GitHub Repository: tensorflow/docs-l10n
Path: blob/master/site/ko/probability/examples/Modeling_with_JointDistribution.ipynb
²⁵¹¹⁸ views

Kernel: Python 3

Copyright 2019 The TensorFlow Authors.

Licensed under the Apache License, Version 2.0 (the "License");

In [ ]:

#@title Licensed under the Apache License, Version 2.0 (the "License"); { display-mode: "form" }
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

공동 분포를 사용한 베이지안 모델링

JointDistributionSequential은 사용자가 베이지안 모델의 프로토타입을 빠르게 만들 수 있도록 새롭게 도입된 분포형 클래스입니다. 여러 분포를 함께 연결하고 람다 함수를 사용하여 종속성을 도입할 수 있습니다. GLM, 혼합 효과 모델, 혼합 모델 등과 같이 일반적으로 사용되는 많은 모델을 포함하여 중소 규모의 베이지안 모델을 구축하도록 설계되었습니다. 사전 예측 샘플링 등 베이지안 워크플로에 필요한 모든 기능이 지원됩니다. 다른 더 큰 베이지안 그래픽 모델 또는 신경망에 플러그인할 수 있습니다. 이 Colab에서는 일상적인 베이지안 워크플로를 위해 JointDistributionSequential을 사용하는 방법의 몇 가지 예를 보여줍니다.

종속성과 전제 조건

In [ ]:

# We will be using ArviZ, a multi-backend Bayesian diagnosis and plotting library
!pip3 install -q git+git://github.com/arviz-devs/arviz.git

In [ ]:

#@title Import and set ups{ display-mode: "form" }

from pprint import pprint
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
import pandas as pd
import arviz as az

import tensorflow.compat.v2 as tf
tf.enable_v2_behavior()

import tensorflow_probability as tfp

sns.reset_defaults()
#sns.set_style('whitegrid')
#sns.set_context('talk')
sns.set_context(context='talk',font_scale=0.7)

%config InlineBackend.figure_format = 'retina'
%matplotlib inline

tfd = tfp.distributions
tfb = tfp.bijectors

dtype = tf.float64

빠르게 처리하세요!

시작하기 전에 이 데모에 GPU를 사용하고 있는지 확인하겠습니다.

확인하려면 'Runtime'-> 'Change runtime type'-> 'Hardware accelerator'-> 'GPU'를 선택합니다.

다음 코드 조각은 GPU에 대한 액세스 권한이 있는지 확인합니다.

In [ ]:

if tf.test.gpu_device_name() != '/device:GPU:0':
  print('WARNING: GPU device not found.')
else:
  print('SUCCESS: Found GPU: {}'.format(tf.test.gpu_device_name()))

SUCCESS: Found GPU: /device:GPU:0

참고: 어떤 이유로 GPU에 액세스할 수 없는 경우에도 이 colab은 계속 동작합니다(훈련은 더 오래 걸림).

JointDistribution

참고: 이 분포 클래스는 간단한 모델만 있는 경우에 유용합니다. "간단한"이란 사슬 모양의 그래프를 의미합니다. 이 접근 방식은 기술적으로 단일 노드에 대해 차수가 최대 255개인 모든 PGM에 대해 작동합니다(Python 함수는 최대 이 정도의 인수를 가질 수 있기 때문).

기본적 개념은 사용자가 PGM의 모든 정점에 대해 하나씩 tfp.Distribution 인스턴스를 생성하는 callable 목록을 지정하도록 하는 것입니다. callable은 목록에 있는 인덱스만큼 많은 인수를 가질 것입니다. (사용자의 편의를 위해 인수는 생성의 역순으로 전달됩니다.) 내부적으로 모든 이전 RV의 값을 각 callable에 전달하여 단순히 "그래프를 탐색"할 것입니다. 그렇게 하면 [확률의 연쇄 법칙]이 구현됩니다(https://en.wikipedia.org/wiki/Chain_rule_(probability)#More_than_two_random_variables): ParseError: KaTeX parse error: Expected '}', got '&' at position 33: …_i^dp( x_i|x* {&̲lt;i}).

아이디어는 Python 코드처럼 매우 간단합니다. 요지는 다음과 같습니다.

# The chain rule of probability, manifest as Python code.
def log_prob(rvs, xs):
  # xs[:i] is rv[i]'s markov blanket. `[::-1]` just reverses the list.
  return sum(rv(*xs[i-1::-1]).log_prob(xs[i])
             for i, rv in enumerate(rvs))

JointDistributionSequential의 docstring에서 더 많은 정보를 찾을 수 있지만 요지는 목록의 일부 분포가 다른 업스트림 분포/변수의 출력에 의존하는 경우 클래스를 초기화하기 위해 분포 목록을 전달한다는 것입니다. 람다 함수로 래핑하면 됩니다. 이제 어떻게 작동하는지 봅시다!

(강력한) 선형 회귀

PyMC3 문서 GLM: 이상값 감지를 사용한 강력한 회귀의 내용

In [ ]:

#@title Get data
#@markdown Cut & pasted directly from the `fetch_hogg2010test()` function. It is identical to the original dataset as hardcoded in the Hogg 2010 paper
dfhogg = pd.DataFrame(np.array([[1, 201, 592, 61, 9, -0.84],
                                 [2, 244, 401, 25, 4, 0.31],
                                 [3, 47, 583, 38, 11, 0.64],
                                 [4, 287, 402, 15, 7, -0.27],
                                 [5, 203, 495, 21, 5, -0.33],
                                 [6, 58, 173, 15, 9, 0.67],
                                 [7, 210, 479, 27, 4, -0.02],
                                 [8, 202, 504, 14, 4, -0.05],
                                 [9, 198, 510, 30, 11, -0.84],
                                 [10, 158, 416, 16, 7, -0.69],
                                 [11, 165, 393, 14, 5, 0.30],
                                 [12, 201, 442, 25, 5, -0.46],
                                 [13, 157, 317, 52, 5, -0.03],
                                 [14, 131, 311, 16, 6, 0.50],
                                 [15, 166, 400, 34, 6, 0.73],
                                 [16, 160, 337, 31, 5, -0.52],
                                 [17, 186, 423, 42, 9, 0.90],
                                 [18, 125, 334, 26, 8, 0.40],
                                 [19, 218, 533, 16, 6, -0.78],
                                 [20, 146, 344, 22, 5, -0.56]]),
                   columns=['id','x','y','sigma_y','sigma_x','rho_xy'])


## for convenience zero-base the 'id' and use as index
dfhogg['id'] = dfhogg['id'] - 1
dfhogg.set_index('id', inplace=True)

## standardize (mean center and divide by 1 sd)
dfhoggs = (dfhogg[['x','y']] - dfhogg[['x','y']].mean(0)) / dfhogg[['x','y']].std(0)
dfhoggs['sigma_y'] = dfhogg['sigma_y'] / dfhogg['y'].std(0)
dfhoggs['sigma_x'] = dfhogg['sigma_x'] / dfhogg['x'].std(0)

def plot_hoggs(dfhoggs):
  ## create xlims ylims for plotting
  xlims = (dfhoggs['x'].min() - np.ptp(dfhoggs['x'])/5,
           dfhoggs['x'].max() + np.ptp(dfhoggs['x'])/5)
  ylims = (dfhoggs['y'].min() - np.ptp(dfhoggs['y'])/5,
           dfhoggs['y'].max() + np.ptp(dfhoggs['y'])/5)

  ## scatterplot the standardized data
  g = sns.FacetGrid(dfhoggs, size=8)
  _ = g.map(plt.errorbar, 'x', 'y', 'sigma_y', 'sigma_x', marker="o", ls='')
  _ = g.axes[0][0].set_ylim(ylims)
  _ = g.axes[0][0].set_xlim(xlims)

  plt.subplots_adjust(top=0.92)
  _ = g.fig.suptitle('Scatterplot of Hogg 2010 dataset after standardization', fontsize=16)
  return g, xlims, ylims
  
g = plot_hoggs(dfhoggs)

/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.
  return ptp(axis=axis, out=out, **kwargs)
/usr/local/lib/python3.6/dist-packages/seaborn/axisgrid.py:230: UserWarning: The `size` paramter has been renamed to `height`; please update your code.
  warnings.warn(msg, UserWarning)

In [ ]:

X_np = dfhoggs['x'].values
sigma_y_np = dfhoggs['sigma_y'].values
Y_np = dfhoggs['y'].values

기존 OLS 모델

이제 간단한 절편 + 기울기 회귀 문제인 선형 모델을 준비해 보겠습니다.

In [ ]:

mdl_ols = tfd.JointDistributionSequential([
    # b0 ~ Normal(0, 1)
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # b1 ~ Normal(0, 1)
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # x ~ Normal(b0+b1*X, 1)
    lambda b1, b0: tfd.Normal(
      # Parameter transformation
      loc=b0 + b1*X_np,
      scale=sigma_y_np)
])

그런 다음 모델의 그래프를 확인하여 종속성을 확인할 수 있습니다. x는 마지막 노드의 이름으로 예약되어 있으며 이를 JointDistributionSequential 모델에서 람다 인수로 확신할 수 없습니다.

In [ ]:

mdl_ols.resolve_graph()

(('b0', ()), ('b1', ()), ('x', ('b1', 'b0')))

모델에서 샘플링하는 것은 매우 간단합니다.

In [ ]:

mdl_ols.sample()

[<tf.Tensor: shape=(), dtype=float64, numpy=-0.50225804634794>,
 <tf.Tensor: shape=(), dtype=float64, numpy=0.682740126293564>,
 <tf.Tensor: shape=(20,), dtype=float64, numpy=
 array([-0.33051382,  0.71443618, -1.91085683,  0.89371173, -0.45060957,
        -1.80448758, -0.21357082,  0.07891058, -0.20689721, -0.62690385,
        -0.55225748, -0.11446535, -0.66624497, -0.86913291, -0.93605552,
        -0.83965336, -0.70988597, -0.95813437,  0.15884761, -0.31113434])>]

...tf.Tensor의 목록을 제공합니다. 이것을 즉시 log_prob 함수에 연결하여 모델의 log_prob를 계산할 수 있습니다.

In [ ]:

prior_predictive_samples = mdl_ols.sample()
mdl_ols.log_prob(prior_predictive_samples)

<tf.Tensor: shape=(20,), dtype=float64, numpy=
array([-4.97502846, -3.98544303, -4.37514505, -3.46933487, -3.80688125,
       -3.42907525, -4.03263074, -3.3646366 , -4.70370938, -4.36178501,
       -3.47823735, -3.94641662, -5.76906319, -4.0944128 , -4.39310708,
       -4.47713894, -4.46307881, -3.98802372, -3.83027747, -4.64777082])>

흠, 뭔가 잘못되었습니다. 스칼라 log_prob를 얻어야 합니다! 실제로 그래픽 모델에서 각 노드의 log_prob를 제공하는 .log_prob_parts를 호출하여 문제가 있는지 추가로 확인할 수 있습니다.

In [ ]:

mdl_ols.log_prob_parts(prior_predictive_samples)

[<tf.Tensor: shape=(), dtype=float64, numpy=-0.9699239562734849>,
 <tf.Tensor: shape=(), dtype=float64, numpy=-3.459364167569284>,
 <tf.Tensor: shape=(20,), dtype=float64, numpy=
 array([-0.54574034,  0.4438451 ,  0.05414307,  0.95995326,  0.62240687,
         1.00021288,  0.39665739,  1.06465152, -0.27442125,  0.06750311,
         0.95105078,  0.4828715 , -1.33977506,  0.33487533,  0.03618104,
        -0.04785082, -0.03379069,  0.4412644 ,  0.59901066, -0.2184827 ])>]

...마지막 노드가 iid 차원/축을 따라 reduce_sum이 아닌 것으로 나타났습니다! 따라서 합계를 수행할 때 처음 두 변수는 잘못 브로드캐스트됩니다.

여기서 트릭은 tfd.Independent를 사용하여 배치 형상을 재해석하는 것입니다(축의 나머지가 올바르게 축소되도록).

In [ ]:

mdl_ols_ = tfd.JointDistributionSequential([
    # b0
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # b1
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # likelihood
    #   Using Independent to ensure the log_prob is not incorrectly broadcasted
    lambda b1, b0: tfd.Independent(
        tfd.Normal(
            # Parameter transformation
            # b1 shape: (batch_shape), X shape (num_obs): we want result to have
            # shape (batch_shape, num_obs)
            loc=b0 + b1*X_np,
            scale=sigma_y_np),
        reinterpreted_batch_ndims=1
    ),
])

이제 모델의 마지막 노드/분포를 확인하겠습니다. 이제 이벤트 형상이 올바르게 해석된 것을 볼 수 있습니다. reinterpreted_batch_ndims를 올바르게 얻으려면 약간의 시행 착오가 필요할 수 있지만 항상 쉽게 분포 또는 샘플링된 텐서를 출력하여 형상을 다시 확인할 수 있습니다!

In [ ]:

print(mdl_ols_.sample_distributions()[0][-1])
print(mdl_ols.sample_distributions()[0][-1])

tfp.distributions.Independent("JointDistributionSequential_sample_distributions_IndependentJointDistributionSequential_sample_distributions_Normal", batch_shape=[], event_shape=[20], dtype=float64)
tfp.distributions.Normal("JointDistributionSequential_sample_distributions_Normal", batch_shape=[20], event_shape=[], dtype=float64)

In [ ]:

prior_predictive_samples = mdl_ols_.sample()
mdl_ols_.log_prob(prior_predictive_samples)  # <== Getting a scalar correctly

<tf.Tensor: shape=(), dtype=float64, numpy=-2.543425661013286>

기타 `JointDistribution*` API

In [ ]:

mdl_ols_named = tfd.JointDistributionNamed(dict(
    likelihood = lambda b0, b1: tfd.Independent(
        tfd.Normal(
            loc=b0 + b1*X_np,
            scale=sigma_y_np),
        reinterpreted_batch_ndims=1
    ),
    b0         = tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    b1         = tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
))

mdl_ols_named.log_prob(mdl_ols_named.sample())

<tf.Tensor: shape=(), dtype=float64, numpy=-5.99620966071338>

In [ ]:

mdl_ols_named.sample()  # output is a dictionary

{'b0': <tf.Tensor: shape=(), dtype=float64, numpy=0.26364058399428225>,
 'b1': <tf.Tensor: shape=(), dtype=float64, numpy=-0.27209402374432207>,
 'likelihood': <tf.Tensor: shape=(20,), dtype=float64, numpy=
 array([ 0.6482155 , -0.39314108,  0.62744764, -0.24587987, -0.20544617,
         1.01465392, -0.04705611, -0.16618702,  0.36410134,  0.3943299 ,
         0.36455291, -0.27822219, -0.24423928,  0.24599518,  0.82731092,
        -0.21983033,  0.56753169,  0.32830481, -0.15713064,  0.23336351])>}

In [ ]:

Root = tfd.JointDistributionCoroutine.Root  # Convenient alias.
def model():
  b1 = yield Root(tfd.Normal(loc=tf.cast(0, dtype), scale=1.))
  b0 = yield Root(tfd.Normal(loc=tf.cast(0, dtype), scale=1.))
  yhat = b0 + b1*X_np
  likelihood = yield tfd.Independent(
        tfd.Normal(loc=yhat, scale=sigma_y_np),
        reinterpreted_batch_ndims=1
    )

mdl_ols_coroutine = tfd.JointDistributionCoroutine(model)
mdl_ols_coroutine.log_prob(mdl_ols_coroutine.sample())

<tf.Tensor: shape=(), dtype=float64, numpy=-4.566678123520463>

In [ ]:

mdl_ols_coroutine.sample()  # output is a tuple

(<tf.Tensor: shape=(), dtype=float64, numpy=0.06811002171170354>,
 <tf.Tensor: shape=(), dtype=float64, numpy=-0.37477064754116807>,
 <tf.Tensor: shape=(20,), dtype=float64, numpy=
 array([-0.91615096, -0.20244718, -0.47840159, -0.26632479, -0.60441105,
        -0.48977789, -0.32422329, -0.44019322, -0.17072643, -0.20666025,
        -0.55932191, -0.40801868, -0.66893181, -0.24134135, -0.50403536,
        -0.51788596, -0.90071876, -0.47382338, -0.34821655, -0.38559724])>)

MLE

이제 추론을 할 수 있습니다! 옵티마이저를 사용하여 최대 우도 추정을 찾을 수 있습니다.

In [ ]:

#@title Define some helper functions

# bfgs and lbfgs currently requries a function that returns both the value and
# gradient re the input.
import functools

def _make_val_and_grad_fn(value_fn):
  @functools.wraps(value_fn)
  def val_and_grad(x):
    return tfp.math.value_and_gradient(value_fn, x)
  return val_and_grad

# Map a list of tensors (e.g., output from JDSeq.sample([...])) to a single tensor
# modify from tfd.Blockwise
from tensorflow_probability.python.internal import dtype_util
from tensorflow_probability.python.internal import prefer_static as ps
from tensorflow_probability.python.internal import tensorshape_util

class Mapper:
  """Basically, this is a bijector without log-jacobian correction."""
  def __init__(self, list_of_tensors, list_of_bijectors, event_shape):
    self.dtype = dtype_util.common_dtype(
        list_of_tensors, dtype_hint=tf.float32)
    self.list_of_tensors = list_of_tensors
    self.bijectors = list_of_bijectors
    self.event_shape = event_shape

  def flatten_and_concat(self, list_of_tensors):
    def _reshape_map_part(part, event_shape, bijector):
      part = tf.cast(bijector.inverse(part), self.dtype)
      static_rank = tf.get_static_value(ps.rank_from_shape(event_shape))
      if static_rank == 1:
        return part
      new_shape = ps.concat([
          ps.shape(part)[:ps.size(ps.shape(part)) - ps.size(event_shape)], 
          [-1]
      ], axis=-1)
      return tf.reshape(part, ps.cast(new_shape, tf.int32))

    x = tf.nest.map_structure(_reshape_map_part,
                              list_of_tensors,
                              self.event_shape,
                              self.bijectors)
    return tf.concat(tf.nest.flatten(x), axis=-1)

  def split_and_reshape(self, x):
    assertions = []
    message = 'Input must have at least one dimension.'
    if tensorshape_util.rank(x.shape) is not None:
      if tensorshape_util.rank(x.shape) == 0:
        raise ValueError(message)
    else:
      assertions.append(assert_util.assert_rank_at_least(x, 1, message=message))
    with tf.control_dependencies(assertions):
      splits = [
          tf.cast(ps.maximum(1, ps.reduce_prod(s)), tf.int32)
          for s in tf.nest.flatten(self.event_shape)
      ]
      x = tf.nest.pack_sequence_as(
          self.event_shape, tf.split(x, splits, axis=-1))
      def _reshape_map_part(part, part_org, event_shape, bijector):
        part = tf.cast(bijector.forward(part), part_org.dtype)
        static_rank = tf.get_static_value(ps.rank_from_shape(event_shape))
        if static_rank == 1:
          return part
        new_shape = ps.concat([ps.shape(part)[:-1], event_shape], axis=-1)
        return tf.reshape(part, ps.cast(new_shape, tf.int32))

      x = tf.nest.map_structure(_reshape_map_part,
                                x, 
                                self.list_of_tensors,
                                self.event_shape,
                                self.bijectors)
    return x

In [ ]:

mapper = Mapper(mdl_ols_.sample()[:-1],
                [tfb.Identity(), tfb.Identity()],
                mdl_ols_.event_shape[:-1])

# mapper.split_and_reshape(mapper.flatten_and_concat(mdl_ols_.sample()[:-1]))

In [ ]:

@_make_val_and_grad_fn
def neg_log_likelihood(x):
  # Generate a function closure so that we are computing the log_prob
  # conditioned on the observed data. Note also that tfp.optimizer.* takes a 
  # single tensor as input.
  return -mdl_ols_.log_prob(mapper.split_and_reshape(x) + [Y_np])

lbfgs_results = tfp.optimizer.lbfgs_minimize(
    neg_log_likelihood,
    initial_position=tf.zeros(2, dtype=dtype),
    tolerance=1e-20,
    x_tolerance=1e-8
)

In [ ]:

b0est, b1est = lbfgs_results.position.numpy()

g, xlims, ylims = plot_hoggs(dfhoggs);
xrange = np.linspace(xlims[0], xlims[1], 100)
g.axes[0][0].plot(xrange, b0est + b1est*xrange, 
                  color='r', label='MLE of OLE model')
plt.legend();

/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.
  return ptp(axis=axis, out=out, **kwargs)
/usr/local/lib/python3.6/dist-packages/seaborn/axisgrid.py:230: UserWarning: The `size` paramter has been renamed to `height`; please update your code.
  warnings.warn(msg, UserWarning)

배치 버전 모델 및 MCMC

베이지안 추론에서는 일반적으로 MCMC 샘플로 작업해야 할 것입니다. 샘플이 뒤쪽에 있을 때 기대치를 계산하기 위해 임의의 함수에 연결할 수 있기 때문입니다. 그러나 MCMC API에서는 배치 친화적 모델을 작성해야 하며 sample([...])을 호출하여 모델이 실제로 '배치 처리 가능'하지 않은지 확인할 수 있습니다.

mdl_ols_.sample(5)  # <== error as some computation could not be broadcasted.

이 경우 모델 내부에 선형 함수만 있으므로 비교적 간단합니다. 형상을 확장하는 것이 방법입니다.

In [ ]:

mdl_ols_batch = tfd.JointDistributionSequential([
    # b0
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # b1
    tfd.Normal(loc=tf.cast(0, dtype), scale=1.),
    # likelihood
    #   Using Independent to ensure the log_prob is not incorrectly broadcasted
    lambda b1, b0: tfd.Independent(
        tfd.Normal(
            # Parameter transformation
            loc=b0[..., tf.newaxis] + b1[..., tf.newaxis]*X_np[tf.newaxis, ...],
            scale=sigma_y_np[tf.newaxis, ...]),
        reinterpreted_batch_ndims=1
    ),
])

mdl_ols_batch.resolve_graph()

(('b0', ()), ('b1', ()), ('x', ('b1', 'b0')))

몇 가지 검사를 수행하기 위해 log_prob_parts를 다시 샘플링하고 평가할 수 있습니다.

In [ ]:

b0, b1, y = mdl_ols_batch.sample(4)
mdl_ols_batch.log_prob_parts([b0, b1, y])

[<tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1.25230168, -1.45281432, -1.87110061, -1.07665206])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1.07019936, -1.59562117, -2.53387765, -1.01557632])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([ 0.45841406,  2.56829635, -4.84973951, -5.59423992])>]

몇 가지 참고 사항:

모델의 배치 버전으로 작업하려고 하는데, 이것이 다중 체인 MCMC에서 가장 빠르기 때문입니다. 모델을 배치 처리된 버전으로 다시 작성할 수 없는 경우(예: ODE 모델) tf.map_fn를 사용하여 log_prob 함수를 매핑하면 동일한 효과를 거둘 수 있습니다.
이제 scaler_tensor[:, None]을 수행할 수 없으므로 mdl_ols_batch.sample()이 이전에 스케일러가 있는 것처럼 작동하지 않을 수 있습니다. 여기서는 tfd.Sample(..., sample_shape=1)을 래핑하여 스케일러 텐서를 1순위로 확장하는 것이 해결책입니다.
하이퍼 파라미터와 같은 설정을 훨씬 쉽게 변경할 수 있도록 모델을 함수로 작성하는 것이 좋습니다.

In [ ]:

def gen_ols_batch_model(X, sigma, hyperprior_mean=0, hyperprior_scale=1):
  hyper_mean = tf.cast(hyperprior_mean, dtype)
  hyper_scale = tf.cast(hyperprior_scale, dtype)
  return tfd.JointDistributionSequential([
      # b0
      tfd.Sample(tfd.Normal(loc=hyper_mean, scale=hyper_scale), sample_shape=1),
      # b1
      tfd.Sample(tfd.Normal(loc=hyper_mean, scale=hyper_scale), sample_shape=1),
      # likelihood
      lambda b1, b0: tfd.Independent(
          tfd.Normal(
              # Parameter transformation
              loc=b0 + b1*X,
              scale=sigma),
          reinterpreted_batch_ndims=1
      ),
  ], validate_args=True)

mdl_ols_batch = gen_ols_batch_model(X_np[tf.newaxis, ...],
                                    sigma_y_np[tf.newaxis, ...])

_ = mdl_ols_batch.sample()
_ = mdl_ols_batch.sample(4)
_ = mdl_ols_batch.sample([3, 4])

In [ ]:

# Small helper function to validate log_prob shape (avoid wrong broadcasting)
def validate_log_prob_part(model, batch_shape=1, observed=-1):
  samples = model.sample(batch_shape)
  logp_part = list(model.log_prob_parts(samples))
  
  # exclude observed node
  logp_part.pop(observed)
  for part in logp_part:
    tf.assert_equal(part.shape, logp_part[-1].shape)

validate_log_prob_part(mdl_ols_batch, 4)

In [ ]:

#@title More checks: comparing the generated log_prob fucntion with handwrittent TFP log_prob function. { display-mode: "form" }

def ols_logp_batch(b0, b1, Y):
  b0_prior = tfd.Normal(loc=tf.cast(0, dtype), scale=1.) # b0
  b1_prior = tfd.Normal(loc=tf.cast(0, dtype), scale=1.) # b1
  likelihood = tfd.Normal(loc=b0 + b1*X_np[None, :],
                          scale=sigma_y_np) # likelihood
  return tf.reduce_sum(b0_prior.log_prob(b0), axis=-1) +\
         tf.reduce_sum(b1_prior.log_prob(b1), axis=-1) +\
         tf.reduce_sum(likelihood.log_prob(Y), axis=-1)

b0, b1, x = mdl_ols_batch.sample(4)
print(mdl_ols_batch.log_prob([b0, b1, Y_np]).numpy()) 
print(ols_logp_batch(b0, b1, Y_np).numpy())

[-227.37899384 -327.10043743 -570.44162789 -702.79808683]
[-227.37899384 -327.10043743 -570.44162789 -702.79808683]

No-U-Turn Sampler를 사용하는 MCMC

In [ ]:

#@title A common `run_chain` function
@tf.function(autograph=False, experimental_compile=True)
def run_chain(init_state, step_size, target_log_prob_fn, unconstraining_bijectors,
              num_steps=500, burnin=50):

  def trace_fn(_, pkr):
    return (
        pkr.inner_results.inner_results.target_log_prob,
        pkr.inner_results.inner_results.leapfrogs_taken,
        pkr.inner_results.inner_results.has_divergence,
        pkr.inner_results.inner_results.energy,
        pkr.inner_results.inner_results.log_accept_ratio
           )
  
  kernel = tfp.mcmc.TransformedTransitionKernel(
    inner_kernel=tfp.mcmc.NoUTurnSampler(
      target_log_prob_fn,
      step_size=step_size),
    bijector=unconstraining_bijectors)

  hmc = tfp.mcmc.DualAveragingStepSizeAdaptation(
    inner_kernel=kernel,
    num_adaptation_steps=burnin,
    step_size_setter_fn=lambda pkr, new_step_size: pkr._replace(
        inner_results=pkr.inner_results._replace(step_size=new_step_size)),
    step_size_getter_fn=lambda pkr: pkr.inner_results.step_size,
    log_accept_prob_getter_fn=lambda pkr: pkr.inner_results.log_accept_ratio
  )

  # Sampling from the chain.
  chain_state, sampler_stat = tfp.mcmc.sample_chain(
      num_results=num_steps,
      num_burnin_steps=burnin,
      current_state=init_state,
      kernel=hmc,
      trace_fn=trace_fn)
  return chain_state, sampler_stat

In [ ]:

nchain = 10
b0, b1, _ = mdl_ols_batch.sample(nchain)
init_state = [b0, b1]
step_size = [tf.cast(i, dtype=dtype) for i in [.1, .1]]
target_log_prob_fn = lambda *x: mdl_ols_batch.log_prob(x + (Y_np, ))

# bijector to map contrained parameters to real
unconstraining_bijectors = [
    tfb.Identity(),
    tfb.Identity(),
]

samples, sampler_stat = run_chain(
    init_state, step_size, target_log_prob_fn, unconstraining_bijectors)

In [ ]:

# using the pymc3 naming convention
sample_stats_name = ['lp', 'tree_size', 'diverging', 'energy', 'mean_tree_accept']
sample_stats = {k:v.numpy().T for k, v in zip(sample_stats_name, sampler_stat)}
sample_stats['tree_size'] = np.diff(sample_stats['tree_size'], axis=1)

var_name = ['b0', 'b1']
posterior = {k:np.swapaxes(v.numpy(), 1, 0) 
             for k, v in zip(var_name, samples)}

az_trace = az.from_dict(posterior=posterior, sample_stats=sample_stats)

In [ ]:

az.plot_trace(az_trace);

In [ ]:

az.plot_forest(az_trace,
               kind='ridgeplot',
               linewidth=4,
               combined=True,
               ridgeplot_overlap=1.5,
               figsize=(9, 4));

In [ ]:

k = 5
b0est, b1est = az_trace.posterior['b0'][:, -k:].values, az_trace.posterior['b1'][:, -k:].values

g, xlims, ylims = plot_hoggs(dfhoggs);
xrange = np.linspace(xlims[0], xlims[1], 100)[None, :]
g.axes[0][0].plot(np.tile(xrange, (k, 1)).T,
                  (np.reshape(b0est, [-1, 1]) + np.reshape(b1est, [-1, 1])*xrange).T,
                  alpha=.25, color='r')
plt.legend([g.axes[0][0].lines[-1]], ['MCMC OLE model']);

/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.
  return ptp(axis=axis, out=out, **kwargs)
/usr/local/lib/python3.6/dist-packages/seaborn/axisgrid.py:230: UserWarning: The `size` paramter has been renamed to `height`; please update your code.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:8: MatplotlibDeprecationWarning: cycling among columns of inputs with non-matching shapes is deprecated.
  

스튜던트-T 방법

지금부터는 항상 모델의 배치 버전으로 작업합니다.

In [ ]:

def gen_studentt_model(X, sigma,
                       hyper_mean=0, hyper_scale=1, lower=1, upper=100):
  loc = tf.cast(hyper_mean, dtype)
  scale = tf.cast(hyper_scale, dtype)
  low = tf.cast(lower, dtype)
  high = tf.cast(upper, dtype)
  return tfd.JointDistributionSequential([
      # b0 ~ Normal(0, 1)
      tfd.Sample(tfd.Normal(loc, scale), sample_shape=1),
      # b1 ~ Normal(0, 1)
      tfd.Sample(tfd.Normal(loc, scale), sample_shape=1),
      # df ~ Uniform(a, b)
      tfd.Sample(tfd.Uniform(low, high), sample_shape=1),
      # likelihood ~ StudentT(df, f(b0, b1), sigma_y)
      #   Using Independent to ensure the log_prob is not incorrectly broadcasted.
      lambda df, b1, b0: tfd.Independent(
          tfd.StudentT(df=df, loc=b0 + b1*X, scale=sigma)),
  ], validate_args=True)

mdl_studentt = gen_studentt_model(X_np[tf.newaxis, ...],
                                  sigma_y_np[tf.newaxis, ...])
mdl_studentt.resolve_graph()

(('b0', ()), ('b1', ()), ('df', ()), ('x', ('df', 'b1', 'b0')))

In [ ]:

validate_log_prob_part(mdl_studentt, 4)

순방향 샘플(사전 예측 샘플링)

In [ ]:

b0, b1, df, x = mdl_studentt.sample(1000)
x.shape

TensorShape([1000, 20])

MLE

In [ ]:

# bijector to map contrained parameters to real
a, b = tf.constant(1., dtype), tf.constant(100., dtype),

# Interval transformation
tfp_interval = tfb.Inline(
    inverse_fn=(
        lambda x: tf.math.log(x - a) - tf.math.log(b - x)),
    forward_fn=(
        lambda y: (b - a) * tf.sigmoid(y) + a),
    forward_log_det_jacobian_fn=(
        lambda x: tf.math.log(b - a) - 2 * tf.nn.softplus(-x) - x),
    forward_min_event_ndims=0,
    name="interval")

unconstraining_bijectors = [
    tfb.Identity(),
    tfb.Identity(),
    tfp_interval,
]

mapper = Mapper(mdl_studentt.sample()[:-1],
                unconstraining_bijectors,
                mdl_studentt.event_shape[:-1])

In [ ]:

@_make_val_and_grad_fn
def neg_log_likelihood(x):
  # Generate a function closure so that we are computing the log_prob
  # conditioned on the observed data. Note also that tfp.optimizer.* takes a 
  # single tensor as input, so we need to do some slicing here:
  return -tf.squeeze(mdl_studentt.log_prob(
      mapper.split_and_reshape(x) + [Y_np]))

lbfgs_results = tfp.optimizer.lbfgs_minimize(
    neg_log_likelihood,
    initial_position=mapper.flatten_and_concat(mdl_studentt.sample()[:-1]),
    tolerance=1e-20,
    x_tolerance=1e-20
)

In [ ]:

b0est, b1est, dfest = lbfgs_results.position.numpy()

g, xlims, ylims = plot_hoggs(dfhoggs);
xrange = np.linspace(xlims[0], xlims[1], 100)
g.axes[0][0].plot(xrange, b0est + b1est*xrange, 
                  color='r', label='MLE of StudentT model')
plt.legend();

/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.
  return ptp(axis=axis, out=out, **kwargs)
/usr/local/lib/python3.6/dist-packages/seaborn/axisgrid.py:230: UserWarning: The `size` paramter has been renamed to `height`; please update your code.
  warnings.warn(msg, UserWarning)

MCMC

In [ ]:

nchain = 10
b0, b1, df, _ = mdl_studentt.sample(nchain)
init_state = [b0, b1, df]
step_size = [tf.cast(i, dtype=dtype) for i in [.1, .1, .05]]

target_log_prob_fn = lambda *x: mdl_studentt.log_prob(x + (Y_np, ))

samples, sampler_stat = run_chain(
    init_state, step_size, target_log_prob_fn, unconstraining_bijectors, burnin=100)

In [ ]:

# using the pymc3 naming convention
sample_stats_name = ['lp', 'tree_size', 'diverging', 'energy', 'mean_tree_accept']
sample_stats = {k:v.numpy().T for k, v in zip(sample_stats_name, sampler_stat)}
sample_stats['tree_size'] = np.diff(sample_stats['tree_size'], axis=1)

var_name = ['b0', 'b1', 'df']
posterior = {k:np.swapaxes(v.numpy(), 1, 0) 
             for k, v in zip(var_name, samples)}

az_trace = az.from_dict(posterior=posterior, sample_stats=sample_stats)

In [ ]:

az.summary(az_trace)

In [ ]:

az.plot_trace(az_trace);

In [ ]:

az.plot_forest(az_trace,
               kind='ridgeplot',
               linewidth=4,
               combined=True,
               ridgeplot_overlap=1.5,
               figsize=(9, 4));

In [ ]:

plt.hist(az_trace.sample_stats['tree_size'], np.linspace(.5, 25.5, 26), alpha=.5);

In [ ]:

k = 5
b0est, b1est = az_trace.posterior['b0'][:, -k:].values, az_trace.posterior['b1'][:, -k:].values

g, xlims, ylims = plot_hoggs(dfhoggs);
xrange = np.linspace(xlims[0], xlims[1], 100)[None, :]
g.axes[0][0].plot(np.tile(xrange, (k, 1)).T,
                  (np.reshape(b0est, [-1, 1]) + np.reshape(b1est, [-1, 1])*xrange).T,
                  alpha=.25, color='r')
plt.legend([g.axes[0][0].lines[-1]], ['MCMC StudentT model']);

/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.
  return ptp(axis=axis, out=out, **kwargs)
/usr/local/lib/python3.6/dist-packages/seaborn/axisgrid.py:230: UserWarning: The `size` paramter has been renamed to `height`; please update your code.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:8: MatplotlibDeprecationWarning: cycling among columns of inputs with non-matching shapes is deprecated.
  

계층적 부분 풀링

PyMC3 Efron과 Morris(1975)의 선수 18명에 대한 야구 데이터에서 가져왔습니다.

In [ ]:

data = pd.read_table('https://raw.githubusercontent.com/pymc-devs/pymc3/master/pymc3/examples/data/efron-morris-75-data.tsv',
                     sep="\t")
at_bats, hits = data[['At-Bats', 'Hits']].values.T
n = len(at_bats)

In [ ]:

def gen_baseball_model(at_bats, rate=1.5, a=0, b=1):
  return tfd.JointDistributionSequential([
    # phi
    tfd.Uniform(low=tf.cast(a, dtype), high=tf.cast(b, dtype)),
    # kappa_log
    tfd.Exponential(rate=tf.cast(rate, dtype)),
    # thetas
    lambda kappa_log, phi: tfd.Sample(
        tfd.Beta(
            concentration1=tf.exp(kappa_log)*phi,
            concentration0=tf.exp(kappa_log)*(1.0-phi)),
        sample_shape=n
    ),
    # likelihood
    lambda thetas: tfd.Independent(
        tfd.Binomial(
            total_count=tf.cast(at_bats, dtype),
            probs=thetas
        )), 
])

mdl_baseball = gen_baseball_model(at_bats)
mdl_baseball.resolve_graph()

(('phi', ()),
 ('kappa_log', ()),
 ('thetas', ('kappa_log', 'phi')),
 ('x', ('thetas',)))

순방향 샘플(사전 예측 샘플링)

In [ ]:

phi, kappa_log, thetas, y = mdl_baseball.sample(4)
# phi, kappa_log, thetas, y

다시 말하지만, Independent를 사용하지 않으면 잘못된 batch_shape를 가진 log_prob로 끝나게 된다는 점에 주목하세요.

In [ ]:

  # check logp
pprint(mdl_baseball.log_prob_parts([phi, kappa_log, thetas, hits]))
print(mdl_baseball.log_prob([phi, kappa_log, thetas, hits]))

[<tf.Tensor: shape=(4,), dtype=float64, numpy=array([0., 0., 0., 0.])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([ 0.1721297 , -0.95946498, -0.72591188,  0.23993813])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([59.35192283,  7.0650634 ,  0.83744911, 74.14370935])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-3279.75191016,  -931.10438484,  -512.59197688, -1131.08043597])>]
tf.Tensor([-3220.22785762  -924.99878641  -512.48043966 -1056.69678849], shape=(4,), dtype=float64)

MLE

tfp.optimizer의 매우 놀라운 기능은 시작점의 k 배치에 대해 병렬로 최적화하고 stopping_condition kwarg를 지정할 수 있다는 것입니다. 이것을 tfp.optimizer.converged_all로 설정하여 동일한 최소값을 찾는지 확인하고 tfp.optimizer.converged_any를 사용하여 로컬 솔루션을 빠르게 찾습니다.

In [ ]:

unconstraining_bijectors = [
    tfb.Sigmoid(),
    tfb.Exp(),
    tfb.Sigmoid(),
]

phi, kappa_log, thetas, y = mdl_baseball.sample(10)

mapper = Mapper([phi, kappa_log, thetas],
                unconstraining_bijectors,
                mdl_baseball.event_shape[:-1])

In [ ]:

@_make_val_and_grad_fn
def neg_log_likelihood(x):
  return -mdl_baseball.log_prob(mapper.split_and_reshape(x) + [hits])

start = mapper.flatten_and_concat([phi, kappa_log, thetas])

lbfgs_results = tfp.optimizer.lbfgs_minimize(
    neg_log_likelihood,
    num_correction_pairs=10,
    initial_position=start,
    # lbfgs actually can work in batch as well
    stopping_condition=tfp.optimizer.converged_any,
    tolerance=1e-50,
    x_tolerance=1e-50,
    parallel_iterations=10,
    max_iterations=200
)

In [ ]:

lbfgs_results.converged.numpy(), lbfgs_results.failed.numpy()

(array([False, False, False, False, False, False, False, False, False,
        False]),
 array([ True,  True,  True,  True,  True,  True,  True,  True,  True,
         True]))

In [ ]:

result = lbfgs_results.position[lbfgs_results.converged & ~lbfgs_results.failed]
result

<tf.Tensor: shape=(0, 20), dtype=float64, numpy=array([], shape=(0, 20), dtype=float64)>

LBFGS는 수렴되지 않았습니다.

In [ ]:

if result.shape[0] > 0:
  phi_est, kappa_est, theta_est = mapper.split_and_reshape(result)
  phi_est, kappa_est, theta_est

MCMC

In [ ]:

target_log_prob_fn = lambda *x: mdl_baseball.log_prob(x + (hits, ))

nchain = 4
phi, kappa_log, thetas, _ = mdl_baseball.sample(nchain)
init_state = [phi, kappa_log, thetas]
step_size=[tf.cast(i, dtype=dtype) for i in [.1, .1, .1]]

samples, sampler_stat = run_chain(
    init_state, step_size, target_log_prob_fn, unconstraining_bijectors,
    burnin=200)

In [ ]:

# using the pymc3 naming convention
sample_stats_name = ['lp', 'tree_size', 'diverging', 'energy', 'mean_tree_accept']
sample_stats = {k:v.numpy().T for k, v in zip(sample_stats_name, sampler_stat)}
sample_stats['tree_size'] = np.diff(sample_stats['tree_size'], axis=1)

var_name = ['phi', 'kappa_log', 'thetas']
posterior = {k:np.swapaxes(v.numpy(), 1, 0) 
             for k, v in zip(var_name, samples)}

az_trace = az.from_dict(posterior=posterior, sample_stats=sample_stats)

In [ ]:

az.plot_trace(az_trace, compact=True);

In [ ]:

az.plot_forest(az_trace,
               var_names=['thetas'],
               kind='ridgeplot',
               linewidth=4,
               combined=True,
               ridgeplot_overlap=1.5,
               figsize=(9, 8));

혼합 효과 모델(Radon)

PyMC3 문서: 다단계 모델링을 위한 베이지안 방법 입문서의 마지막 모델

이전의 일부 변경 사항(더 작은 스케일 등)

In [ ]:

#@title Load raw data and clean up
srrs2 = pd.read_csv('https://raw.githubusercontent.com/pymc-devs/pymc3/master/pymc3/examples/data/srrs2.dat')

srrs2.columns = srrs2.columns.map(str.strip)
srrs_mn = srrs2[srrs2.state=='MN'].copy()
srrs_mn['fips'] = srrs_mn.stfips*1000 + srrs_mn.cntyfips

cty = pd.read_csv('https://raw.githubusercontent.com/pymc-devs/pymc3/master/pymc3/examples/data/cty.dat')
cty_mn = cty[cty.st=='MN'].copy()
cty_mn[ 'fips'] = 1000*cty_mn.stfips + cty_mn.ctfips

srrs_mn = srrs_mn.merge(cty_mn[['fips', 'Uppm']], on='fips')
srrs_mn = srrs_mn.drop_duplicates(subset='idnum')
u = np.log(srrs_mn.Uppm)

n = len(srrs_mn)

srrs_mn.county = srrs_mn.county.map(str.strip)
mn_counties = srrs_mn.county.unique()
counties = len(mn_counties)
county_lookup = dict(zip(mn_counties, range(len(mn_counties))))

county = srrs_mn['county_code'] = srrs_mn.county.replace(county_lookup).values
radon = srrs_mn.activity
srrs_mn['log_radon'] = log_radon = np.log(radon + 0.1).values
floor_measure = srrs_mn.floor.values.astype('float')

# Create new variable for mean of floor across counties
xbar = srrs_mn.groupby('county')['floor'].mean().rename(county_lookup).values

복잡한 변환이 있는 모델의 경우, 기능적 스타일로 구현하면 작성과 테스트가 훨씬 쉬워집니다. 또한 입력된 데이터의 (미니 배치) 조건을 지정하는 log_prob 함수를 프로그래밍 방식으로 훨씬 쉽게 생성할 수 있습니다.

In [ ]:

def affine(u_val, x_county, county, floor, gamma, eps, b):
  """Linear equation of the coefficients and the covariates, with broadcasting."""
  return (tf.transpose((gamma[..., 0]
                      + gamma[..., 1]*u_val[:, None]
                      + gamma[..., 2]*x_county[:, None]))
          + tf.gather(eps, county, axis=-1)
          + b*floor)


def gen_radon_model(u_val, x_county, county, floor,
                    mu0=tf.zeros([], dtype, name='mu0')):
  """Creates a joint distribution representing our generative process."""
  return tfd.JointDistributionSequential([
      # sigma_a
      tfd.HalfCauchy(loc=mu0, scale=5.),
      # eps
      lambda sigma_a: tfd.Sample(
          tfd.Normal(loc=mu0, scale=sigma_a), sample_shape=counties),
      # gamma
      tfd.Sample(tfd.Normal(loc=mu0, scale=100.), sample_shape=3),
      # b
      tfd.Sample(tfd.Normal(loc=mu0, scale=100.), sample_shape=1),
      # sigma_y
      tfd.Sample(tfd.HalfCauchy(loc=mu0, scale=5.), sample_shape=1),
      # likelihood
      lambda sigma_y, b, gamma, eps: tfd.Independent(
          tfd.Normal(
              loc=affine(u_val, x_county, county, floor, gamma, eps, b),
              scale=sigma_y
          ),
          reinterpreted_batch_ndims=1
      ),
  ])

contextual_effect2 = gen_radon_model(
    u.values,  xbar[county], county, floor_measure)

@tf.function(autograph=False)
def unnormalized_posterior_log_prob(sigma_a, gamma, eps, b, sigma_y):
  """Computes `joint_log_prob` pinned at `log_radon`."""
  return contextual_effect2.log_prob(
      [sigma_a, gamma, eps, b, sigma_y, log_radon])

assert [4] == unnormalized_posterior_log_prob(
    *contextual_effect2.sample(4)[:-1]).shape

In [ ]:

samples = contextual_effect2.sample(4)
pprint([s.shape for s in samples])

[TensorShape([4]),
 TensorShape([4, 85]),
 TensorShape([4, 3]),
 TensorShape([4, 1]),
 TensorShape([4, 1]),
 TensorShape([4, 919])]

In [ ]:

contextual_effect2.log_prob_parts(list(samples)[:-1] + [log_radon])

[<tf.Tensor: shape=(4,), dtype=float64, numpy=array([-3.95681828, -2.45693443, -2.53310078, -4.7717536 ])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-340.65975204, -217.11139018, -246.50498667, -369.79687704])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-20.49822449, -20.38052557, -18.63843525, -17.83096972])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-5.94765605, -5.91460848, -6.66169402, -5.53894593])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-2.10293999, -4.34186631, -2.10744955, -3.016717  ])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=
 array([-29022322.1413861 ,   -114422.36893361,  -8708500.81752865,
           -35061.92497235])>]

변이 추론

JointDistribution*의 매우 강력한 기능 중 하나는 VI에 대한 근사치를 쉽게 생성할 수 있다는 것입니다. 예를 들어, meanfield ADVI를 수행하려면 그래프를 검사하고 관찰되지 않은 모든 분포를 정규 분포로 바꾸기만 하면 됩니다.

Meanfield ADVI

또한 tensorflow_probability/python/experimental/vi의 실험적 기능을 사용하여 아래에서 사용된 것과 본질적으로 동일하지만 무한 공간이 아닌 원래 공간에서 근사 출력을 가진 논리(예: JointDistribution을 사용하여 근사값 빌드)인 변이 근사값을 빌드할 수 있습니다.

In [ ]:

from tensorflow_probability.python.mcmc.transformed_kernel import (
    make_transform_fn, make_transformed_log_prob)

In [ ]:

# Wrap logp so that all parameters are in the Real domain
# copied and edited from tensorflow_probability/python/mcmc/transformed_kernel.py
unconstraining_bijectors = [
    tfb.Exp(),
    tfb.Identity(),
    tfb.Identity(),
    tfb.Identity(),
    tfb.Exp()
]

unnormalized_log_prob = lambda *x: contextual_effect2.log_prob(x + (log_radon,))

contextual_effect_posterior = make_transformed_log_prob(
    unnormalized_log_prob,
    unconstraining_bijectors,
    direction='forward',
    # TODO(b/72831017): Disable caching until gradient linkage
    # generally works.
    enable_bijector_caching=False)

In [ ]:

# debug
if True:
  # Check the two versions of log_prob - they should be different given the Jacobian
  rv_samples = contextual_effect2.sample(4)

  _inverse_transform = make_transform_fn(unconstraining_bijectors, 'inverse')
  _forward_transform = make_transform_fn(unconstraining_bijectors, 'forward')

  pprint([
      unnormalized_log_prob(*rv_samples[:-1]),
      contextual_effect_posterior(*_inverse_transform(rv_samples[:-1])),
      unnormalized_log_prob(
          *_forward_transform(
              tf.zeros_like(a, dtype=dtype) for a in rv_samples[:-1])
      ),
      contextual_effect_posterior(
          *[tf.zeros_like(a, dtype=dtype) for a in rv_samples[:-1]]
      ),
  ])

