ['ds', 'y']
m = Prophet()
m.fit(df3)
future = m.make_future_dataframe(periods=10)
forecast = m.predict(future)
#forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail(31)
fig = plot_plotly(m, forecast) # This returns a plotly Figure
fig.update_layout(
title_text='<b>Covid-19 Total Recovery cases Forecast in karnataka<b>',
title_x=0.5,
paper_bgcolor='LightGreen',
plot_bgcolor="LightGreen",
)
fig.show()
return render_template("karnataka.html")
def iterate():
folder_path = "../convert_data/idle_res_us_timestamp"
for filename in os.listdir(folder_path):
if filename.endswith(".csv"):
df = pd.read_csv(folder_path + "/" + filename,
names=[
"ds", "bandwidth_up", "bandwidth_down",
"connections", "ports", "ips"
])
df['ds'] = df['ds'].apply(
lambda x: time.strftime('%Y-%m-%d %H:%M:%S', time.gmtime(x)))
df.head()
m = Prophet(changepoint_prior_scale=0.01, interval_width=0.95)
for col in df.columns:
if col != "ds":
subdf = df[['ds', col]].dropna()
subdf = subdf.rename(columns={'ds': 'ds', col: 'y'})
m.fit(subdf)
break
future = m.make_future_dataframe(periods=1)
future.tail()
forecast = m.predict(future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()
m.plot_components(forecast).savefig(
filename.split('.')[0] + '.png') #
fig = plot_plotly(m, forecast) # This returns a plotly Figure
file_name_html = filename.split('.')[0] + ".html"
py.plot(fig, filename=file_name_html)
dataset[dataset.Country == 'China'].groupby(
'Date')['Confirmed'].sum().reset_index()).rename(columns={
'Date': 'ds',
'Confirmed': 'y'
})
confirmed_training_dataset.head()
prophet = Prophet(yearly_seasonality=False,
weekly_seasonality=False,
daily_seasonality=True,
seasonality_mode='additive')
prophet.fit(confirmed_training_dataset)
future = prophet.make_future_dataframe(periods=7)
confirmed_forecast = prophet.predict(future)
fig = plot_plotly(prophet, confirmed_forecast)
annotations = []
annotations.append(
dict(xref='paper',
yref='paper',
x=0.0,
y=1.05,
xanchor='left',
yanchor='bottom',
text='确诊人数预测',
font=dict(family='Arial', size=30, color='rgb(37,37,37)'),
showarrow=False))
fig.update_layout(annotations=annotations)
prophet = Prophet(yearly_seasonality=False,
weekly_seasonality=False,
df3 = df2[['date', 'confirmed']]
#Renaming column names according to fb prophet
df3.columns = ['ds', 'y']
#model
m = Prophet()
#fitting the model
m.fit(df3)
#forecast
future = m.make_future_dataframe(periods=20)
forecast = m.predict(future)
from fbprophet.plot import plot_plotly
fig4 = plot_plotly(m, forecast) # This returns a plotly Figure
fig4.update_layout(
autosize=False,
width=750,
height=800,
title_text='<b>Covid-19 Total cases Forecast<b>',
title_x=0.5,
paper_bgcolor='snow',
plot_bgcolor="snow",
)
maha = df[df.state == 'Maharashtra']
maha = maha[['date', 'confirmed']]
maha.columns = ['ds', 'y']
p = Prophet()
p.fit(maha)
<div style="font-family: 'Suez One';font-size:70px; background-color:white; color:black"><center>Covid-19 Global Predictor</center></div>
''',unsafe_allow_html=True
)
st.header(" ")
st.header("Region-wise predictions")
option1 = st.selectbox("Country",countries)
option2 = st.number_input("Forecast period in days",min_value=1,max_value=30)
option3 = st.selectbox("Select the forecasted variable",list(type_of_analysis.keys()))
if( len(option1) != 0):
m=Prophet(seasonality_mode='additive',yearly_seasonality=True,weekly_seasonality=True,daily_seasonality=True)
#,growth='logistic'
data=df[df['location']==option1][['date',type_of_analysis[option3]]].rename(columns={"date":"ds",type_of_analysis[option3]:"y"})
#data['floor']=0
#data['cap']=data['y'].max()+data['y'].mean()
m.fit(data)
future=m.make_future_dataframe(periods=option2)
#future['floor']=0
#future['cap']=data['y'].max()+data['y'].mean()
forecast = m.predict(future)
fig=plot_plotly(m, forecast)
fig.update_layout(
title="Forecasting",
xaxis_title="Date",
yaxis_title=option3)
st.plotly_chart(fig,config = {'displayModeBar': False})
cross_validation_results = cross_validation(m,initial='100 days',horizon="50 days")
#horizon='5 days'
#, initial='210 days', period='15 days',
performance_metrics_results = performance_metrics(cross_validation_results)
st.write(performance_metrics_results)
year = [1977, 1989, 1996, 1997, 2006, 2011, 2018]
speed = [62136, 173736, 27000, 122000, 58536, 209000, 343180]
plt.plot(year, speed, color='red')
plt.xlabel('Year Launched')
plt.ylabel('Fastest Speed Recorded (km/h)')
plt.title('Fastest Spacecrafts Launched in the Last 50 Years')
#create a dataframe for manipulation
data = [['1977-09-05', 62136], ['1989-10-18', 173736], ['2011-08-05', 209000], ['2018-08-12', 343180]]
df = pd.DataFrame(data, columns = ['ds', 'y'])
#use fbprophet to predict groth
m = Prophet()
m.fit(df)
future = m.make_future_dataframe(periods=365*243)
future.tail(n=10)
forecast = m.predict(future)
forecast.tail(n=3)
py.init_notebook_mode()
fig = plot_plotly(m, forecast) # This returns a plotly Figure
py.iplot(fig)
m.plot_components(forecast)
y = travelTime(50034460)
timeConver(y)
data2 = (data2['Price'].dropna().to_frame().reset_index().rename(columns={
"index": "ds",
'Price': "y"
}))
windows = st.sidebar.slider('prediction window',
min_value=80,
max_value=250,
value=90,
step=1)
model = Prophet()
model.fit(data2)
future = model.make_future_dataframe(periods=windows)
forecast = model.predict(future)
fig = plot_plotly(model, forecast)
fig.update_layout(
title='title',
yaxis_title=asset,
xaxis_title="Date",
)
st.write("# Forecast Prices")
st.plotly_chart(fig)
##########################################
model.plot_components(forecast)
#plt.title("Global Products")
plt.legend()
st.pyplot()
m.fit(df_train)
future = m.make_future_dataframe(periods=periode)
forecast = m.predict(future)
return forecast
#Visualisasi Ramalan
#selectbox untuk tampilkan plot
st.subheader("Tampilkan Visualisasi Awal")
check_1 = st.checkbox('tampilkan canddle stick')
check_2 = st.checkbox('tampilkan grafik garis')
if check_1:
plot_data_awal_canddlestick(df)
if check_2:
plot_data_awal_linechart(df)
st.subheader("Ramalan")
check_3 = st.checkbox('Lakukan Training')
if check_3:
hasil_training = forecast_stock(df)
st.write(hasil_training.tail())
st.subheader("Tampilkan Visualisasi Ramalan")
check_4 = st.checkbox("Visualisasi 1")
if check_4:
st.plotly_chart(plot_plotly(m, hasil_training))
df_treino = df_valores[['Date', 'Close']]
#renomear colunas
df_treino = df_treino.rename(columns={"Date": 'ds', 'Close': 'y'})
modelo = Prophet()
modelo.fit(df_treino)
futuro = modelo.make_future_dataframe(periods=n_dias, freq='B')
previsao = modelo.predict(futuro)
st.subheader('Previsão')
st.write(previsao[['ds', 'yhat','yhat_lower','yhat_upper' ]].tail(n_dias))
#grafico
grafico1 = plot_plotly(modelo, previsao)
st.plotly_chart(grafico1)
#grafico2
grafico2 = plot_components_plotly(modelo, previsao)
st.plotly_chart(grafico2)