From 8ef4dc5bebbc0ccba4b5c55eceb2c40fd39d2414 Mon Sep 17 00:00:00 2001 From: aloctavodia Date: Mon, 5 Jun 2017 16:26:21 +0200 Subject: [PATCH 1/5] add weighted ppc --- pymc3/sampling.py | 52 ++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 42 insertions(+), 10 deletions(-) diff --git a/pymc3/sampling.py b/pymc3/sampling.py index a554c25acf..dde914942a 100644 --- a/pymc3/sampling.py +++ b/pymc3/sampling.py @@ -484,14 +484,16 @@ def _update_start_vals(a, b, model): a.update({k: v for k, v in b.items() if k not in a}) + def sample_ppc(trace, samples=None, model=None, vars=None, size=None, - random_seed=None, progressbar=True): + weights=None, random_seed=None, progressbar=True): """Generate posterior predictive samples from a model given a trace. Parameters ---------- trace : backend, list, or MultiTrace - Trace generated from MCMC sampling + Trace generated from MCMC sampling. If a set of weights is also passed + this can be a list of traces, useful for model averaging. samples : int Number of posterior predictive samples to generate. Defaults to the length of `trace` @@ -503,16 +505,23 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, size : int The number of random draws from the distribution specified by the parameters in each sample of the trace. + weights: array-like + Individuals weights for each trace, useful for model averaging + random_seed : int + Seed for the random number generator. + progressbar : bool + Whether or not to display a progress bar in the command line. The + bar shows the percentage of completion, the sampling speed in + samples per second (SPS), and the estimated remaining time until + completion ("expected time of arrival"; ETA). Returns ------- samples : dict - Dictionary with the variables as keys. The values corresponding - to the posterior predictive samples. + Dictionary with the variables as keys. The values corresponding to the + posterior predictive samples. If a set of weights and a matching number + of traces are provided, then the samples will be weighted. """ - if samples is None: - samples = len(trace) - if model is None: model = modelcontext(model) @@ -521,10 +530,33 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, seed(random_seed) + if weights is not None: + if len(trace) != len(weights): + raise ValueError( + 'The number of traces and weights should be the same') + + weights = np.asarray(weights) + p = weights / np.sum(weights) + + min_tr = min([len(i) for i in trace]) + + n = (min_tr * p).astype('int') + # ensure n sum up to min_tr + idx = np.argmax(n) + n[idx] = n[idx] + min_tr - np.sum(n) + + trace = np.concatenate([np.random.choice(trace[i], j) + for i, j in enumerate(n)]) + + len_trace = len(trace) + + if samples is None: + samples = len_trace + + indices = randint(0, len_trace, samples) + if progressbar: - indices = tqdm(randint(0, len(trace), samples), total=samples) - else: - indices = randint(0, len(trace), samples) + indices = tqdm(indices, total=samples) ppc = defaultdict(list) for idx in indices: From 2852edcd1114ae4eebef7fb929be8e55ad39c7ab Mon Sep 17 00:00:00 2001 From: aloctavodia Date: Wed, 7 Jun 2017 10:09:20 +0200 Subject: [PATCH 2/5] separated function for weighted ppc --- pymc3/sampling.py | 114 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 92 insertions(+), 22 deletions(-) diff --git a/pymc3/sampling.py b/pymc3/sampling.py index dde914942a..8656cef0e1 100644 --- a/pymc3/sampling.py +++ b/pymc3/sampling.py @@ -18,7 +18,7 @@ import sys sys.setrecursionlimit(10000) -__all__ = ['sample', 'iter_sample', 'sample_ppc', 'init_nuts'] +__all__ = ['sample', 'iter_sample', 'sample_ppc', 'sample_ppc_w', 'init_nuts'] STEP_METHODS = (NUTS, HamiltonianMC, Metropolis, BinaryMetropolis, BinaryGibbsMetropolis, Slice, CategoricalGibbsMetropolis) @@ -486,14 +486,13 @@ def _update_start_vals(a, b, model): def sample_ppc(trace, samples=None, model=None, vars=None, size=None, - weights=None, random_seed=None, progressbar=True): + random_seed=None, progressbar=True): """Generate posterior predictive samples from a model given a trace. Parameters ---------- trace : backend, list, or MultiTrace - Trace generated from MCMC sampling. If a set of weights is also passed - this can be a list of traces, useful for model averaging. + Trace generated from MCMC sampling. samples : int Number of posterior predictive samples to generate. Defaults to the length of `trace` @@ -505,8 +504,6 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, size : int The number of random draws from the distribution specified by the parameters in each sample of the trace. - weights: array-like - Individuals weights for each trace, useful for model averaging random_seed : int Seed for the random number generator. progressbar : bool @@ -522,6 +519,9 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, posterior predictive samples. If a set of weights and a matching number of traces are provided, then the samples will be weighted. """ + if samples is None: + samples = len(trace) + if model is None: model = modelcontext(model) @@ -530,23 +530,94 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, seed(random_seed) - if weights is not None: - if len(trace) != len(weights): - raise ValueError( - 'The number of traces and weights should be the same') + indices = randint(0, len(trace), samples) + if progressbar: + indices = tqdm(indices, total=samples) - weights = np.asarray(weights) - p = weights / np.sum(weights) + ppc = defaultdict(list) + for idx in indices: + param = trace[idx] + for var in vars: + ppc[var.name].append(var.distribution.random(point=param, + size=size)) - min_tr = min([len(i) for i in trace]) + return {k: np.asarray(v) for k, v in ppc.items()} - n = (min_tr * p).astype('int') - # ensure n sum up to min_tr - idx = np.argmax(n) - n[idx] = n[idx] + min_tr - np.sum(n) - trace = np.concatenate([np.random.choice(trace[i], j) - for i, j in enumerate(n)]) +def sample_ppc_w(traces, samples=None, models=None, size=None, weights=None, + random_seed=None, progressbar=True): + """Generate weighted posterior predictive samples from a list of models and + a list of traces according to a set of weights. + + Parameters + ---------- + traces : list + List of traces generated from MCMC sampling. The number of traces should + be equal to the number of weights. + samples : int + Number of posterior predictive samples to generate. Defaults to the + length of the shorter trace in traces. + models : list + List of models used to generate the list of traces. The number of models + should be equal to the number of weights and the number of observed RVs + should be the same for all models. + By default a single model will be inferred from `with` context, in this + case results will only be meaningful if all models share the same + distributions for the observed RVs. + size : int + The number of random draws from the distributions specified by the + parameters in each sample of the trace. + weights: array-like + Individual weights for each trace. Default, same weight for each model. + random_seed : int + Seed for the random number generator. + progressbar : bool + Whether or not to display a progress bar in the command line. The + bar shows the percentage of completion, the sampling speed in + samples per second (SPS), and the estimated remaining time until + completion ("expected time of arrival"; ETA). + + Returns + ------- + samples : dict + Dictionary with the variables as keys. The values corresponding to the + posterior predictive samples from the weighted models. + """ + seed(random_seed) + + if models is None: + models = [modelcontext(models)] * len(traces) + + if weights is None: + weights = [1] * len(traces) + + if len(traces) != len(weights): + raise ValueError('The number of traces and weights should be the same') + + if len(models) != len(weights): + raise ValueError('The number of models and weights should be the same') + + lenght_morv = len(models[0].observed_RVs) + if not all(len(i.observed_RVs) == lenght_morv for i in models): + raise ValueError( + 'The number of observed RVs should be the same for all models') + + weights = np.asarray(weights) + p = weights / np.sum(weights) + + min_tr = min([len(i) for i in traces]) + + n = (min_tr * p).astype('int') + # ensure n sum up to min_tr + idx = np.argmax(n) + n[idx] = n[idx] + min_tr - np.sum(n) + + trace = np.concatenate([np.random.choice(traces[i], j) + for i, j in enumerate(n)]) + + variables = [] + for i, m in enumerate(models): + variables.extend(m.observed_RVs * n[i]) len_trace = len(trace) @@ -561,9 +632,8 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, ppc = defaultdict(list) for idx in indices: param = trace[idx] - for var in vars: - ppc[var.name].append(var.distribution.random(point=param, - size=size)) + var = variables[idx] + ppc[var.name].append(var.distribution.random(point=param, size=size)) return {k: np.asarray(v) for k, v in ppc.items()} From dad7d7f2bb33ed50875440cf3dfbd800d30a2297 Mon Sep 17 00:00:00 2001 From: aloctavodia Date: Thu, 8 Jun 2017 13:02:39 +0200 Subject: [PATCH 3/5] add example, fix minor issues --- docs/source/examples.rst | 2 + docs/source/notebooks/model_averaging.ipynb | 495 ++++++++++++++++++ ...omparison.ipynb => model_comparison.ipynb} | 0 pymc3/examples/data/milk.csv | 18 + pymc3/sampling.py | 39 +- 5 files changed, 543 insertions(+), 11 deletions(-) create mode 100644 docs/source/notebooks/model_averaging.ipynb rename docs/source/notebooks/{Model Comparison.ipynb => model_comparison.ipynb} (100%) create mode 100644 pymc3/examples/data/milk.csv diff --git a/docs/source/examples.rst b/docs/source/examples.rst index 3dfa13cc29..f5614180ee 100644 --- a/docs/source/examples.rst +++ b/docs/source/examples.rst @@ -12,6 +12,8 @@ Howto notebooks/sampler-stats.ipynb notebooks/Diagnosing_biased_Inference_with_Divergences.ipynb notebooks/posterior_predictive.ipynb + notebooks/model_comparison.ipynb + notebooks/model_averaging.ipynb notebooks/howto_debugging.ipynb notebooks/PyMC3_tips_and_heuristic.ipynb notebooks/LKJ.ipynb diff --git a/docs/source/notebooks/model_averaging.ipynb b/docs/source/notebooks/model_averaging.ipynb new file mode 100644 index 0000000000..ff43209701 --- /dev/null +++ b/docs/source/notebooks/model_averaging.ipynb @@ -0,0 +1,495 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "%matplotlib inline\n", + "import matplotlib.pyplot as plt\n", + "plt.style.use(['seaborn-darkgrid', 'seaborn-colorblind'])\n", + "import pymc3 as pm\n", + "import numpy as np\n", + "import pandas as pd\n", + "from theano import shared" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "When confronted with more than one model we have several options. One of them is to perform model selection, using for example a given Information Criterion as exemplified [in this notebook](model_comparsion.ipynb) and this other [example](model_comparsion.ipynb). Model selection is appealing for its simplicity, but we are discarding information about the uncertainty in our models. This is somehow similar to computing the full posterior and then keep just keep a point-estimate like the posterior mean; we may become overconfident of what we really know.\n", + "\n", + "One alternative is to perform model selection but discuss the all the different models together with the computed values of a given Information Criterion. It is important to put all these numbers and tests in the context of our problem so that we and our audience can have a better feeling of the possible limitations and shortcomings of our methods. If you are in the academic world you can use this approach to add elements to the discussion section of a paper, presentation, thesis, and so on.\n", + "\n", + "Yet another approach is to perform model averaging. The idea now is to generate a meta-model (and meta-predictions) using a weighted average of the models. One way to compute these weights is to apply this formula:\n", + "\n", + "$$w_i = \\frac {e^{ \\frac{1}{2} dIC_i }} {\\sum_j^M e^{ - \\frac{1}{2} dIC_j }}$$\n", + "\n", + "Where $dIC_i$ is the difference between the i-esim information criterion value and the lowest one. Remember that the lowest the value of the IC, the better. We can use any information criterion we want to compute a set of weights, but, of course, we cannot mix them. \n", + "\n", + "This formula is a heuristic way to compute the relative probability of each model (given a fixed set of models) from the information criteria values. Look how the denominator is just a normalization term to ensure that the weights sum up to one.\n", + "\n", + "Once we have computed the weights we can use them to get a weighted posterior predictive samples. PyMC3 offers functions to perform these steps in a simple way, so let see them in action using an example.\n", + "\n", + "The following example is taken from the superb book [Statistical Rethinking](http://xcelab.net/rm/statistical-rethinking/) by Richard McElreath. You will find more PyMC3 examples from this book in this [repository](https://github.com/aloctavodia/Statistical-Rethinking-with-Python-and-PyMC3). We are going to explore a simplified version of it. Check the book for the whole example and a more thorough discussion of both, the biological motivation for this problem and a theoretical/practical discussion of using Information Criteria to compare, select and average models.\n", + "\n", + "Briefly, our problem is as follows: We want to explore the composition of milk across several primate species, it is hypothesized that females from species of primates with larger brains produce more _nutritious_ milk (loosely speaking this is done _in order to_ support the development of such big brains). This is an important question for evolutionary biologists and try to give and answer we will use 3 variables, two predictor variables: the proportion of neocortex compare to the total mass of the brain and the logarithm of the body mass of the mothers. And for predicted variable, the kilocalories per gram of milk. With these variables we are going to build 3 different linear models:\n", + " \n", + "1. A model using only the neocortex variable\n", + "2. A model using only the logarithm of the mass variable\n", + "3. A model using both variables\n", + "\n", + "Let start by uploading the data and centering the `neocortex` and `log mass` variables, for better sampling." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
kcal.per.gneocortexlog_mass
00.49-0.123706-0.831353
10.47-0.0307060.158647
20.56-0.0307060.181647
30.890.000294-0.579353
40.920.012294-1.885353
\n", + "
" + ], + "text/plain": [ + " kcal.per.g neocortex log_mass\n", + "0 0.49 -0.123706 -0.831353\n", + "1 0.47 -0.030706 0.158647\n", + "2 0.56 -0.030706 0.181647\n", + "3 0.89 0.000294 -0.579353\n", + "4 0.92 0.012294 -1.885353" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "d = pd.read_csv('../data/milk.csv')\n", + "d.iloc[:,1:] = d.iloc[:,1:] - d.iloc[:,1:].mean()\n", + "d.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now that we have the data we are going to build our first model using only the `neocortex`." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using ADVI...\n", + "Average Loss = 12.469: 8%|▊ | 16237/200000 [00:02<00:27, 6644.09it/s]\n", + "Convergence archived at 16300\n", + "Interrupted at 16,300 [8%]: Average Loss = 28.436\n", + "100%|██████████| 2500/2500 [00:03<00:00, 809.12it/s]\n" + ] + } + ], + "source": [ + "with pm.Model() as model_0:\n", + " alpha = pm.Normal('alpha', mu=0, sd=10)\n", + " beta = pm.Normal('beta', mu=0, sd=10)\n", + " epsilon = pm.HalfNormal('epsilon', 10)\n", + " \n", + " mu = alpha + beta * d['neocortex']\n", + " \n", + " kcal = pm.Normal('kcal', mu=mu, sd=epsilon, observed=d['kcal.per.g'])\n", + " trace_0 = pm.sample(2000)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "For the first model I am going to check the posterior using the `traceplot` function, you can do the same for the other models." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA14AAAGoCAYAAABWjDTfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngUAgEAgEAoFAIHQz/x8LtAeAqiThMwAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "pm.traceplot(trace_0);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The second model is exactly the same as the first one, except we now use the logarithm of the mass" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using ADVI...\n", + "Average Loss = 9.5197: 11%|█ | 21758/200000 [00:03<00:26, 6805.33it/s]\n", + "Convergence archived at 22200\n", + "Interrupted at 22,200 [11%]: Average Loss = 26.56\n", + "100%|██████████| 2500/2500 [00:03<00:00, 781.01it/s]\n" + ] + } + ], + "source": [ + "with pm.Model() as model_1:\n", + " alpha = pm.Normal('alpha', mu=0, sd=10)\n", + " beta = pm.Normal('beta', mu=0, sd=1)\n", + " sigma = pm.HalfNormal('sigma', 10)\n", + " \n", + " mu = alpha + beta * d['log_mass']\n", + " \n", + " kcal = pm.Normal('kcal', mu=mu, sd=sigma, observed=d['kcal.per.g'])\n", + " \n", + " trace_1 = pm.sample(2000)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "And finally the third model using the `neocortex` and `log_mass` variables" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using ADVI...\n", + "Average Loss = 9.7679: 11%|█ | 21786/200000 [00:03<00:25, 7027.79it/s]\n", + "Convergence archived at 21900\n", + "Interrupted at 21,900 [10%]: Average Loss = 26.927\n", + "100%|██████████| 2500/2500 [00:04<00:00, 574.80it/s]\n" + ] + } + ], + "source": [ + "with pm.Model() as model_2:\n", + " alpha = pm.Normal('alpha', mu=0, sd=10)\n", + " beta = pm.Normal('beta', mu=0, sd=1, shape=2)\n", + " sigma = pm.HalfNormal('sigma', 10)\n", + "\n", + " mu = alpha + pm.math.dot(beta, d[['neocortex','log_mass']].T)\n", + "\n", + " kcal = pm.Normal('kcal', mu=mu, sd=sigma, observed=d['kcal.per.g'])\n", + "\n", + " trace_2 = pm.sample(2000)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now that we have sampled the posterior for the 3 models, we are going to use WAIC (Widely applicable information criterion) to compare the 3 models. We can do this using the `compare` function included with PyMC3." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
WAICpWAICdWAICweightSEdSEwarning
2-15.55652.4408800.9515964.774901
1-8.994632.011896.561920.03577264.14110.8899761
0-6.912661.98678.643880.01263163.10313.812020
\n", + "
" + ], + "text/plain": [ + " WAIC pWAIC dWAIC weight SE dSE warning\n", + "2 -15.5565 2.44088 0 0.951596 4.7749 0 1\n", + "1 -8.99463 2.01189 6.56192 0.0357726 4.1411 0.889976 1\n", + "0 -6.91266 1.9867 8.64388 0.0126316 3.1031 3.81202 0" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "traces = [trace_0, trace_1, trace_2]\n", + "models = [model_0, model_1, model_2]\n", + "comp = pm.compare(traces, models)\n", + "comp" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can see that the best model is `model_2`, the one with both predictor variables. Notice the DataFrame is ordered from lowest to highest WAIC (_i.e_ from _better_ to _worst_ model). Check [this notebook](model_comparison.ipynb) for a more detailed discussing on model comparison.\n", + "\n", + "We can also see that we get a column with the relative `weight` for each model (according to the first equation at the beginning of this notebook). This weights can be _vaguely_ interpreted as the probability that each model will make the correct predictions on future data. Of course this interpretation is conditional on the models used to compute the weights, if we add or remove models the weights will change. And also is dependent on the assumptions behind WAIC (or any other Information Criterion used). So try to do not overinterpret these `weights`. \n", + "\n", + "We are going to use these weights to generate predictions based not on a single model but on the weighted set of models. This is one way to perform model averaging. Using PyMC3 we can call the `sample_ppc_w` function as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████| 1000/1000 [00:30<00:00, 32.31it/s]\n" + ] + } + ], + "source": [ + "ppc_w = pm.sample_ppc_w(traces, 1000, models, weights=comp.weight)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We are also going to compute PPCs for the lowest-WAIC model" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████| 1000/1000 [00:36<00:00, 28.42it/s]\n" + ] + } + ], + "source": [ + "ppc_2 = pm.sample_ppc(trace_2, 1000, model_2)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A simple way to compare both kind of predictions is to plot their mean and hpd interval" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAEBCAYAAADbxHY7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBv0fn8ahWZJqNysAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "mean_w = ppc_w['kcal'].mean()\n", + "hpd_w = pm.hpd(ppc_w['kcal']).mean(0)\n", + "\n", + "mean = ppc_2['kcal'].mean()\n", + "hpd = pm.hpd(ppc_2['kcal']).mean(0)\n", + "\n", + "plt.errorbar(mean, 1, xerr=[[mean - hpd]], fmt='o', label='model 2')\n", + "plt.errorbar(mean_w, 0, xerr=[[mean_w - hpd_w]], fmt='o', label='weighted models')\n", + "\n", + "plt.yticks([])\n", + "plt.ylim(-2, 3)\n", + "plt.xlabel('kcal per g')\n", + "plt.legend();" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As we can see the mean value is almost the same for both predictions but the uncertainty in the weighted model is larger. We have effectively propagated the uncertainty about which model we should select to the posterior predictive samples.