time series forecasting using prophet model get error during run time

daoudimca

######## python and prophet version details

prophet version=1.1.4
python_version=3.9.12

I am using the same version in my local machine its working file. but i am facing error on DATAIKU

########### Code Sample ########

lag_periods = [1,5, 10, 15]

for lag in lag_periods:
df_prophet[f'Lag_{lag}'] = df_prophet['y'].shift(lag)

df_prophet['Rolling_Mean_7'] = df_prophet['y'].rolling(window=3).mean()
df_prophet['Rolling_Mean_14'] = df_prophet['y'].rolling(window=14).mean()
df_prophet['Rolling_Mean_30'] = df_prophet['y'].rolling(window=30).mean()

df_prophet['Exp_Moving_Avg_3'] = df_prophet['y'].ewm(span=3).mean()
df_prophet['Exp_Moving_Avg_7'] = df_prophet['y'].ewm(span=7).mean()

# Create trend features
df_prophet['Linear_Trend'] = df_prophet['y'].ewm(alpha=0.1).mean()
df_prophet['Quadratic_Trend'] = df_prophet['y'].ewm(alpha=0.1).mean().pow(2)
df_prophet['Exponential_Trend'] = df_prophet['y'].ewm(alpha=0.1).mean().ewm(alpha=0.1).mean()
df_prophet['Cubic_Trend'] = df_prophet['y'].ewm(alpha=0.5).mean().pow(3)

# Create outlier features
#df_prophet['Is_Outlier_Sale'] = df_prophet['y'].apply(lambda x: x > 3 * df['y'].std())

# Create rate of change feature
df_prophet['Rate_of_Change'] = df_prophet['y'].pct_change() * 100

# Initialize and fit Prophet model
prophet_model = Prophet( seasonality_prior_scale=0.05, changepoint_prior_scale=0.2,
yearly_seasonality=True, weekly_seasonality=True, daily_seasonality=True)
prophet_model.add_seasonality(name='monthly', period=30.44, fourier_order=4) # Capturing monthly seasonality
prophet_model.add_country_holidays(country_name='SA')

# Fit the model
prophet_model.fit(df_prophet)

#############################

RuntimeError: Error during optimization! Command '/dataiku/Data_dir_design/code-envs/python/Py39/lib/python3.9/site-packages/prophet/stan_model/prophet_model.bin random seed=91535 data file=/tmp/tmpnvp6v2uv/tvj8jixl.json init=/tmp/tmpnvp6v2uv/czt0d5ix.json output file=/tmp/tmpnvp6v2uv/prophet_modelc9a_94m7/prophet_model-20230919112247.csv method=optimize algorithm=newton iter=10000' failed:

AdrienL

Hi,
It seems the error message was cut off where the details were about to be given. See the end of the error message, that says in essence "the process failed:" and then cuts off. Pasting the rest of the error message would help.

daoudimca

finally, i found the solution. The current model support on below configuration

#prophet==1.1.1, holidays==0.18.0 ,python 3.8.13, Linux