[<tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1.73354969e+04, -5.51622488e+04, -2.77754609e+08, -1.09065161e+07])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1.73331358e+04, -5.51582029e+04, -2.77754602e+08, -1.09065134e+07])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1992.10420767, -1992.10420767, -1992.10420767, -1992.10420767])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-1992.10420767, -1992.10420767, -1992.10420767, -1992.10420767])>]

In [ ]:

# Build meanfield ADVI for a jointdistribution
# Inspect the input jointdistribution and replace the list of distribution with
# a list of Normal distribution, each with the same shape.
def build_meanfield_advi(jd_list, observed_node=-1):
  """
  The inputted jointdistribution needs to be a batch version
  """
  # Sample to get a list of Tensors
  list_of_values = jd_list.sample(1)  # <== sample([]) might not work
  
  # Remove the observed node
  list_of_values.pop(observed_node)
  
  # Iterate the list of Tensor to a build a list of Normal distribution (i.e.,
  # the Variational posterior)
  distlist = []
  for i, value in enumerate(list_of_values):
    dtype = value.dtype
    rv_shape = value[0].shape
    loc = tf.Variable(
        tf.random.normal(rv_shape, dtype=dtype),
        name='meanfield_%s_mu' % i,
        dtype=dtype)
    scale = tfp.util.TransformedVariable(
        tf.fill(rv_shape, value=tf.constant(0.02, dtype)),
        tfb.Softplus(),
        name='meanfield_%s_scale' % i,
    )

    approx_node = tfd.Normal(loc=loc, scale=scale)
    if loc.shape == ():
      distlist.append(approx_node)
    else:
      distlist.append(
          # TODO: make the reinterpreted_batch_ndims more flexible (for 
          # minibatch etc)
          tfd.Independent(approx_node, reinterpreted_batch_ndims=1)
      )

  # pass list to JointDistribution to initiate the meanfield advi
  meanfield_advi = tfd.JointDistributionSequential(distlist)
  return meanfield_advi

In [ ]:

advi = build_meanfield_advi(contextual_effect2, observed_node=-1)

# Check the logp and logq
advi_samples = advi.sample(4)
pprint([
  advi.log_prob(advi_samples),
  contextual_effect_posterior(*advi_samples)
  ])

[<tf.Tensor: shape=(4,), dtype=float64, numpy=array([231.26836839, 229.40755095, 227.10287879, 224.05914594])>,
 <tf.Tensor: shape=(4,), dtype=float64, numpy=array([-10615.93542431, -11743.21420129, -10376.26732337, -11338.00600103])>]

In [ ]:

opt = tf.optimizers.Adam(learning_rate=.1)

@tf.function(experimental_compile=True)
def run_approximation():
  loss_ = tfp.vi.fit_surrogate_posterior(
        contextual_effect_posterior,
        surrogate_posterior=advi,
        optimizer=opt,
        sample_size=10,
        num_steps=300)
  return loss_

loss_ = run_approximation()

In [ ]:

plt.plot(loss_);
plt.xlabel('iter');
plt.ylabel('loss');

In [ ]:

graph_info = contextual_effect2.resolve_graph()
approx_param = dict()
free_param = advi.trainable_variables
for i, (rvname, param) in enumerate(graph_info[:-1]):
  approx_param[rvname] = {"mu": free_param[i*2].numpy(),
                          "sd": free_param[i*2+1].numpy()}

In [ ]:

approx_param.keys()

dict_keys(['sigma_a', 'eps', 'gamma', 'b', 'sigma_y'])

In [ ]:

approx_param['gamma']

{'mu': array([1.28145814, 0.70365287, 1.02689857]),
 'sd': array([-3.6604972 , -2.68153218, -2.04176524])}

In [ ]:

a_means = (approx_param['gamma']['mu'][0] 
         + approx_param['gamma']['mu'][1]*u.values
         + approx_param['gamma']['mu'][2]*xbar[county]
         + approx_param['eps']['mu'][county])
_, index = np.unique(county, return_index=True)
plt.scatter(u.values[index], a_means[index], color='g')

xvals = np.linspace(-1, 0.8)
plt.plot(xvals, 
         approx_param['gamma']['mu'][0]+approx_param['gamma']['mu'][1]*xvals, 
         'k--')
plt.xlim(-1, 0.8)

plt.xlabel('County-level uranium');
plt.ylabel('Intercept estimate');

In [ ]:

y_est = (approx_param['gamma']['mu'][0] 
        + approx_param['gamma']['mu'][1]*u.values
        + approx_param['gamma']['mu'][2]*xbar[county]
        + approx_param['eps']['mu'][county]
        + approx_param['b']['mu']*floor_measure)

_, ax = plt.subplots(1, 1, figsize=(12, 4))
ax.plot(county, log_radon, 'o', alpha=.25, label='observed')
ax.plot(county, y_est, '-o', lw=2, alpha=.5, label='y_hat')
ax.set_xlim(-1, county.max()+1)
plt.legend(loc='lower right')
ax.set_xlabel('County #')
ax.set_ylabel('log(Uranium) level');

FullRank ADVI

전체 순위 ADVI의 경우, 다변량 가우스를 사용하여 사후값을 근사화하려고 합니다.

In [ ]:

USE_FULLRANK = True

In [ ]:

*prior_tensors, _ = contextual_effect2.sample()

mapper = Mapper(prior_tensors,
                [tfb.Identity() for _ in prior_tensors],
                contextual_effect2.event_shape[:-1])

In [ ]:

rv_shape = ps.shape(mapper.flatten_and_concat(mapper.list_of_tensors))
init_val = tf.random.normal(rv_shape, dtype=dtype)
loc = tf.Variable(init_val, name='loc', dtype=dtype)

if USE_FULLRANK:
  # cov_param = tfp.util.TransformedVariable(
  #     10. * tf.eye(rv_shape[0], dtype=dtype),
  #     tfb.FillScaleTriL(),
  #     name='cov_param'
  #     )
  FillScaleTriL = tfb.FillScaleTriL(
        diag_bijector=tfb.Chain([
          tfb.Shift(tf.cast(.01, dtype)),
          tfb.Softplus(),
          tfb.Shift(tf.cast(np.log(np.expm1(1.)), dtype))]),
        diag_shift=None)
  cov_param = tfp.util.TransformedVariable(
      .02 * tf.eye(rv_shape[0], dtype=dtype), 
      FillScaleTriL,
      name='cov_param')
  advi_approx = tfd.MultivariateNormalTriL(
      loc=loc, scale_tril=cov_param)
else:
  # An alternative way to build meanfield ADVI.
  cov_param = tfp.util.TransformedVariable(
      .02 * tf.ones(rv_shape, dtype=dtype),
      tfb.Softplus(),
      name='cov_param'
      )
  advi_approx = tfd.MultivariateNormalDiag(
      loc=loc, scale_diag=cov_param)

contextual_effect_posterior2 = lambda x: contextual_effect_posterior(
    *mapper.split_and_reshape(x)
)

# Check the logp and logq
advi_samples = advi_approx.sample(7)
pprint([
  advi_approx.log_prob(advi_samples),
  contextual_effect_posterior2(advi_samples)
  ])

[<tf.Tensor: shape=(7,), dtype=float64, numpy=
array([238.81841799, 217.71022639, 234.57207103, 230.0643819 ,
       243.73140943, 226.80149702, 232.85184209])>,
 <tf.Tensor: shape=(7,), dtype=float64, numpy=
array([-3638.93663169, -3664.25879314, -3577.69371677, -3696.25705312,
       -3689.12130489, -3777.53698383, -3659.4982734 ])>]

In [ ]:

learning_rate = tf.optimizers.schedules.ExponentialDecay(
    initial_learning_rate=1e-2,
    decay_steps=10,
    decay_rate=0.99,
    staircase=True)

opt = tf.optimizers.Adam(learning_rate=learning_rate)

@tf.function(experimental_compile=True)
def run_approximation():
  loss_ = tfp.vi.fit_surrogate_posterior(
        contextual_effect_posterior2,
        surrogate_posterior=advi_approx,
        optimizer=opt,
        sample_size=10,
        num_steps=1000)
  return loss_

loss_ = run_approximation()

In [ ]:

plt.plot(loss_);
plt.xlabel('iter');
plt.ylabel('loss');

In [ ]:

# debug
if True:
  _, ax = plt.subplots(1, 2, figsize=(10, 5))
  ax[0].plot(mapper.flatten_and_concat(advi.mean()), advi_approx.mean(), 'o', alpha=.5)
  ax[1].plot(mapper.flatten_and_concat(advi.stddev()), advi_approx.stddev(), 'o', alpha=.5)
  ax[0].set_xlabel('MeanField')
  ax[0].set_ylabel('FullRank')

In [ ]:

graph_info = contextual_effect2.resolve_graph()
approx_param = dict()

free_param_mean = mapper.split_and_reshape(advi_approx.mean())
free_param_std = mapper.split_and_reshape(advi_approx.stddev())
for i, (rvname, param) in enumerate(graph_info[:-1]):
  approx_param[rvname] = {"mu": free_param_mean[i].numpy(),
                          "cov_info": free_param_std[i].numpy()}

In [ ]:

a_means = (approx_param['gamma']['mu'][0] 
         + approx_param['gamma']['mu'][1]*u.values
         + approx_param['gamma']['mu'][2]*xbar[county]
         + approx_param['eps']['mu'][county])
_, index = np.unique(county, return_index=True)
plt.scatter(u.values[index], a_means[index], color='g')

xvals = np.linspace(-1, 0.8)
plt.plot(xvals, 
         approx_param['gamma']['mu'][0]+approx_param['gamma']['mu'][1]*xvals, 
         'k--')
plt.xlim(-1, 0.8)

plt.xlabel('County-level uranium');
plt.ylabel('Intercept estimate');

In [ ]:

y_est = (approx_param['gamma']['mu'][0] 
         + approx_param['gamma']['mu'][1]*u.values
         + approx_param['gamma']['mu'][2]*xbar[county]
         + approx_param['eps']['mu'][county]
         + approx_param['b']['mu']*floor_measure)

_, ax = plt.subplots(1, 1, figsize=(12, 4))
ax.plot(county, log_radon, 'o', alpha=.25, label='observed')
ax.plot(county, y_est, '-o', lw=2, alpha=.5, label='y_hat')
ax.set_xlim(-1, county.max()+1)
plt.legend(loc='lower right')
ax.set_xlabel('County #')
ax.set_ylabel('log(Uranium) level');

베타-베르누이 혼합 모델

여러 검토자가 일부 항목에 레이블을 지정하고 알려지지 않은 (진정한) 잠재적 레이블을 사용하는 혼합 모델입니다.

In [ ]:

dtype = tf.float32

In [ ]:

n = 50000    # number of examples reviewed
p_bad_ = 0.1 # fraction of bad events
m = 5        # number of reviewers for each example
rcl_ = .35 + np.random.rand(m)/10
prc_ = .65 + np.random.rand(m)/10

# PARAMETER TRANSFORMATION
tpr = rcl_
fpr = p_bad_*tpr*(1./prc_-1.)/(1.-p_bad_)
tnr = 1 - fpr

# broadcast to m reviewer.
batch_prob = np.asarray([tpr, fpr]).T
mixture = tfd.Mixture(
    tfd.Categorical(
        probs=[p_bad_, 1-p_bad_]),
    [
        tfd.Independent(tfd.Bernoulli(probs=tpr), 1),
        tfd.Independent(tfd.Bernoulli(probs=fpr), 1),
    ])
# Generate reviewer response
X_tf = mixture.sample([n])

# run once to always use the same array as input
# so we can compare the estimation from different
# inference method.
X_np = X_tf.numpy()

In [ ]:

# batched Mixture model
mdl_mixture = tfd.JointDistributionSequential([
    tfd.Sample(tfd.Beta(5., 2.), m),
    tfd.Sample(tfd.Beta(2., 2.), m),
    tfd.Sample(tfd.Beta(1., 10), 1),
    lambda p_bad, rcl, prc: tfd.Sample(
        tfd.Mixture(
            tfd.Categorical(
                probs=tf.concat([p_bad, 1.-p_bad], -1)),
            [
              tfd.Independent(tfd.Bernoulli(
                  probs=rcl), 1),
              tfd.Independent(tfd.Bernoulli(
                  probs=p_bad*rcl*(1./prc-1.)/(1.-p_bad)), 1)
             ]
      ), (n, )), 
    ])

mdl_mixture.resolve_graph()

(('prc', ()), ('rcl', ()), ('p_bad', ()), ('x', ('p_bad', 'rcl', 'prc')))

In [ ]:

prc, rcl, p_bad, x = mdl_mixture.sample(4)
x.shape

TensorShape([4, 50000, 5])

In [ ]:

mdl_mixture.log_prob_parts([prc, rcl, p_bad, X_np[np.newaxis, ...]])

[<tf.Tensor: shape=(4,), dtype=float32, numpy=array([1.4828572, 2.957961 , 2.9355168, 2.6116824], dtype=float32)>,
 <tf.Tensor: shape=(4,), dtype=float32, numpy=array([-0.14646745,  1.3308513 ,  1.1205603 ,  0.5441705 ], dtype=float32)>,
 <tf.Tensor: shape=(4,), dtype=float32, numpy=array([1.3733709, 1.8020535, 2.1865845, 1.5701319], dtype=float32)>,
 <tf.Tensor: shape=(4,), dtype=float32, numpy=array([-54326.664, -52683.93 , -64407.67 , -55007.895], dtype=float32)>]

추론(NUTS)

In [ ]:

nchain = 10
prc, rcl, p_bad, _ = mdl_mixture.sample(nchain)
initial_chain_state = [prc, rcl, p_bad]

# Since MCMC operates over unconstrained space, we need to transform the
# samples so they live in real-space.
unconstraining_bijectors = [
    tfb.Sigmoid(),       # Maps R to [0, 1].
    tfb.Sigmoid(),       # Maps R to [0, 1].
    tfb.Sigmoid(),       # Maps R to [0, 1].
]
step_size = [tf.cast(i, dtype=dtype) for i in [1e-3, 1e-3, 1e-3]]

X_expanded = X_np[np.newaxis, ...]
target_log_prob_fn = lambda *x: mdl_mixture.log_prob(x + (X_expanded, ))

samples, sampler_stat = run_chain(
    initial_chain_state, step_size, target_log_prob_fn, 
    unconstraining_bijectors, burnin=100)

In [ ]:

# using the pymc3 naming convention
sample_stats_name = ['lp', 'tree_size', 'diverging', 'energy', 'mean_tree_accept']
sample_stats = {k:v.numpy().T for k, v in zip(sample_stats_name, sampler_stat)}
sample_stats['tree_size'] = np.diff(sample_stats['tree_size'], axis=1)

var_name = ['Precision', 'Recall', 'Badness Rate']
posterior = {k:np.swapaxes(v.numpy(), 1, 0) 
             for k, v in zip(var_name, samples)}

az_trace = az.from_dict(posterior=posterior, sample_stats=sample_stats)

In [ ]:

axes = az.plot_trace(az_trace, compact=True);

Copyright 2019 The TensorFlow Authors.

공동 분포를 사용한 베이지안 모델링

종속성과 전제 조건

빠르게 처리하세요!

JointDistribution

(강력한) 선형 회귀

기존 OLS 모델

기타 `JointDistribution*` API

MLE

배치 버전 모델 및 MCMC

No-U-Turn Sampler를 사용하는 MCMC

스튜던트-T 방법

순방향 샘플(사전 예측 샘플링)

MLE

MCMC

계층적 부분 풀링

순방향 샘플(사전 예측 샘플링)

MLE

MCMC

혼합 효과 모델(Radon)

변이 추론

Meanfield ADVI

FullRank ADVI

베타-베르누이 혼합 모델

추론(NUTS)

Product

Resources

Company

Copyright 2019 The TensorFlow Authors.

공동 분포를 사용한 베이지안 모델링

종속성과 전제 조건

빠르게 처리하세요!

JointDistribution

(강력한) 선형 회귀

기존 OLS 모델

기타 JointDistribution* API

MLE

배치 버전 모델 및 MCMC

No-U-Turn Sampler를 사용하는 MCMC

스튜던트-T 방법

순방향 샘플(사전 예측 샘플링)

MLE

MCMC

계층적 부분 풀링

순방향 샘플(사전 예측 샘플링)

MLE

MCMC

혼합 효과 모델(Radon)

변이 추론

Meanfield ADVI

FullRank ADVI

베타-베르누이 혼합 모델

추론(NUTS)

기타 `JointDistribution*` API