\n", + "\n", + "**Final notes:** \n", + "\n", + "There are other ways to average models such as, for example, explicitly building a meta-model that includes all the models we have. We then perform parameter inference while jumping between the models. One problem with this approach is that jumping between models could hamper the proper sampling of the posterior.\n", + "\n", + "Besides averaging discrete models we can sometimes think of continuous versions of them. A toy example is to imagine that we have a coin and we want to estimated it's degree of bias, a number between 0 and 1 being 0.5 equal chance of head and tails. We could think of two separated models one with a prior biased towards heads and one towards tails. We could fit both separate models and then average them using, for example, IC-derived weights. As an alternative, is to build a hierarchical model to estimate the prior distribution, instead of contemplating two discrete models we will be computing a continuous model that includes these two both discrete models as particular cases. Which approach is better? That depends on our concrete problem. Do we have good reasons to think about two discrete models, or is our problem better represented with a continuous bigger model?" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.5.3" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/source/notebooks/Model Comparison.ipynb b/docs/source/notebooks/model_comparison.ipynb similarity index 100% rename from docs/source/notebooks/Model Comparison.ipynb rename to docs/source/notebooks/model_comparison.ipynb diff --git a/pymc3/examples/data/milk.csv b/pymc3/examples/data/milk.csv new file mode 100644 index 0000000000..c0b8fbb832 --- /dev/null +++ b/pymc3/examples/data/milk.csv @@ -0,0 +1,18 @@ +kcal.per.g,neocortex,log_mass +0.490,0.552,0.668 +0.470,0.645,1.658 +0.560,0.645,1.681 +0.890,0.676,0.920 +0.920,0.688,-0.386 +0.800,0.589,-2.120 +0.460,0.617,-0.755 +0.710,0.603,-1.139 +0.680,0.700,0.438 +0.970,0.704,1.176 +0.840,0.734,2.510 +0.620,0.675,1.681 +0.540,0.713,3.569 +0.490,0.726,4.375 +0.480,0.702,3.707 +0.550,0.763,3.500 +0.710,0.755,4.006 diff --git a/pymc3/sampling.py b/pymc3/sampling.py index 8656cef0e1..c81ffe7893 100644 --- a/pymc3/sampling.py +++ b/pymc3/sampling.py @@ -534,12 +534,20 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, if progressbar: indices = tqdm(indices, total=samples) - ppc = defaultdict(list) - for idx in indices: - param = trace[idx] - for var in vars: - ppc[var.name].append(var.distribution.random(point=param, - size=size)) + try: + ppc = defaultdict(list) + for idx in indices: + param = trace[idx] + for var in vars: + ppc[var.name].append(var.distribution.random(point=param, + size=size)) + + except KeyboardInterrupt: + pass + + finally: + if progressbar: + indices.close() return {k: np.asarray(v) for k, v in ppc.items()} @@ -629,11 +637,20 @@ def sample_ppc_w(traces, samples=None, models=None, size=None, weights=None, if progressbar: indices = tqdm(indices, total=samples) - ppc = defaultdict(list) - for idx in indices: - param = trace[idx] - var = variables[idx] - ppc[var.name].append(var.distribution.random(point=param, size=size)) + try: + ppc = defaultdict(list) + for idx in indices: + param = trace[idx] + var = variables[idx] + ppc[var.name].append(var.distribution.random(point=param, + size=size)) + + except KeyboardInterrupt: + pass + + finally: + if progressbar: + indices.close() return {k: np.asarray(v) for k, v in ppc.items()} From db0da223cf80ed54013f7bd037587ff6a16dea14 Mon Sep 17 00:00:00 2001 From: Osvaldo Martin Date: Sun, 11 Jun 2017 18:51:39 +0200 Subject: [PATCH 4/5] Create sampling.py fix error docstring --- pymc3/sampling.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/pymc3/sampling.py b/pymc3/sampling.py index c81ffe7893..2d2df24ab3 100644 --- a/pymc3/sampling.py +++ b/pymc3/sampling.py @@ -516,8 +516,7 @@ def sample_ppc(trace, samples=None, model=None, vars=None, size=None, ------- samples : dict Dictionary with the variables as keys. The values corresponding to the - posterior predictive samples. If a set of weights and a matching number - of traces are provided, then the samples will be weighted. + posterior predictive samples. """ if samples is None: samples = len(trace) From e282df28d322c5f3e5bd92d83019c9926414cf78 Mon Sep 17 00:00:00 2001 From: aloctavodia Date: Tue, 27 Jun 2017 09:54:33 +0200 Subject: [PATCH 5/5] rename --- docs/source/notebooks/model_comparison.ipynb | 230 ++++++++++--------- 1 file changed, 116 insertions(+), 114 deletions(-) diff --git a/docs/source/notebooks/model_comparison.ipynb b/docs/source/notebooks/model_comparison.ipynb index 32f9bc4ebf..6f428f120d 100644 --- a/docs/source/notebooks/model_comparison.ipynb +++ b/docs/source/notebooks/model_comparison.ipynb @@ -12,9 +12,7 @@ { "cell_type": "code", "execution_count": 1, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", @@ -35,9 +33,7 @@ { "cell_type": "code", "execution_count": 2, - "metadata": { - "collapsed": true - }, + "metadata": {}, "outputs": [], "source": [ "J = 8\n", @@ -55,16 +51,18 @@ { "cell_type": "code", "execution_count": 3, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "Average ELBO = -42.09: 100%|██████████| 200000/200000 [00:17<00:00, 11675.35it/s] , 9446.54it/s]\n", - "100%|██████████| 2000/2000 [00:01<00:00, 1747.24it/s]\n" + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using ADVI...