From a3ae4715bbc0a18cde35f7b199a671f53df80168 Mon Sep 17 00:00:00 2001
From: Maxim Grishin <igevorse@gmail.com>
Date: Sun, 29 Sep 2019 23:24:38 +0300
Subject: [PATCH] Make Python version consistent with R

---
 R/R/prophet.R                  | 12 ++++++------
 python/fbprophet/forecaster.py |  8 ++++----
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/R/R/prophet.R b/R/R/prophet.R
index fe851f4..c8c8948 100644
--- a/R/R/prophet.R
+++ b/R/R/prophet.R
@@ -1264,8 +1264,8 @@ fit.prophet <- function(m, df, ...) {
       iter = m$mcmc.samples
     )
     args <- utils::modifyList(args, list(...))
-    stan.fit <- do.call(rstan::sampling, args)
-    m$params <- rstan::extract(stan.fit)
+    m$stan.fit <- do.call(rstan::sampling, args)
+    m$params <- rstan::extract(m$stan.fit)
     n.iteration <- length(m$params$k)
   } else {
     args <- list(
@@ -1277,15 +1277,15 @@ fit.prophet <- function(m, df, ...) {
       as_vector = FALSE
     )
     args <- utils::modifyList(args, list(...))
-    stan.fit <- do.call(rstan::optimizing, args)
-    if (stan.fit$return_code != 0) {
+    m$stan.fit <- do.call(rstan::optimizing, args)
+    if (m$stan.fit$return_code != 0) {
       message(
         'Optimization terminated abnormally. Falling back to Newton optimizer.'
       )
       args$algorithm = 'Newton'
-      stan.fit <- do.call(rstan::optimizing, args)
+      m$stan.fit <- do.call(rstan::optimizing, args)
     }
-    m$params <- stan.fit$par
+    m$params <- m$stan.fit$par
     n.iteration <- 1
   }
   
diff --git a/python/fbprophet/forecaster.py b/python/fbprophet/forecaster.py
index 3635581..16086b2 100644
--- a/python/fbprophet/forecaster.py
+++ b/python/fbprophet/forecaster.py
@@ -1131,17 +1131,17 @@ class Prophet(object):
             )
             args.update(kwargs)
             try:
-                params = model.optimizing(**args)
+                self.stan_fit = model.optimizing(**args)
             except RuntimeError:
                 # Fall back on Newton
                 logger.warning(
                     'Optimization terminated abnormally. Falling back to Newton.'
                 )
                 args['algorithm'] = 'Newton'
-                params = model.optimizing(**args)
+                self.stan_fit = model.optimizing(**args)
 
-            for par in params:
-                self.params[par] = params[par].reshape((1, -1))
+            for par in self.stan_fit:
+                self.params[par] = self.stan_fit[par].reshape((1, -1))
 
         # If no changepoints were requested, replace delta with 0s
         if len(self.changepoints) == 0: