update bayes training
This commit is contained in:
ritchie46 committed
1 parent
8805a9b83a
commit
17b2045f12
1 file changed
+390
-90
@@ -2,7 +2,7 @@
|
|||||||
"cells": [
|
"cells": [
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 1,
|
"execution_count": 2,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -43,7 +43,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 43,
|
"execution_count": 3,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -73,7 +73,7 @@
|
|||||||
"<IPython.core.display.HTML object>"
|
"<IPython.core.display.HTML object>"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 43,
|
"execution_count": 3,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
}
|
}
|
||||||
@@ -132,7 +132,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 14,
|
"execution_count": 4,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -141,7 +141,7 @@
|
|||||||
"outputs": [
|
"outputs": [
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 432x288 with 1 Axes>"
|
"<Figure size 432x288 with 1 Axes>"
|
||||||
]
|
]
|
||||||
@@ -162,10 +162,10 @@
|
|||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"0.4954"
|
"0.496"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 14,
|
"execution_count": 4,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
}
|
}
|
||||||
@@ -413,7 +413,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 43,
|
"execution_count": 5,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -430,13 +430,13 @@
|
|||||||
" <a list of 10 Patch objects>)"
|
" <a list of 10 Patch objects>)"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 43,
|
"execution_count": 5,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 1008x288 with 1 Axes>"
|
"<Figure size 1008x288 with 1 Axes>"
|
||||||
]
|
]
|
||||||
@@ -478,26 +478,34 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 45,
|
"execution_count": 6,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"outputs": [
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stderr",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"/opt/miniconda3/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
|
||||||
|
" return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<matplotlib.axes._subplots.AxesSubplot at 0x7f056adab550>"
|
"<matplotlib.axes._subplots.AxesSubplot at 0x7f04a8ee9be0>"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 45,
|
"execution_count": 6,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 1008x288 with 1 Axes>"
|
"<Figure size 1008x288 with 1 Axes>"
|
||||||
]
|
]
|
||||||
@@ -528,7 +536,7 @@
|
|||||||
}
|
}
|
||||||
},
|
},
|
||||||
"source": [
|
"source": [
|
||||||
"## What dis we just do?\n",
|
"## What did we just do?\n",
|
||||||
"\n",
|
"\n",
|
||||||
"The gaussian we fitted is actually a model that explains the data we've seen.\n",
|
"The gaussian we fitted is actually a model that explains the data we've seen.\n",
|
||||||
"\n",
|
"\n",
|
||||||
@@ -816,7 +824,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 147,
|
"execution_count": 7,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -831,14 +839,14 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 22000/22000 [00:03<00:00, 6271.39draws/s]\n"
|
"Sampling 4 chains: 100%|██████████| 14000/14000 [00:02<00:00, 5751.58draws/s]\n"
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"source": [
|
"source": [
|
||||||
"coin_flips = [1, 1]\n",
|
"coin_flips = [1, 1]\n",
|
||||||
"with pm.Model():\n",
|
"with pm.Model():\n",
|
||||||
" p = pm.Uniform('p', 0.001, 0.999)\n",
|
" p = pm.Uniform('p', 0, 1)\n",
|
||||||
" \n",
|
" \n",
|
||||||
" likelihood = pm.Bernoulli('observed', p=p, observed=coin_flips)\n",
|
" likelihood = pm.Bernoulli('observed', p=p, observed=coin_flips)\n",
|
||||||
" \n",
|
" \n",
|
||||||
@@ -849,37 +857,28 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 148,
|
"execution_count": 8,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"outputs": [
|
"outputs": [
|
||||||
{
|
|
||||||
"name": "stderr",
|
|
||||||
"output_type": "stream",
|
|
||||||
"text": [
|
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
|
||||||
},
|
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f054b83a400>,\n",
|
"array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f049d161630>,\n",
|
||||||
" <matplotlib.axes._subplots.AxesSubplot object at 0x7f054b813c88>]],\n",
|
" <matplotlib.axes._subplots.AxesSubplot object at 0x7f049d18f5f8>]],\n",
|
||||||
" dtype=object)"
|
" dtype=object)"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 148,
|
"execution_count": 8,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -896,7 +895,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 156,
|
"execution_count": 9,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -907,15 +906,15 @@
|
|||||||
"name": "stdout",
|
"name": "stdout",
|
||||||
"output_type": "stream",
|
"output_type": "stream",
|
||||||
"text": [
|
"text": [
|
||||||
"p: 0.3: 0.08795\n",
|
"P(p < 0.3): 0.09008333333333333\n",
|
||||||
"p: 0.5: 0.25275\n",
|
"P(p < 0.5): 0.24941666666666668\n",
|
||||||
"p: 0.8: 0.63875\n"
|
"P(p < 0.8): 0.6446666666666667\n"
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"source": [
|
"source": [
|
||||||
"for p in [.3, .5, .8]:\n",
|
"for p in [.3, .5, .8]:\n",
|
||||||
" print(f'p: {p}:', (trace['p'] < p).sum() / trace['p'].shape[0])"
|
" print(f'P(p < {p}):', (trace['p'] < p).sum() / trace['p'].shape[0])"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -934,7 +933,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 161,
|
"execution_count": 10,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -944,16 +943,16 @@
|
|||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"[<matplotlib.lines.Line2D at 0x7f054d5a2198>]"
|
"[<matplotlib.lines.Line2D at 0x7f049d0b2080>]"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 161,
|
"execution_count": 10,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 432x288 with 1 Axes>"
|
"<Figure size 432x288 with 1 Axes>"
|
||||||
]
|
]
|
||||||
@@ -971,7 +970,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 162,
|
"execution_count": 11,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -986,8 +985,7 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 14000/14000 [00:02<00:00, 6167.84draws/s]\n",
|
"Sampling 4 chains: 100%|██████████| 14000/14000 [00:03<00:00, 4603.02draws/s]\n"
|
||||||
"The acceptance probability does not match the target. It is 0.8795086571448326, but should be close to 0.8. Try to increase the number of tuning steps.\n"
|
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
@@ -1003,37 +1001,28 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 163,
|
"execution_count": 12,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"outputs": [
|
"outputs": [
|
||||||
{
|
|
||||||
"name": "stderr",
|
|
||||||
"output_type": "stream",
|
|
||||||
"text": [
|
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
|
||||||
},
|
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f0539ff65f8>,\n",
|
"array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f048d4c3630>,\n",
|
||||||
" <matplotlib.axes._subplots.AxesSubplot object at 0x7f0539fac400>]],\n",
|
" <matplotlib.axes._subplots.AxesSubplot object at 0x7f048d4e2048>]],\n",
|
||||||
" dtype=object)"
|
" dtype=object)"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 163,
|
"execution_count": 12,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1058,7 +1047,7 @@
|
|||||||
"source": [
|
"source": [
|
||||||
"## Who will be elected?\n",
|
"## Who will be elected?\n",
|
||||||
"\n",
|
"\n",
|
||||||
"There will be a new referendum about a possible 'Dexit'.\n",
|
"There will be a new referendum about a possible 'Nexit'.\n",
|
||||||
"\n",
|
"\n",
|
||||||
"People have to vote $1$ for staying and $0$ for leaving.\n",
|
"People have to vote $1$ for staying and $0$ for leaving.\n",
|
||||||
"\n",
|
"\n",
|
||||||
@@ -1067,17 +1056,17 @@
|
|||||||
"\\begin{align}\n",
|
"\\begin{align}\n",
|
||||||
"f(x)= \n",
|
"f(x)= \n",
|
||||||
"\\begin{cases}\n",
|
"\\begin{cases}\n",
|
||||||
" 0.2,& \\text{staying } \\\\\n",
|
" 0.8,& \\text{staying } \\\\\n",
|
||||||
" 0.8, & \\text{leaving}\n",
|
" 0.2, & \\text{leaving}\n",
|
||||||
"\\end{cases}\n",
|
"\\end{cases}\n",
|
||||||
"\\end{align}\n",
|
"\\end{align}\n",
|
||||||
"\n",
|
"\n",
|
||||||
"Meanwhile the first votes are coming in. What is the probability of a 'Dexit' occurring?"
|
"Meanwhile the first votes are coming in. What is the probability of a 'Nexit' occurring?"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 3,
|
"execution_count": 13,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -1090,7 +1079,7 @@
|
|||||||
"array([1, 1, 1, 1, 1, 1, 1, 0, 0, 1])"
|
"array([1, 1, 1, 1, 1, 1, 1, 0, 0, 1])"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 3,
|
"execution_count": 13,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
}
|
}
|
||||||
@@ -1102,7 +1091,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 3,
|
"execution_count": 14,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -1112,16 +1101,16 @@
|
|||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"[<matplotlib.lines.Line2D at 0x7f9297e93978>]"
|
"[<matplotlib.lines.Line2D at 0x7f048b80eba8>]"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 3,
|
"execution_count": 14,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 432x288 with 1 Axes>"
|
"<Figure size 432x288 with 1 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1133,13 +1122,15 @@
|
|||||||
}
|
}
|
||||||
],
|
],
|
||||||
"source": [
|
"source": [
|
||||||
|
"# translate Maurice's probabilities to a prior believe\n",
|
||||||
|
"\n",
|
||||||
"x = np.linspace(0, 1)\n",
|
"x = np.linspace(0, 1)\n",
|
||||||
"plt.plot(x, stats.beta(1.3, 3).pdf(x))"
|
"plt.plot(x, stats.beta(1.3, 3).pdf(x))"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 4,
|
"execution_count": 15,
|
||||||
"metadata": {
|
"metadata": {
|
||||||
"slideshow": {
|
"slideshow": {
|
||||||
"slide_type": "slide"
|
"slide_type": "slide"
|
||||||
@@ -1154,10 +1145,8 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:06<00:00, 6487.09draws/s]\n",
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5357.62draws/s]\n",
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
"The number of effective samples is smaller than 25% for some parameters.\n"
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1169,7 +1158,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1187,10 +1176,7 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:06<00:00, 6164.67draws/s]\n",
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5601.20draws/s]\n"
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1202,7 +1188,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1220,10 +1206,7 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:06<00:00, 6377.00draws/s]\n",
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5444.02draws/s]\n"
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1235,7 +1218,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1253,10 +1236,9 @@
|
|||||||
"Initializing NUTS using jitter+adapt_diag...\n",
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
"NUTS: [p]\n",
|
"NUTS: [p]\n",
|
||||||
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5921.57draws/s]\n",
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5619.20draws/s]\n",
|
||||||
"/opt/miniconda3/lib/python3.7/site-packages/matplotlib/axes/_base.py:3455: MatplotlibDeprecationWarning: \n",
|
"The acceptance probability does not match the target. It is 0.8897717351599277, but should be close to 0.8. Try to increase the number of tuning steps.\n",
|
||||||
"The `ymin` argument was deprecated in Matplotlib 3.0 and will be removed in 3.2. Use `bottom` instead.\n",
|
"The acceptance probability does not match the target. It is 0.8937936259844127, but should be close to 0.8. Try to increase the number of tuning steps.\n"
|
||||||
" alternative='`bottom`', obj_type='argument')\n"
|
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
@@ -1268,7 +1250,7 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"data": {
|
"data": {
|
||||||
"image/png": "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 truncated
|
"image/png": "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 truncated
|
||||||
"text/plain": [
|
"text/plain": [
|
||||||
"<Figure size 864x144 with 2 Axes>"
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
]
|
]
|
||||||
@@ -1294,12 +1276,330 @@
|
|||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 5,
|
"execution_count": 16,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"outputs": [],
|
"outputs": [],
|
||||||
"source": [
|
"source": [
|
||||||
"np.save('data/votes.npy', stats.bernoulli(0.68).rvs(1000))"
|
"np.save('data/votes.npy', stats.bernoulli(0.68).rvs(1000))"
|
||||||
]
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "markdown",
|
||||||
|
"metadata": {},
|
||||||
|
"source": [
|
||||||
|
"## A/B/C testing\n",
|
||||||
|
"\n",
|
||||||
|
"* Three different colors are tested for the Xomnia website, and randomly assigned to visitors (not uniformly)\n",
|
||||||
|
"* Should we switch colors?\n",
|
||||||
|
"* Assume Xomnia goes bankrupt if we don't get at least 35 calls. We expect 1000 visitors next month, what is the probability that we don't make it? Which color should we use?\n",
|
||||||
|
"\n",
|
||||||
|
"color | number of visitors | number of calls\n",
|
||||||
|
"--- | --- | ---\n",
|
||||||
|
"blue | 2000 | 35\n",
|
||||||
|
"red | 500 | 8\n",
|
||||||
|
"green | 1000 | 18"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 17,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stderr",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"Auto-assigning NUTS sampler...\n",
|
||||||
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
|
"NUTS: [p]\n",
|
||||||
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5808.91draws/s]\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"# Simple model for one color\n",
|
||||||
|
"\n",
|
||||||
|
"with pm.Model():\n",
|
||||||
|
" \n",
|
||||||
|
" # prior\n",
|
||||||
|
" p = pm.Uniform('p', lower=0, upper=1)\n",
|
||||||
|
" conversions = pm.Binomial('conversions', 2000, p, observed = [35])\n",
|
||||||
|
" \n",
|
||||||
|
" samples = pm.sample(10000)\n",
|
||||||
|
" \n",
|
||||||
|
"pm.traceplot(samples);"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "markdown",
|
||||||
|
"metadata": {},
|
||||||
|
"source": [
|
||||||
|
"## Combining parameters\n",
|
||||||
|
"\n",
|
||||||
|
"* You can easily combine multiple parameters and observations in the same model"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 18,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stderr",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"Auto-assigning NUTS sampler...\n",
|
||||||
|
"Initializing NUTS using jitter+adapt_diag...\n",
|
||||||
|
"Multiprocess sampling (4 chains in 4 jobs)\n",
|
||||||
|
"NUTS: [p]\n",
|
||||||
|
"Sampling 4 chains: 100%|██████████| 42000/42000 [00:07<00:00, 5600.55draws/s]\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 864x144 with 2 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"with pm.Model():\n",
|
||||||
|
" \n",
|
||||||
|
" p = pm.Uniform('p', lower=0, upper=1, shape=3)\n",
|
||||||
|
" conversions = pm.Binomial('conv', [2000, 500, 1000], p, observed = [35, 8, 18])\n",
|
||||||
|
" \n",
|
||||||
|
" trace = pm.sample(10000)\n",
|
||||||
|
" \n",
|
||||||
|
"pm.traceplot(trace)\n",
|
||||||
|
"\n",
|
||||||
|
"blue = trace['p'][:, 0]\n",
|
||||||
|
"red = trace['p'][:, 1]\n",
|
||||||
|
"green = trace['p'][:, 2]"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 19,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stderr",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"/opt/miniconda3/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
|
||||||
|
" return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"text/plain": [
|
||||||
|
"<matplotlib.axes._subplots.AxesSubplot at 0x7f048b8700f0>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"execution_count": 19,
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "execute_result"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 864x288 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"plt.figure(figsize=(12, 4))\n",
|
||||||
|
"sns.distplot(blue, color='blue', hist=False)\n",
|
||||||
|
"sns.distplot(red, color='red', hist=False)\n",
|
||||||
|
"sns.distplot(green, color='green', hist=False)"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "markdown",
|
||||||
|
"metadata": {},
|
||||||
|
"source": [
|
||||||
|
"## Posterior samples\n",
|
||||||
|
"\n",
|
||||||
|
"* `pymc3` does not generate a real distribution, but samples from the posterior distribution.\n",
|
||||||
|
"* With those samples you can \n",
|
||||||
|
" * Understand the data, draw conclusions\n",
|
||||||
|
" * Generate predictions\n",
|
||||||
|
"\n",
|
||||||
|
"## Exercise\n",
|
||||||
|
"1. What is the probability that the conversion rate of blue is higher than 2%?\n",
|
||||||
|
"2. Which color should we choose?\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 20,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"text/plain": [
|
||||||
|
"0.2357"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"execution_count": 20,
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "execute_result"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"(blue > 0.02).sum() / blue.shape[0]"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 34,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 432x288 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"calls = stats.binom(n=1000, p=blue).rvs()\n",
|
||||||
|
"fig = plt.hist(calls, bins=calls.max() - calls.min())"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 59,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"text/plain": [
|
||||||
|
"array([[16, 23, 17, ..., 19, 24, 19],\n",
|
||||||
|
" [13, 19, 23, ..., 8, 28, 17],\n",
|
||||||
|
" [17, 20, 20, ..., 21, 10, 29]])"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"execution_count": 59,
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "execute_result"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"calls = stats.binom(n=1000, p=np.stack([blue, red, green])).rvs()\n",
|
||||||
|
"calls"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 74,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"text/plain": [
|
||||||
|
"(0, 40)"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"execution_count": 74,
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "execute_result"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1152x576 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"plt.figure(figsize=(16, 8))\n",
|
||||||
|
"plt.title('Calls histogram per color')\n",
|
||||||
|
"plt.hist(calls.T, (calls.max(1) - calls.min(1)).min(), color='brg', ec='black')\n",
|
||||||
|
"plt.xlabel('Calls')\n",
|
||||||
|
"plt.xlim(0, 40)"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 72,
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"[0.00165 0.019875 0.009775]\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"text/plain": [
|
||||||
|
"<BarContainer object of 3 artists>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"execution_count": 72,
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "execute_result"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1152x576 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {
|
||||||
|
"needs_background": "light"
|
||||||
|
},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"mean = (calls > 35).mean(1)\n",
|
||||||
|
"print(mean)\n",
|
||||||
|
"\n",
|
||||||
|
"plt.figure(figsize=(16, 8))\n",
|
||||||
|
"plt.title('Probability of evading bankrupty')\n",
|
||||||
|
"plt.bar(np.array(list('brg')), mean, color='brg', ec='black')\n"
|
||||||
|
]
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"metadata": {
|
"metadata": {
|
||||||
|
|||||||
Reference in new issue
Block a user