\n", + "Average Loss = 43.342: 4%|▍ | 7754/200000 [00:00<00:12, 14950.83it/s]\n", + "Convergence archived at 9400\n", + "Interrupted at 9,400 [4%]: Average Loss = 43.902\n", + "100%|██████████| 1500/1500 [00:00<00:00, 2870.72it/s]\n" ] } ], @@ -74,21 +72,19 @@ " \n", " obs = pm.Normal('obs', mu, sd=sigma, observed=y)\n", " \n", - " trace_p = pm.sample(2000)" + " trace_p = pm.sample(1000)" ] }, { "cell_type": "code", "execution_count": 4, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAACCCAYAAABxYpjTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAACECAYAAABiSeFfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -109,16 +105,20 @@ { "cell_type": "code", "execution_count": 5, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "Average ELBO = -43.175: 100%|██████████| 200000/200000 [00:23<00:00, 8435.73it/s]3, 8478.86it/s]\n", - "100%|██████████| 2000/2000 [00:03<00:00, 477.25it/s]\n" + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using ADVI...\n", + "Average Loss = 43.29: 11%|█▏ | 22935/200000 [00:02<00:14, 12516.51it/s] \n", + "Convergence archived at 23000\n", + "Interrupted at 23,000 [11%]: Average Loss = 44.107\n", + " 94%|█████████▍| 1412/1500 [00:01<00:00, 1234.30it/s]/Users/fonnescj/Repos/pymc3/pymc3/step_methods/hmc/nuts.py:456: UserWarning: Chain 0 contains 1 diverging samples after tuning. If increasing `target_accept` does not help try to reparameterize.\n", + " % (self._chain_id, n_diverging))\n", + "100%|██████████| 1500/1500 [00:01<00:00, 1162.34it/s]\n" ] } ], @@ -133,21 +133,19 @@ " \n", " obs = pm.Normal('obs', theta, sd=sigma, observed=y)\n", " \n", - " trace_h = pm.sample(2000)" + " trace_h = pm.sample(1000)" ] }, { "cell_type": "code", "execution_count": 6, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAACCCAYAAAC0EAJCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIYQQQogd0SCLEEIIIYQQQuzo/wDXvtrR6JVm9gAAAABJRU5ErkJggg==\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAACECAYAAABiSeFfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyCKEEEIIIYQQL/p/ZWq7Xso5Sg0AAAAASUVORK5CYII=\n", "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -161,15 +159,13 @@ { "cell_type": "code", "execution_count": 7, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAg8AAAF8CAYAAABIe1hQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhYAAAF8CAYAAACTwPJyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAADgNxQLAADgNxQLAADgNxQLAADgN/8Pe/KaAGscqGIAAAAASUVORK5CYII=\n", "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -192,14 +188,12 @@ { "cell_type": "code", "execution_count": 8, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "90.832077760613302" + "90.882480923881175" ] }, "execution_count": 8, @@ -216,14 +210,12 @@ { "cell_type": "code", "execution_count": 9, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "124.82406598767808" + "124.2888830531669" ] }, "execution_count": 9, @@ -249,14 +241,12 @@ { "cell_type": "code", "execution_count": 10, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "61.121151094964134" + "61.136455281870752" ] }, "execution_count": 10, @@ -273,14 +263,12 @@ { "cell_type": "code", "execution_count": 11, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "61.570270537072844" + "61.350518836043364" ] }, "execution_count": 11, @@ -304,14 +292,25 @@ { "cell_type": "code", "execution_count": 12, - "metadata": { - "collapsed": false - }, + "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", + "\n", "\n", " \n", " \n", @@ -328,22 +327,22 @@ " \n", " \n", " \n", - " \n", - " \n", + " \n", + " \n", " \n", - " \n", - " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -352,8 +351,8 @@ ], "text/plain": [ " WAIC pWAIC dWAIC weight SE dSE warning\n", - "1 61.1212 0.674471 0 0.555905 2.20084 0 0\n", - "0 61.5703 1.08163 0.449119 0.444095 1.96714 0.0259316 0" + "1 61.1365 0.676001 0 0.526732 2.20685 0 0\n", + "0 61.3505 0.985125 0.214064 0.473268 1.94631 0.0389335 0" ] }, "execution_count": 12, @@ -397,16 +396,14 @@ }, { "cell_type": "code", - "execution_count": 13, - "metadata": { - "collapsed": false - }, + "execution_count": 14, + "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAFXCAYAAADu/TSqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAFqlJREFUeJzt3X90lnX9x/HXvY2bDaZfVo7CgAHlQA9HFGJGuVYZkS3g\nDBScxY7K18NRQwwUtQ7mKQ6OlRrxB2R0MhAXc56ZBAsOcTwjBISxgfwYRsE8y8TFOMVuRtvuXd8/\n+DoD9gvYe/fnM5+Pv7jvm933mw83vPa5dr3uKxQEQSAAANDt4mI9AAAAvRUhCwCAEUIWAAAjhCwA\nAEYIWQAAjBCyAAAYSejuJ6ytPd3dT9mulJR+OnXqTI+93scN62urq+u7Zs0qSdKsWf9rPVKvwvvX\nFuv7kdTUq9p9zOudbEJCfKxH6NVYX1usry3W1xbr2zVehywAAC7r9sPFALoXh4kBf7GTBQDACCEL\nOG7Hjm3asWNbrMcAcBkIWcBxR48e0dGjR2I9BoDLQMgCAGCEkAUAwAhnFwMOa2ho0N695Tp58qT6\n9bta2dlTlJiYGOuxAHRRl3ay+/bt06xZs6xnwWUqKSlWVtYEDRqUoqysCSopKY71SOgGFRXlysgY\noz179qi5uVlFRYUaP/5GVVSUx3o0AF3U6U72V7/6lV5//XUlJSX1xDy4RCUlxZoz577W24cPH2y9\nnZNzR6zGwhVqaGhQXl6uCgqe1+23Z7feX1q6QXl5udq9ez87WsADoSAIgo5+w6ZNmzRy5EgtXLhQ\nRUVFnT5hd3928bhxo9t9LC4upJaWDsfv9d5//x9qamq66P4+ffro058edEXPzfra6mh9I5GIzpyJ\nKDV14EWP1dZ+oH79+qt///7WI3qN96+tnl7f8vIDPfZal6qjzy7udCc7adIk1dTUdPnFUlL6detn\nWsbFha7o8d6urYD98P7uWJuP+/paa299o9FmhcNhxcWFNG7cOElSefm5w8ThcFjRaDN/N13AGtnq\nyfXtKMhc1u0nPnX3VRl273673cdSU6/q0av+uCgra4IOHz540f033DBab7zx5hU9N+trq6P1ffXV\nIhUVFWrdupLWq/CsXPmiJGnmzBzNmJGr6dNn9NSoXuL9a6un19flv8teexUeSI88sqDN++fNm9/D\nk6A7ZWdP0aFDB1VauuG8+0tLN+jQoYPKzp4So8kAXAoqPJ778OSmZcue0zvvVCk9fZTmzZvPSU+e\nS0xM1OrVhcrLy9WAAf+jwYOH6A9/OBewq1cXctIT4IlOT3y6VD25pedwkC3W11ZX1rehoUE/+MF8\nnTxZp6lTp9OTvQS8f22xvh+5ohOfAMROUlKSxo49d+ITP4MF/EPIAo7jerKAvzjxCQAAI4Qs4Diu\nJwv4i5AFHMf1ZAF/EbIAABghZAEAMELIAgBghJAFAMAIPVnAcfRkAX+xkwUAwAghCziOnizgL0IW\ncBw9WcBfhCwAAEYIWQAAjBCyAAAYIWQBADBCTxZwHD1ZwF/sZAEAMELIAo6jJwv4i5AFHEdPFvAX\nIQsAgBFCFgAAI4QsAABGCFkAAIzQkwUcR08W8Bc7WQAAjBCygOPoyQL+ImQBx9GTBfxFyAIAYISQ\nBQDACCELAIARQhYAACP0ZAHH0ZMF/MVOFgAAI4Qs4Dh6soC/CFnAcfRkAX8RsgAAGCFkAQAwQsgC\nAGCEkAUAwAg9WcBx9GQBf7GTBQDACCELOI6eLOAvQhZwHD1ZwF+ELAAARghZAACMELIAABghZAEA\nMEJPFnAcPVnAX+xkAQAwQsgCjqMnC/iLkAUcR08W8BchCwCAEUIWAAAjhCwAAEYIWQAAjNCTBRxH\nTxbwFztZAACMELKA4+jJAv4iZAHH0ZMF/EXIAgBghJAFAMAIIQsAgBFCFgAAI/RkAcfRkwX8xU4W\nAAAjhCzgOHqygL8IWcBx9GQBfxGyAAAYIWQBADBCyAIAYISQBQDACD1ZwHH0ZAF/sZMFAMAIIQs4\njp4s4C9CFnAcPVnAX4QsAABGCFkAAIwQsgAAGCFkAQAwQk8WcBw9WcBf7GQBADBCyAKOoycL+IuQ\nBRxHTxbwFyELAIARQhYAACOELAAARghZAACM0JMFHEdPFvAXO1kAAIwQsoDj6MkC/iJkAcfRkwX8\nRcgCAGCEkAUAwAghCwCAEUIWAAAj9GQBx9GTBfzFThYAACOELOA4erKAvwhZwHH0ZAF/EbIAABgh\nZAEAMELIAgBghJAFAMAIPVnAcfRkAX+xkwUAwAghCziOnizgL0IWcBw9WcBfhCwAAEYIWQAAjBCy\nAAAYIWQBADBCTxZwHD1ZwF/sZAEAMELIAo6jJwv4i5AFHEdPFvAXIQsAgBFCFgAAI4QsAABGCFkA\nAIzQkwUcR08W8Bc7WQAAjBCygOPoyQL+ImQBx9GTBfxFyAIAYISQBQDACCELAICRTis8LS0tevrp\np3XkyBGFw2EtXrxYaWlpPTEbAMedOPEPnT17Vmlpw2M9CuCkTkN2y5Ytamxs1Lp161RZWan8/Hyt\nWLGiJ2YDILd7svv3V+if//xAQ4akKS6OA2PAhTr9V1FeXq7MzExJ0k033aQDBw6YDwXAfb/97Qv6\ny1+qdOpUnX7zm5WxHgdwUqc72fr6eiUnJ7fejo+PV3NzsxIS2v7SlJR+SkiI774JO5GaelWPvdbH\nEetrqyvru2XLFknS17/+detxLklzc1Prrxsb/6Pk5AQlJSXFcKKL8f61xfp2rtOQTU5OViQSab3d\n0tLSbsBK0qlTZ7pnsi5ITb1KtbWne+z1Pm5YX1tdXd/9+9+WJI0Zc4v1SF0WjUYVH58g6T+SpCAI\nVFq6WZmZX4vtYP+F968t1vcjHX2z0enh4rFjx6qsrEySVFlZqfT09O6bDICXjh49ojNnIufdd/Dg\nfp06VRejiQA3dRqyEydOVDgc1l133aVnnnlGTz75ZE/MBcBhb79dedF9LS0t2r79jZ4fBnBYp4eL\n4+Li9OMf/7gnZgHgiTvuuDvWIwBe4Jx7AACMcD1ZwHEu92QBdIydLAAARghZwHFcTxbwFyELOI7r\nyQL+ImQBADBCyAIAYISQBQDACCELAIARerKA4+jJAv5iJwsAgBFCFnAcPVnAX4Qs4Dh6soC/CFkA\nAIwQsgAAGCFkAQAwQsgCAGCEnizgOHqygL/YyQIAYISQBRxHTxbwFyELOI6eLOAvQhYAACOELAAA\nRghZAACMELIAABihJws4jp4s4C92sgAAGCFkAcfRkwX8RcgCjqMnC/iLkAUAwAghCwCAEUIWAAAj\nhCwAAEboyQKOoycL+IudLAAARghZwHH0ZAF/EbKA4+jJAv4iZAEAMELIAgBghJAFAMAIIQsAgBF6\nsoDj6MkC/mInCwCAEUIWcBw9WcBfhCzgOHqygL8IWQAAjBCyAAAYIWQBADBCyAIAYISeLOA4erKA\nv9jJAgBghJAFHEdPFvAXIQs4jp4s4C9CFgAAI4QsAABGCFkAAIwQsgAAGKEnCziOnizgL3ayAAAY\nIWQBx9GTBfxFyAKOoycL+IuQBQDACCELAIARQhYAACOELAAARujJAo6jJwv4i50sAABGCFnAcfRk\nAX8RsoDj6MkC/iJkAQAwQsgCAGCEkAUAwAghCwCAEXqygOPoyQL+YicLAIARQhZwHD1ZwF+ELOA4\nerKAvwhZAACMELIAABghZAEAMELIAgBghJ4s4Dh6soC/2MkCAGCEkAUcR08W8BchCziOnizgL0IW\nAAAjhCwAAEYIWQAAjBCyAAAYoScLOI6eLOAvdrIAABghZAHH0ZMF/EXIAo6jJwv4i5AFAMAIIQsA\ngBFCFgAAI4QsAABG6MkCjqMnC/iLnSwAAEYIWcBx9GQBfxGygOPoyQL+ImQBADBCyAIAYISQBQDA\nCCELAIARerKA4+jJAv5iJwsAgBFCFnAcPVnAX4Qs4Dh6soC/CFkAAIwQsgAAGCFkAQAwQsgCAGCE\nnizgOHqygL/YyQIAYISQBRxHTxbwFyELOI6eLOAvQhYAACOELAAARghZAACMELIAABihJws4jp4s\n4C92sgAAGCFkAcfRkwX8RcgCjqMnC/iLkAUAwAghCwCAEc4uBnBZGhoatHHjeh0/fkzDhg1XdvYU\nJSYmxnoswCmd7mRbWlr01FNPaebMmZo1a5aqq6t7Yi6vlJQUKytrggYNSlFW1gSVlBTHeiTAVEVF\nuTIyxqioqFCNjf9RUVGhxo+/URUV5bEeDXBKpzvZLVu2qLGxUevWrVNlZaXy8/O1YsWKnpjNCyUl\nxZoz577W24cPH2y9nZNzR6zGQi/iWk+2oaFBeXm5Kih4Xrffnt16f2npBuXl5Wr37v3saIH/FwqC\nIOjoNzzzzDO68cYblZ197h9TZmamtm1rv05QW3u6eyfsQGrqVT36em3Jypqgw4cPXnT/DTeM1htv\nvHnJzzdu3OjuGKtbxMWF1NLS4dsDV8DX9Y1EIjpzJqLU1IEXPVZb+4H69euv/v37x2Cy8/m6vr7w\neX3Lyw906/Olpl7V7mOd7mTr6+uVnJzcejs+Pl7Nzc1KSGj7S1NS+ikhIf4yxrw8Hf3hesI771S1\ne//lzBYXF7rSkbqVa/P0Nl1Z33HjxkmSysvdOBQbjTYrHA63OXs4HFY02uzM+8aVOXorX9e3J3Oj\n05BNTk5WJBJpvd3S0tJuwErSqVNnumeyLnBhJ5uePqrNnWx6+qjLmm337re7Y6xu4cL69mZdXd81\na1ZJklaufNF4oq559dUiFRUVat26kosemzkzRzNm5Gr69BkxmOx8vH9t+by+3T13R6Hd6YlPY8eO\nVVlZmSSpsrJS6enp3TdZL/DIIwvavH/evPk9PAnQM7Kzp+jQoYMqLd1w3v2lpRt06NBBZWdPidFk\ngHs63clOnDhR27dv11133aUgCLRkyZKemMsbH57ctGzZc3rnnSqlp4/SvHnzOekJvVZiYqJWry5U\nXl6uXnxxlcaMuVn79lXo0KGDWr26kJOegP/S6YlPl+rjduJTb8b62rrUw8UunmW8ceN6VVcfV1ra\nMOd6srx/bbG+H7miE58AoC1JSUlO/OwVcBkhCzjOtR0sgK7js4sBADBCyAKO43qygL8IWcBxXE8W\n8BchCwCAEUIWAAAjhCwAAEYIWQAAjHT7Jz4BAIBz2MkCAGCEkAUAwAghCwCAEUIWAAAjhCwAAEYI\nWQAAjHh1qbucnBwlJydLkgYPHqzs7Gz97Gc/U1JSkjIzM/Xggw/GeEK//fKXv9TWrVvV1NSk3Nxc\nZWRk6IknnlAoFNJ1112nH/3oR4qL4/uyy3Xh+t55552SpCVLlmj48OHKzc2N8YR+u3B9R48erZ/8\n5CeKj49XOBzW0qVLdc0118R6TC9duLY333yzFi1apCAINGzYMC1evFgJCV7FSc8JPHH27Nlg6tSp\nrbej0WiQlZUVvPvuu0EQBMGCBQuC3bt3x2o87+3cuTOYM2dOEI1Gg/r6+uAXv/hFMGfOnGDnzp1B\nEATBokWLgs2bN8d4Sn+1tb4nT54MZs+eHdx2223Byy+/HOsRvdbW+n7nO98JDh06FARBEBQWFgZL\nliyJ8ZR+amttH3jggeCtt94KgiAIHn/8cf5v6IA333pUVVWpoaFB9913n5qbmzV37lxdffXVGjJk\niCRp7Nix2rt3rz7/+c/HeFI//fnPf1Z6eroeeugh1dfXa+HChSoqKlJGRoYk6ctf/rK2b9+uiRMn\nxnhSP7W1vpFIRHPnzlVZWVmsx/NeW+s7c+ZMDRw4UJIUjUbVt2/fGE/pp7bW9sEHH1R8fLwaGxtV\nW1vbeoQRF/MmZBMTEzV79mzdeeedOn78uO6//361tLTor3/9q4YNG6aysjKNGjUq1mN669SpU3rv\nvfe0cuVK1dTU6IEHHlAQBAqFQpKk/v376/Tp0zGe0l9tre8f//hHDRkyhJDtBu2tryTt3btXL730\nktauXRvjKf3U3tr+/e9/17333qvk5GT+7+2ANyE7fPhwpaWlKRQKafjw4RowYICeeOIJPf300wqH\nw0pPT1dKSkqsx/TWgAEDNGLECIXDYY0YMUJ9+/bV+++/3/p4JBLR1VdfHcMJ/dbW+tbV1emTn/xk\nrEfrFdpb3127dmnFihV64YUX9IlPfCLWY3qpvbX9zGc+o82bN+uVV15Rfn6+li5dGutRneTNWSzF\nxcXKz8+XJJ04cUL19fXauXOnfv3rX2vVqlV699139cUvfjHGU/pr3Lhx2rZtm4Ig0IkTJ9TQ0KAJ\nEyZo165dkqSysjIOxV+BttZ3wIABsR6r12hrfcvKyvTSSy9pzZo1rT9WwqVra21/+MMf6vjx45LO\nHeXihMj2eXOBgMbGRj355JN67733FAqF9Oijj+ro0aNau3atEhMTNXnyZH33u9+N9ZheKygo0K5d\nuxQEgb7//e9r8ODBWrRokZqamjRixAgtXrxY8fHxsR7TWxeub2ZmpiRp+fLluuaaazi7+ApduL4L\nFizQoEGDWo/AjB8/Xg8//HCMp/TThWvbv39/FRQUqE+fPkpKStLixYtbf/6N83kTsgAA+IY9PgAA\nRghZAACMELIAABghZAEAMELIAgBgxJsPowB6i5qaGn3zm9/UZz/7WUnS2bNnNXLkSD311FOX/AH2\ny5Yt0+jRo3XbbbdZjArgClHhAXpYTU2N8vLytHXrVklSEAR67rnnVF5erpdffjnG0wHoTuxkgRgL\nhUKaO3euvvSlL6mqqkplZWUqLS1VNBrVrbfeqscee0z5+fkaOHCgZs+eLUl6+OGH9e1vf1tbt25V\nRkaGpk2bpueff147duzQv/71L6WkpGj58uVKTU3VrbfeqkmTJqm8vFzx8fH6+c9/riFDhujNN99U\nfn6+giDQtddeq2effVZJSUkqKCjQW2+9pWg0qmnTpumee+6J7QIBHuNnsoADwuGw0tLSVFVVpQMH\nDqi4uFivvfaaTpw4oddff11Tp07Vhg0bJEn19fXau3evvvKVr7R+fXV1tf72t7/pd7/7nTZt2qSh\nQ4dq/fr1kqTa2lpNmDBBr732msaPH6+1a9eqsbFRjz76qJYuXar169dr5MiRKikpUVFRkSSppKRE\nxcXF+tOf/qQ9e/b0+HoAvQU7WcARoVBIq1evVl1dnaZNmybp3M9rr732Wk2dOlWNjY2qrq5WRUWF\nvvrVryocDrd+bVpamh5//HG98sorOnbsmCorKzV06NDWxz/8CMfrrrtOe/bs0ZEjR/SpT31K119/\nvSRp/vz5ks7tkA8fPqydO3dKks6cOaMjR47wudXAZSJkAQc0Njbq2LFjuuWWWzR58mTde++9kqR/\n//vfrZ8XPWXKFG3cuFEVFRW6//77z/v6AwcOaMGCBbrnnns0adIkxcXF6b9Pt/jwWqqhUEhBEKhP\nnz7nff3p06cViUQUjUb12GOP6Rvf+IYkqa6uTv369TP7cwO9HYeLgRhraWnR8uXLNWbMGE2fPl2/\n//3vFYlE1NzcrIceekibNm2SJE2ePFkbN25UdXX1RTvL3bt3KyMjQ7m5ufrc5z6n7du3KxqNtvua\nw4cPV11dnY4ePSpJWrVqlQoLC/WFL3xBRUVFampqUiQS0d133619+/bZ/eGBXo6dLBADH3zwgaZO\nnSrpXMhef/31evbZZzVgwABVVVVpxowZikajyszMVE5OjiRp0KBBSklJ0U033aRQKHTe833rW9/S\n9773PU2ePFl9+vTRyJEjVVNT0+7r9+3bVz/96U+1cOFCNTU1aejQoSooKFA4HFZ1dbVycnLU3Nys\nadOm6ZZbbrFbCKCXo8IDAIARDhcDAGCEkAUAwAghCwCAEUIWAAAjhCwAAEYIWQAAjBCyAAAYIWQB\nADDyf2IvK/9TovU3AAAAAElFTkSuQmCC\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAFXCAYAAADu/TSqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWWPl/+MUipIAAAAASUVORK5CYII=\n", "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -414,7 +411,7 @@ } ], "source": [ - "pm.compare_plot(df_comp_WAIC);" + "pm.compareplot(df_comp_WAIC);" ] }, { @@ -441,16 +438,14 @@ }, { "cell_type": "code", - "execution_count": 14, - "metadata": { - "collapsed": false - }, + "execution_count": 15, + "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "/home/osvaldo/anaconda3/lib/python3.5/site-packages/pymc3/stats.py:208: UserWarning: Estimated shape parameter of Pareto distribution is\n", + "/Users/fonnescj/Repos/pymc3/pymc3/stats.py:255: UserWarning: Estimated shape parameter of Pareto distribution is\n", " greater than 0.7 for one or more samples.\n", " You should consider using a more robust model, this is\n", " because importance sampling is less likely to work well if the marginal\n", @@ -462,10 +457,10 @@ { "data": { "text/plain": [ - "61.591020562529295" + "61.555247474354104" ] }, - "execution_count": 14, + "execution_count": 15, "metadata": {}, "output_type": "execute_result" } @@ -478,16 +473,14 @@ }, { "cell_type": "code", - "execution_count": 15, - "metadata": { - "collapsed": false - }, + "execution_count": 16, + "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "/home/osvaldo/anaconda3/lib/python3.5/site-packages/pymc3/stats.py:208: UserWarning: Estimated shape parameter of Pareto distribution is\n", + "/Users/fonnescj/Repos/pymc3/pymc3/stats.py:255: UserWarning: Estimated shape parameter of Pareto distribution is\n", " greater than 0.7 for one or more samples.\n", " You should consider using a more robust model, this is\n", " because importance sampling is less likely to work well if the marginal\n", @@ -499,10 +492,10 @@ { "data": { "text/plain": [ - "61.655191059793324" + "61.413990925656876" ] }, - "execution_count": 15, + "execution_count": 16, "metadata": {}, "output_type": "execute_result" } @@ -522,15 +515,26 @@ }, { "cell_type": "code", - "execution_count": 16, - "metadata": { - "collapsed": false - }, + "execution_count": 17, + "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", + "\n", "
161.12120.67447161.13650.67600100.5559052.200840.5267322.2068500
061.57031.081630.4491190.4440951.967140.025931661.35050.9851250.2140640.4732681.946310.03893350
\n", " \n", " \n", @@ -546,23 +550,23 @@ " \n", " \n", " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", " \n", - " \n", - " \n", + " \n", + " \n", " \n", " \n", " \n", " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", " \n", " \n", " \n", @@ -570,12 +574,12 @@ "" ], "text/plain": [ - " LOO pLOO dLOO weight SE dSE warning\n", - "1 61.591 0.909406 0 0.508021 2.13514 0 1\n", - "0 61.6552 1.12409 0.0641705 0.491979 1.9971 0.0258501 1" + " LOO pLOO dLOO weight SE dSE warning\n", + "0 61.414 1.01686 0 0.51765 1.95466 0 1\n", + "1 61.5552 0.885397 0.141257 0.48235 2.1805 0.0347903 1" ] }, - "execution_count": 16, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -596,16 +600,14 @@ }, { "cell_type": "code", - "execution_count": 17, - "metadata": { - "collapsed": false - }, + "execution_count": 18, + "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAFXCAYAAADu/TSqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9H5Up+UAAAAASUVORK5CYII=\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAFXCAYAAADu/TSqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n", "text/plain": [ - "" + "" ] }, "metadata": {}, @@ -613,7 +615,7 @@ } ], "source": [ - "pm.compare_plot(df_comp_LOO);" + "pm.compareplot(df_comp_LOO);" ] }, { @@ -648,9 +650,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.5.2" + "version": "3.6.1" } }, "nbformat": 4, - "nbformat_minor": 0 + "nbformat_minor": 1 }
161.5910.909406061.4141.0168600.5080212.135140.517651.9546601
061.65521.124090.06417050.4919791.99710.0258501161.55520.8853970.1412570.482352.18050.03479031