mise en évidence pb logmodel.predict

module4/challenger.org

@@ -193,6 +193,8 @@ plt.savefig("fig1.png")
 #+END_SRC
 
 #+RESULTS:
+: /home/eliox/miniconda3/envs/mooc-rr-emacs/lib/python3.8/site-packages/pandas/plotting/_matplotlib/core.py:320: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).
+:   fig = self.plt.figure(figsize=self.figsize)
 : 'c' argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with 'x' & 'y'.  Please use a 2-D array with a single row if you really want to specify the same RGB or RGBA value for all points.
 
 [[./fig1.png]]
@@ -206,7 +208,7 @@ Let's assume O-rings independently fail with the same probability which
 solely depends on temperature. A logistic regression should allow us to
 estimate the influence of temperature.
 
-#+BEGIN_SRC python :session :exports both :results value
+#+BEGIN_SRC python :session :exports both :results output
 import statsmodels.api as sm
 
 data["Success"]=data.Count-data.Malfunction
@@ -215,11 +217,14 @@ data["Intercept"]=1
 logmodel=sm.GLM(data['Frequency'], data[['Intercept','Temperature']],
                 family=sm.families.Binomial(sm.families.links.logit)).fit()
 
-logmodel.summary()
+print(logmodel.summary())
 #+END_SRC
 
 #+RESULTS:
 #+begin_example
+/tmp/babel-tMc2Ef/python-Skn7ZF:7: DeprecationWarning: Calling Family(..) with a link class as argument is deprecated.
+Use an instance of a link class instead.
+  family=sm.families.Binomial(sm.families.links.logit)).fit()
                  Generalized Linear Model Regression Results                  
 ==============================================================================
 Dep. Variable:              Frequency   No. Observations:                   23
@@ -227,8 +232,8 @@ Model:                            GLM   Df Residuals:                       21
 Model Family:                Binomial   Df Model:                            1
 Link Function:                  logit   Scale:                          1.0000
 Method:                          IRLS   Log-Likelihood:                -3.9210
-Date:               mar., 05 mai 2020   Deviance:                       3.0144
-Time:                        22:25:31   Pearson chi2:                     5.00
+Date:               mer., 06 mai 2020   Deviance:                       3.0144
+Time:                        18:03:21   Pearson chi2:                     5.00
 No. Iterations:                     6                                         
 Covariance Type:            nonrobust                                         
 ===============================================================================
@@ -269,8 +274,8 @@ Model:                            GLM   Df Residuals:                       21
 Model Family:                Binomial   Df Model:                            1
 Link Function:                  logit   Scale:                          1.0000
 Method:                          IRLS   Log-Likelihood:                -23.526
-Date:               mar., 05 mai 2020   Deviance:                       18.086
-Time:                        22:25:31   Pearson chi2:                     30.0
+Date:               mer., 06 mai 2020   Deviance:                       18.086
+Time:                        18:03:21   Pearson chi2:                     30.0
 No. Iterations:                     6                                         
 Covariance Type:            nonrobust                                         
 ===============================================================================
@@ -300,14 +305,31 @@ estimate the failure probability for such temperature using our model.:
 #+BEGIN_SRC python :session :results output :exports both
 #%matplotlib inline
 data_pred = pd.DataFrame({'Temperature': np.linspace(start=30, stop=90, num=121), 'Intercept': 1})
+print(logmodel.predict(data_pred))
+"""
 data_pred['Frequency'] = logmodel.predict(data_pred)
-data_pred.plot(x="Temperature",y="Frequency",kind="line",ylim=[0,1])
+data_pred.plot(x="Temperature",y="Frequency",kind="scatter",ylim=[0,1])
 plt.scatter(x=data["Temperature"],y=data["Frequency"])
 plt.grid(True)
 plt.savefig("fig2.png")
+"""
 #+END_SRC
 
 #+RESULTS:
+#+begin_example
+0      1.0
+1      1.0
+2      1.0
+3      1.0
+4      1.0
+      ... 
+116    1.0
+117    1.0
+118    1.0
+119    1.0
+120    1.0
+Length: 121, dtype: float64
+#+end_example
 
 [[./fig2.png]]
 
@@ -325,6 +347,7 @@ I use regplot.
 
 #+BEGIN_SRC python :session :results output :exports both
 sns.set(color_codes=True)
+plt.figure(figsize=(5,3))
 plt.xlim(30,90)
 plt.ylim(0,1)
 sns.regplot(x='Temperature', y='Frequency', data=data, logistic=True)
@@ -332,6 +355,8 @@ plt.savefig("fig3.png")
 #+END_SRC
 
 #+RESULTS:
+: /tmp/babel-tMc2Ef/python-j6HG2E:2: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).
+:   plt.figure(figsize=(5,3))
 
 [[./fig3.png]]