Skip to content

M
mooc-rr
Commit bd72d97a authored by Agathe Schmider's avatar Agathe Schmider
Browse files
Options

end exercice 2

parent 9c9ca2c3
Show whitespace changes
Inline Side-by-side
Showing with 452 additions and 381 deletions
module4/.ipynb_checkpoints/challenger-checkpoint.ipynb
View file @ bd72d97a
This source diff could not be displayed because it is too large. You can view the blob instead.
module4/challenger.ipynb
View file @ bd72d97a
...@@ -538,10 +538,10 @@ ...@@ -538,10 +538,10 @@
" <th>Method:</th> <td>IRLS</td> <th> Log-Likelihood: </th> <td> -3.9210</td>\n", " <th>Method:</th> <td>IRLS</td> <th> Log-Likelihood: </th> <td> -3.9210</td>\n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>Date:</th> <td>Tue, 14 Apr 2020</td> <th> Deviance: </th> <td> 3.0144</td>\n", " <th>Date:</th> <td>Wed, 15 Apr 2020</td> <th> Deviance: </th> <td> 3.0144</td>\n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>Time:</th> <td>10:27:51</td> <th> Pearson chi2: </th> <td> 5.00</td> \n", " <th>Time:</th> <td>15:47:34</td> <th> Pearson chi2: </th> <td> 5.00</td> \n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>No. Iterations:</th> <td>6</td> <th> </th> <td> </td> \n", " <th>No. Iterations:</th> <td>6</td> <th> </th> <td> </td> \n",
...@@ -572,8 +572,8 @@ ...@@ -572,8 +572,8 @@
"Model Family: Binomial Df Model: 1\n", "Model Family: Binomial Df Model: 1\n",
"Link Function: logit Scale: 1.0000\n", "Link Function: logit Scale: 1.0000\n",
"Method: IRLS Log-Likelihood: -3.9210\n", "Method: IRLS Log-Likelihood: -3.9210\n",
"Date: Tue, 14 Apr 2020 Deviance: 3.0144\n", "Date: Wed, 15 Apr 2020 Deviance: 3.0144\n",
"Time: 10:27:51 Pearson chi2: 5.00\n", "Time: 15:47:34 Pearson chi2: 5.00\n",
"No. Iterations: 6 \n", "No. Iterations: 6 \n",
"Covariance Type: nonrobust \n", "Covariance Type: nonrobust \n",
"===============================================================================\n", "===============================================================================\n",
...@@ -644,10 +644,10 @@ ...@@ -644,10 +644,10 @@
" <th>Method:</th> <td>IRLS</td> <th> Log-Likelihood: </th> <td> -23.526</td>\n", " <th>Method:</th> <td>IRLS</td> <th> Log-Likelihood: </th> <td> -23.526</td>\n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>Date:</th> <td>Tue, 14 Apr 2020</td> <th> Deviance: </th> <td> 18.086</td>\n", " <th>Date:</th> <td>Wed, 15 Apr 2020</td> <th> Deviance: </th> <td> 18.086</td>\n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>Time:</th> <td>10:27:51</td> <th> Pearson chi2: </th> <td> 30.0</td> \n", " <th>Time:</th> <td>15:47:34</td> <th> Pearson chi2: </th> <td> 30.0</td> \n",
"</tr>\n", "</tr>\n",
"<tr>\n", "<tr>\n",
" <th>No. Iterations:</th> <td>6</td> <th> </th> <td> </td> \n", " <th>No. Iterations:</th> <td>6</td> <th> </th> <td> </td> \n",
...@@ -678,8 +678,8 @@ ...@@ -678,8 +678,8 @@
"Model Family: Binomial Df Model: 1\n", "Model Family: Binomial Df Model: 1\n",
"Link Function: logit Scale: 1.0000\n", "Link Function: logit Scale: 1.0000\n",
"Method: IRLS Log-Likelihood: -23.526\n", "Method: IRLS Log-Likelihood: -23.526\n",
"Date: Tue, 14 Apr 2020 Deviance: 18.086\n", "Date: Wed, 15 Apr 2020 Deviance: 18.086\n",
"Time: 10:27:51 Pearson chi2: 30.0\n", "Time: 15:47:34 Pearson chi2: 30.0\n",
"No. Iterations: 6 \n", "No. Iterations: 6 \n",
"Covariance Type: nonrobust \n", "Covariance Type: nonrobust \n",
"===============================================================================\n", "===============================================================================\n",
...@@ -780,55 +780,6 @@ ...@@ -780,55 +780,6 @@
"plt.grid(True)" "plt.grid(True)"
] ]
}, },
{
"cell_type": "markdown",
"metadata": {},
"source": [
"There were warnings during the construction of the log model. let's try and change it"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"ename": "DistributionNotFound",
"evalue": "The 'statsmodel==0.9.0' distribution was not found and is required by the application",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mDistributionNotFound\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-12-0d6a0b26ac77>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mpkg_resources\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mpkg_resources\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrequire\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"statsmodel==0.9.0\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mstatsmodel\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m~/miniconda3/lib/python3.7/site-packages/pkg_resources/__init__.py\u001b[0m in \u001b[0;36mrequire\u001b[0;34m(self, *requirements)\u001b[0m\n\u001b[1;32m 899\u001b[0m \u001b[0mincluded\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0meven\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mthey\u001b[0m \u001b[0mwere\u001b[0m \u001b[0malready\u001b[0m \u001b[0mactivated\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mthis\u001b[0m \u001b[0mworking\u001b[0m \u001b[0mset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 900\u001b[0m \"\"\"\n\u001b[0;32m--> 901\u001b[0;31m \u001b[0mneeded\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresolve\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparse_requirements\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequirements\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 902\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 903\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mdist\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mneeded\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m~/miniconda3/lib/python3.7/site-packages/pkg_resources/__init__.py\u001b[0m in \u001b[0;36mresolve\u001b[0;34m(self, requirements, env, installer, replace_conflicting, extras)\u001b[0m\n\u001b[1;32m 785\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mdist\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 786\u001b[0m \u001b[0mrequirers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrequired_by\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 787\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mDistributionNotFound\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrequirers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 788\u001b[0m \u001b[0mto_activate\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 789\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mdist\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mreq\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mDistributionNotFound\u001b[0m: The 'statsmodel==0.9.0' distribution was not found and is required by the application"
]
}
],
"source": [
"import pkg_resources\n",
"pkg_resources.require(\"statsmodel==0.9.0\")\n",
"import statsmodel"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import statsmodels.api as sm\n",
"\n",
"data[\"Success\"]=data.Count-data.Malfunction\n",
"data[\"Intercept\"]=1\n",
"\n",
"logmodel=sm.GLM(data['Frequency'], data[['Intercept','Temperature']], \n",
" family=sm.families.Binomial(sm.families.links.logit)).fit()\n",
"\n",
"logmodel.summary()\n"
]
},
{ {
"cell_type": "markdown", "cell_type": "markdown",
"metadata": { "metadata": {
...@@ -861,7 +812,7 @@ ...@@ -861,7 +812,7 @@
"outputs": [ "outputs": [
{ {
"data": { "data": {
"image/png": "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\n", "image/png": "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\n",
"text/plain": [ "text/plain": [
"<Figure size 432x288 with 1 Axes>" "<Figure size 432x288 with 1 Axes>"
] ]
...@@ -884,174 +835,6 @@ ...@@ -884,174 +835,6 @@
"source": [ "source": [
"**I think I have managed to correctly compute and plot the uncertainty of my prediction.** Although the shaded area seems very similar to [the one obtained by with R](https://app-learninglab.inria.fr/moocrr/gitlab/moocrr-session3/moocrr-reproducibility-study/tree/master/challenger.pdf), I can spot a few differences (e.g., the blue point for temperature 63 is outside)... Could this be a numerical error ? Or a difference in the statistical method ? It is not clear which one is \"right\"." "**I think I have managed to correctly compute and plot the uncertainty of my prediction.** Although the shaded area seems very similar to [the one obtained by with R](https://app-learninglab.inria.fr/moocrr/gitlab/moocrr-session3/moocrr-reproducibility-study/tree/master/challenger.pdf), I can spot a few differences (e.g., the blue point for temperature 63 is outside)... Could this be a numerical error ? Or a difference in the statistical method ? It is not clear which one is \"right\"."
] ]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Temperature</th>\n",
" <th>Intercept</th>\n",
" <th>Frequency</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>30.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>30.5</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>31.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>31.5</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>32.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>116</th>\n",
" <td>88.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>117</th>\n",
" <td>88.5</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>118</th>\n",
" <td>89.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>119</th>\n",
" <td>89.5</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>120</th>\n",
" <td>90.0</td>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>121 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" Temperature Intercept Frequency\n",
"0 30.0 1 1.0\n",
"1 30.5 1 1.0\n",
"2 31.0 1 1.0\n",
"3 31.5 1 1.0\n",
"4 32.0 1 1.0\n",
".. ... ... ...\n",
"116 88.0 1 1.0\n",
"117 88.5 1 1.0\n",
"118 89.0 1 1.0\n",
"119 89.5 1 1.0\n",
"120 90.0 1 1.0\n",
"\n",
"[121 rows x 3 columns]"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data_pred"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0 1.0\n",
"1 1.0\n",
"2 1.0\n",
"3 1.0\n",
"4 1.0\n",
" ... \n",
"116 1.0\n",
"117 1.0\n",
"118 1.0\n",
"119 1.0\n",
"120 1.0\n",
"Length: 121, dtype: float64"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data_pred = pd.DataFrame({'Temperature': np.linspace(start=30, stop=90, num=121), 'Intercept': 1})\n",
"logmodel.predict(data_pred)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
} }
], ],
"metadata": { "metadata": {
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment