TFP Sürüm Notları not defteri (0.12.1)

Bu not defterinin amacı, bazı küçük parçalar aracılığıyla TFP 0.12.1'in "canlanmasına" yardımcı olmaktır - TFP ile elde edebileceğiniz şeylerin küçük demoları.

TensorFlow.org'da görüntüleyin Google Colab'da çalıştırın Kaynağı GitHub'da görüntüleyin Not defterini indir

Yüklemeler ve içe aktarmalar

Bijektörler

Glow

Kağıttan bir bijector ters çevrilebilir 1x1 konvolusyonları ile Üretken Akış: Glow , Kingma ve Dhariwal tarafından.

Bir dağıtımdan nasıl resim çizileceği aşağıda açıklanmıştır (dağıtımın burada hiçbir şey "öğrenmediğini" unutmayın).

image_shape = (32, 32, 4)  # 32 x 32 RGBA image

glow = tfb.Glow(output_shape=image_shape,
                coupling_bijector_fn=tfb.GlowDefaultNetwork,
                exit_bijector_fn=tfb.GlowDefaultExitNetwork)

pz = tfd.Sample(tfd.Normal(0., 1.), tf.reduce_prod(image_shape))

# Calling glow on distribution p(z) creates our glow distribution over images.
px = glow(pz)

# Take samples from the distribution to get images from your dataset.
image = px.sample(1)[0].numpy()

# Rescale to [0, 1].
image = (image - image.min()) / (image.max() - image.min())
plt.imshow(image);

png

RayleighCDF

İçin Bijector Rayleigh dağılımınızın CDF'nin. Bir kullanım, Rayleigh dağılımından tek tip örnekler alarak ve ardından bunları CDF'nin tersinden geçirerek örnekleme yapmaktır.

bij = tfb.RayleighCDF()
uniforms = tfd.Uniform().sample(10_000)
plt.hist(bij.inverse(uniforms), bins='auto');

png

Ascending() yerine Invert(Ordered())

x = tfd.Normal(0., 1.).sample(5)
print(tfb.Ascending()(x))
print(tfb.Invert(tfb.Ordered())(x))
tf.Tensor([1.9363368 2.650928  3.4936204 4.1817293 5.6920815], shape=(5,), dtype=float32)
WARNING:tensorflow:From <ipython-input-5-1406b9939c00>:3: Ordered.__init__ (from tensorflow_probability.python.bijectors.ordered) is deprecated and will be removed after 2021-01-09.
Instructions for updating:
`Ordered` bijector is deprecated; please use `tfb.Invert(tfb.Ascending())` instead.
tf.Tensor([1.9363368 2.650928  3.4936204 4.1817293 5.6920815], shape=(5,), dtype=float32)

Ekle low arg: Softplus(low=2.)

x = tf.linspace(-4., 4., 100)

for low in (-1., 0., 1.):
  bij = tfb.Softplus(low=low)
  plt.plot(x, bij(x));

png

tfb.ScaleMatvecLinearOperatorBlock tümceye destekler LinearOperator , çok parçalı args

op_1 = tf.linalg.LinearOperatorDiag(diag=[1., -1., 3.])
op_2 = tf.linalg.LinearOperatorFullMatrix([[12., 5.], [-1., 3.]])
scale = tf.linalg.LinearOperatorBlockDiag([op_1, op_2], is_non_singular=True)

bij = tfb.ScaleMatvecLinearOperatorBlock(scale)
bij([[1., 2., 3.], [0., 1.]])
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/linalg/linear_operator_block_diag.py:223: LinearOperator.graph_parents (from tensorflow.python.ops.linalg.linear_operator) is deprecated and will be removed in a future version.
Instructions for updating:
Do not call `graph_parents`.
[<tf.Tensor: shape=(3,), dtype=float32, numpy=array([ 1., -2.,  9.], dtype=float32)>,
 <tf.Tensor: shape=(2,), dtype=float32, numpy=array([5., 3.], dtype=float32)>]

dağıtımlar

Skellam

İki farklılıklarından ötürü Dağıtım Poisson RVs. Bu dağılımdan alınan örneklerin negatif olabileceğini unutmayın.

x = tf.linspace(-5., 10., 10 - -5 + 1)

rates = (4, 2)
for i, rate in enumerate(rates):
  plt.bar(x - .3 * (1 - i), tfd.Poisson(rate).prob(x), label=f'Poisson({rate})', alpha=0.5, width=.3)
plt.bar(x.numpy() + .3, tfd.Skellam(*rates).prob(x).numpy(), color='k', alpha=0.25, width=.3,
        label=f'Skellam{rates}')

plt.legend();

png

JointDistributionCoroutine[AutoBatched] üretmek namedtuple benzeri örnekler

Açıkça belirtmek sample_dtype=[...] yaşlı için tuple davranışı.

@tfd.JointDistributionCoroutineAutoBatched
def model():
  x = yield tfd.Normal(0., 1., name='x')
  y = x + 4.
  yield tfd.Normal(y, 1., name='y')

draw = model.sample(10_000)

plt.hist(draw.x, bins='auto', alpha=0.5)
plt.hist(draw.y, bins='auto', alpha=0.5);
WARNING:tensorflow:Note that RandomStandardNormal inside pfor op may not give same output as inside a sequential loop.
WARNING:tensorflow:Note that RandomStandardNormal inside pfor op may not give same output as inside a sequential loop.

png

VonMisesFisher destekleri dim > 5 , entropy()

Von Mises-Fisher dağılımı üzerinde bir dağıtım \(n-1\) boyutlu küre \(\mathbb{R}^n\).

dist = tfd.VonMisesFisher([0., 1, 0, 1, 0, 1], concentration=1.)
draws = dist.sample(3)

print(dist.entropy())
tf.reduce_sum(draws ** 2, axis=1)  # each draw has length 1
tf.Tensor(3.3533673, shape=(), dtype=float32)
<tf.Tensor: shape=(3,), dtype=float32, numpy=array([1.0000002 , 0.99999994, 1.0000001 ], dtype=float32)>

ExpGamma , ExpInverseGamma

log_rate parametre ilave Gamma . Sayısal iyileştirmeler düşük konsantrasyon örnekleme Beta , Dirichlet ve arkadaşları. Her durumda örtük yeniden parametrelendirme gradyanları.

plt.figure(figsize=(10, 3))
plt.subplot(121)
plt.hist(tfd.Beta(.02, .02).sample(10_000), bins='auto')
plt.title('Beta(.02, .02)')
plt.subplot(122)
plt.title('GamX/(GamX+GamY) [the old way]')
g = tfd.Gamma(.02, 1); s0, s1 = g.sample(10_000), g.sample(10_000)
plt.hist(s0 / (s0 + s1), bins='auto')
plt.show()

plt.figure(figsize=(10, 3))
plt.subplot(121)
plt.hist(tfd.ExpGamma(.02, 1.).sample(10_000), bins='auto')
plt.title('ExpGamma(.02, 1)')
plt.subplot(122)
plt.hist(tfb.Log()(tfd.Gamma(.02, 1.)).sample(10_000), bins='auto')
plt.title('tfb.Log()(Gamma(.02, 1)) [the old way]');

png

png

JointDistribution*AutoBatched tekrarlanabilir örnekleme destek (uzunluk-2 başlığın ile / tensör tohumlar)

@tfd.JointDistributionCoroutineAutoBatched
def model():
  x = yield tfd.Normal(0, 1, name='x')
  y = yield tfd.Normal(x + 4, 1, name='y')

print(model.sample(seed=(1, 2)))
print(model.sample(seed=(1, 2)))
StructTuple(
  x=<tf.Tensor: shape=(), dtype=float32, numpy=-0.59835213>,
  y=<tf.Tensor: shape=(), dtype=float32, numpy=6.2380724>
)
StructTuple(
  x=<tf.Tensor: shape=(), dtype=float32, numpy=-0.59835213>,
  y=<tf.Tensor: shape=(), dtype=float32, numpy=6.2380724>
)

KL(VonMisesFisher || SphericalUniform)

# Build vMFs with the same mean direction, batch of increasing concentrations.
vmf = tfd.VonMisesFisher(tf.math.l2_normalize(tf.random.normal([10])),
                         concentration=[0., .1, 1., 10.])
# KL increases with concentration, since vMF(conc=0) == SphericalUniform.
print(tfd.kl_divergence(vmf, tfd.SphericalUniform(10)))
tf.Tensor([4.7683716e-07 4.9877167e-04 4.9384594e-02 2.4844694e+00], shape=(4,), dtype=float32)

parameter_properties

Dağıtım sınıfları artık açığa parameter_properties(dtype=tf.float32, num_classes=None) dağılımların birçok sınıfların otomatik inşaat etkinleştirebilirsiniz sınıf yöntemi.

print('Gamma:', tfd.Gamma.parameter_properties())
print('Categorical:', tfd.Categorical.parameter_properties(dtype=tf.float64, num_classes=7))
Gamma: {'concentration': ParameterProperties(event_ndims=0, shape_fn=<function ParameterProperties.<lambda> at 0x7ff6bbfcdd90>, default_constraining_bijector_fn=<function Gamma._parameter_properties.<locals>.<lambda> at 0x7ff6afd95510>, is_preferred=True), 'rate': ParameterProperties(event_ndims=0, shape_fn=<function ParameterProperties.<lambda> at 0x7ff6bbfcdd90>, default_constraining_bijector_fn=<function Gamma._parameter_properties.<locals>.<lambda> at 0x7ff6afd95ea0>, is_preferred=False), 'log_rate': ParameterProperties(event_ndims=0, shape_fn=<function ParameterProperties.<lambda> at 0x7ff6bbfcdd90>, default_constraining_bijector_fn=<class 'tensorflow_probability.python.bijectors.identity.Identity'>, is_preferred=True)}
Categorical: {'logits': ParameterProperties(event_ndims=1, shape_fn=<function Categorical._parameter_properties.<locals>.<lambda> at 0x7ff6afd95510>, default_constraining_bijector_fn=<class 'tensorflow_probability.python.bijectors.identity.Identity'>, is_preferred=True), 'probs': ParameterProperties(event_ndims=1, shape_fn=<function Categorical._parameter_properties.<locals>.<lambda> at 0x7ff6afdc91e0>, default_constraining_bijector_fn=<class 'tensorflow_probability.python.bijectors.softmax_centered.SoftmaxCentered'>, is_preferred=False)}

experimental_default_event_space_bijector

Artık bazı dağıtım parçalarını sabitleyen ek argümanları kabul ediyor.

@tfd.JointDistributionCoroutineAutoBatched
def model():
  scale = yield tfd.Gamma(1, 1, name='scale')
  obs = yield tfd.Normal(0, scale, name='obs')

model.experimental_default_event_space_bijector(obs=.2).forward(
    [tf.random.uniform([3], -2, 2.)])
StructTuple(
  scale=<tf.Tensor: shape=(3,), dtype=float32, numpy=array([0.6630705, 1.5401832, 1.0777743], dtype=float32)>
)

JointDistribution.experimental_pin

Dönen iğneler bazı ortak dağılım parçaları, JointDistributionPinned ortak normalleştirilmemiş yoğunluğunu temsil eden bir nesne.

Çalışma experimental_default_event_space_bijector , bu mantıklı varsayılanlarla çok daha basit varyasyon çıkarım veya MCMC yapıyor yapar. Aşağıdaki örnekte, ilk iki satır sample yapmak bir esinti MCMC çalışan.

dist = tfd.JointDistributionSequential([
    tfd.HalfNormal(1.),
    lambda scale: tfd.Normal(0., scale, name='observed')])

@tf.function
def sample():
  bij = dist.experimental_default_event_space_bijector(observed=1.)
  target_log_prob = dist.experimental_pin(observed=1.).unnormalized_log_prob

  kernel = tfp.mcmc.TransformedTransitionKernel(
      tfp.mcmc.HamiltonianMonteCarlo(target_log_prob,
                                     step_size=0.6,
                                     num_leapfrog_steps=16),
      bijector=bij)
  return tfp.mcmc.sample_chain(500, 
                               current_state=tf.ones([8]),  # multiple chains
                               kernel=kernel,
                               trace_fn=None)

draws = sample()

fig, (hist, trace) = plt.subplots(ncols=2, figsize=(16, 3))
trace.plot(draws, alpha=0.5)
for col in tf.transpose(draws):
  sns.kdeplot(col, ax=hist);

png

tfd.NegativeBinomial.experimental_from_mean_dispersion

Alternatif parametreleştirme. Diğer dağıtımlar için benzer sınıf yöntemleri eklemek için tfprobability@tensorflow.org adresine e-posta gönderin veya bize bir PR gönderin.

nb = tfd.NegativeBinomial.experimental_from_mean_dispersion(30., .01)
plt.hist(nb.sample(10_000), bins='auto');

png

tfp.experimental.distribute

DistributionStrategy Cihazlar arası olabilirlik hesaplamaları sağlayan, eklem dağılımları-uyumlu. Kanatlı bir Independent ve Sample dağılımları.

# Note: 2-logical devices are configured in the install/import cell at top.
strategy = tf.distribute.MirroredStrategy()
assert strategy.num_replicas_in_sync == 2

@tfp.experimental.distribute.JointDistributionCoroutine
def model():
  root = tfp.experimental.distribute.JointDistributionCoroutine.Root
  group_scale = yield root(tfd.Sample(tfd.Exponential(1), 3, name='group_scale'))
  _ = yield tfp.experimental.distribute.ShardedSample(tfd.Independent(tfd.Normal(0, group_scale), 1),
                                                      sample_shape=[4], name='x')

seed1, seed2 = tfp.random.split_seed((1, 2))

@tf.function
def sample(seed):
  return model.sample(seed=seed)
xs = strategy.run(sample, (seed1,))
print("""
Note that the global latent `group_scale` is shared across devices, whereas
the local `x` is sampled independently on each device.
""")
print('sample:', xs)
print('another sample:', strategy.run(sample, (seed2,)))

@tf.function
def log_prob(x):
  return model.log_prob(x)
print("""
Note that each device observes the same log_prob (local latent log_probs are
summed across devices).
""")
print('log_prob:', strategy.run(log_prob, (xs,)))

@tf.function
def grad_log_prob(x):
  return tfp.math.value_and_gradient(model.log_prob, x)[1]

print("""
Note that each device observes the same log_prob gradient (local latents have
independent gradients, global latents have gradients aggregated across devices).
""")
print('grad_log_prob:', strategy.run(grad_log_prob, (xs,)))
WARNING:tensorflow:There are non-GPU devices in `tf.distribute.Strategy`, not using nccl allreduce.
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:CPU:0', '/job:localhost/replica:0/task:0/device:CPU:1')

Note that the global latent `group_scale` is shared across devices, whereas 
the local `x` is sampled independently on each device.

sample: StructTuple(
  group_scale=PerReplica:{
      0: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([2.6355493, 1.1805456, 1.245112 ], dtype=float32)>,
      1: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([2.6355493, 1.1805456, 1.245112 ], dtype=float32)>
    },
  x=PerReplica:{
      0: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[-0.90548456,  0.7675636 ,  0.27627748],
           [-0.3475989 ,  2.0194046 , -1.2531326 ]], dtype=float32)>,
      1: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[ 3.251305  , -0.5790973 ,  0.42745453],
           [-1.562331  ,  0.3006323 ,  0.635732  ]], dtype=float32)>
    }
)
another sample: StructTuple(
  group_scale=PerReplica:{
      0: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([2.41133   , 0.10307606, 0.5236566 ], dtype=float32)>,
      1: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([2.41133   , 0.10307606, 0.5236566 ], dtype=float32)>
    },
  x=PerReplica:{
      0: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[-3.2476294 ,  0.07213175, -0.39536062],
           [-1.2319602 , -0.05505352,  0.06356457]], dtype=float32)>,
      1: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[ 5.6028705 ,  0.11919801, -0.48446828],
           [-1.5938259 ,  0.21123725,  0.28979057]], dtype=float32)>
    }
)

Note that each device observes the same log_prob (local latent log_probs are
summed across devices).

INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0', '/job:localhost/replica:0/task:0/device:CPU:1').
log_prob: PerReplica:{
  0: tf.Tensor(-25.05747, shape=(), dtype=float32),
  1: tf.Tensor(-25.05747, shape=(), dtype=float32)
}

Note that each device observes the same log_prob gradient (local latents have
independent gradients, global latents are aggregated across devices).

INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0', '/job:localhost/replica:0/task:0/device:CPU:1').
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0', '/job:localhost/replica:0/task:0/device:CPU:1').
grad_log_prob: StructTuple(
  group_scale=PerReplica:{
      0: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([-1.7555585, -1.2928739, -3.0554674], dtype=float32)>,
      1: <tf.Tensor: shape=(3,), dtype=float32, numpy=array([-1.7555585, -1.2928739, -3.0554674], dtype=float32)>
    },
  x=PerReplica:{
      0: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[ 0.13035832, -0.5507428 , -0.17820862],
           [ 0.05004217, -1.4489648 ,  0.80831426]], dtype=float32)>,
      1: <tf.Tensor: shape=(2, 3), dtype=float32, numpy=
    array([[-0.46807498,  0.41551432, -0.27572307],
           [ 0.22492138, -0.21570992, -0.41006932]], dtype=float32)>
    }
)

PSD Çekirdekleri

GeneralizedMatern

GeneralizedMatern pozitif yarı kesin çekirdek genelleştirir MaternOneHalf , MAterhThreeHalves ve MaternFiveHalves .

gm = tfpk.GeneralizedMatern(df=[0.5, 1.5, 2.5], length_scale=1., amplitude=0.5)
m1 = tfpk.MaternOneHalf(length_scale=1., amplitude=0.5)
m2 = tfpk.MaternThreeHalves(length_scale=1., amplitude=0.5)
m3 = tfpk.MaternFiveHalves(length_scale=1., amplitude=0.5)
xs = tf.linspace(-1.5, 1.5, 100)

gm_matrix = gm.matrix([[0.]], xs[..., tf.newaxis])
plt.plot(xs, gm_matrix[0][0])
plt.plot(xs, m1.matrix([[0.]], xs[..., tf.newaxis])[0])
plt.show()
plt.plot(xs, gm_matrix[1][0])
plt.plot(xs, m2.matrix([[0.]], xs[..., tf.newaxis])[0])
plt.show()
plt.plot(xs, gm_matrix[2][0])
plt.plot(xs, m3.matrix([[0.]], xs[..., tf.newaxis])[0])
plt.show()

png

png

png

Parabolic (Epanechnikov)

epa = tfpk.Parabolic()
xs = tf.linspace(-1.05, 1.05, 100)
plt.plot(xs, epa.matrix([[0.]], xs[..., tf.newaxis])[0]);

png

VI

build_asvi_surrogate_posterior

Önceki dağıtımın grafik yapısını içerecek şekilde VI için otomatik olarak yapılandırılmış bir vekil posterior oluşturun. Bu, otomatik Yapısal Varyasyon Çıkarım (kağıt anlatılan yöntem kullanmaktadır https://arxiv.org/abs/2002.00643 ).

# Import a Brownian Motion model from TFP's inference gym.
model = gym.targets.BrownianMotionMissingMiddleObservations()
prior = model.prior_distribution()
ground_truth = ground_truth = model.sample_transformations['identity'].ground_truth_mean
target_log_prob = lambda *values: model.log_likelihood(values) + prior.log_prob(values)

Bu, bir Gauss gözlem modeliyle bir Brownian Hareket sürecini modellemektedir. 30 zaman adımından oluşur, ancak ortadaki 10 zaman adımı gözlemlenemez.

  locs[0] ~ Normal(loc=0, scale=innovation_noise_scale)
  for t in range(1, num_timesteps):
    locs[t] ~ Normal(loc=locs[t - 1], scale=innovation_noise_scale)

  for t in range(num_timesteps):
    observed_locs[t] ~ Normal(loc=locs[t], scale=observation_noise_scale)

Amaç değerlerini anlaması için ise locs gürültülü gözlemlerden (dan observed_locs ). Orta 10 dilimler gözlemlenebilir olduğu, observed_locs olan NaN dilimler [10,19] değerler.

# The observed loc values in the Brownian Motion inference gym model
OBSERVED_LOC = np.array([
    0.21592641, 0.118771404, -0.07945447, 0.037677474, -0.27885845, -0.1484156,
    -0.3250906, -0.22957903, -0.44110894, -0.09830782, np.nan, np.nan, np.nan,
    np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, -0.8786016,
    -0.83736074, -0.7384849, -0.8939254, -0.7774566, -0.70238715, -0.87771565,
    -0.51853573, -0.6948214, -0.6202789
]).astype(dtype=np.float32)

# Plot the prior and the likelihood observations
plt.figure()
plt.title('Brownian Motion Prior Samples and Observations')

num_samples = 15
prior_samples = prior.sample(num_samples)

plt.plot(prior_samples, c='blue', alpha=0.1)
plt.plot(prior_samples[0][0], label="Prior Samples", c='blue', alpha=0.1)

plt.scatter(x=range(30),y=OBSERVED_LOC, c='black', alpha=0.5, label="Observations")

plt.legend(bbox_to_anchor=(1.05, 1), borderaxespad=0.);

png

logging.getLogger('tensorflow').setLevel(logging.ERROR)  # suppress pfor warnings

# Construct and train an ASVI Surrogate Posterior.
asvi_surrogate_posterior = tfp.experimental.vi.build_asvi_surrogate_posterior(prior)

asvi_losses = tfp.vi.fit_surrogate_posterior(target_log_prob,
                                        asvi_surrogate_posterior,
                                        optimizer=tf.optimizers.Adam(learning_rate=0.1),
                                        num_steps=500)
logging.getLogger('tensorflow').setLevel(logging.NOTSET)
# Construct and train a Mean-Field Surrogate Posterior.
factored_surrogate_posterior = tfp.experimental.vi.build_factored_surrogate_posterior(event_shape=prior.event_shape)

factored_losses = tfp.vi.fit_surrogate_posterior(target_log_prob,
                                        factored_surrogate_posterior,
                                        optimizer=tf.optimizers.Adam(learning_rate=0.1),
                                        num_steps=500)
logging.getLogger('tensorflow').setLevel(logging.ERROR)  # suppress pfor warnings

# Sample from the posteriors.
asvi_posterior_samples = asvi_surrogate_posterior.sample(num_samples)
factored_posterior_samples = factored_surrogate_posterior.sample(num_samples)

logging.getLogger('tensorflow').setLevel(logging.NOTSET)

Hem ASVI hem de ortalama alan vekil posterior dağılımları birleşti ve ASVI vekil posterior dağılımları daha düşük bir nihai kayba sahipti (negatif ELBO değeri).

# Plot the loss curves.
plt.figure()
plt.title('Loss Curves for ASVI vs Mean-Field Surrogate Posteriors')

plt.plot(asvi_losses, c='orange', label='ASVI', alpha = 0.4)
plt.plot(factored_losses, c='green', label='Mean-Field', alpha = 0.4)
plt.ylim(-50, 300)

plt.legend(bbox_to_anchor=(1.3, 1), borderaxespad=0.);

png

Posteriorlardan alınan örnekler, ASVI vekil posteriorun gözlemler olmaksızın zaman adımlarının belirsizliğini ne kadar güzel yakaladığını vurgular. Öte yandan, ortalama alan vekil posterior gerçek belirsizliği yakalamak için mücadele eder.

# Plot samples from the ASVI and Mean-Field Surrogate Posteriors.
plt.figure()
plt.title('Posterior Samples from ASVI vs Mean-Field Surrogate Posterior')

plt.plot(asvi_posterior_samples, c='orange', alpha = 0.25)
plt.plot(asvi_posterior_samples[0][0], label='ASVI Surrogate Posterior', c='orange', alpha = 0.25)

plt.plot(factored_posterior_samples, c='green', alpha = 0.25)
plt.plot(factored_posterior_samples[0][0], label='Mean-Field Surrogate Posterior', c='green', alpha = 0.25)

plt.scatter(x=range(30),y=OBSERVED_LOC, c='black', alpha=0.5, label='Observations')

plt.plot(ground_truth, c='black', label='Ground Truth')

plt.legend(bbox_to_anchor=(1.585, 1), borderaxespad=0.);

png

MCMC

ProgressBarReducer

Örnekleyicinin ilerlemesini görselleştirin. (Nominal bir performans cezası olabilir; şu anda JIT derlemesi kapsamında desteklenmemektedir.)

kernel = tfp.mcmc.HamiltonianMonteCarlo(lambda x: -x**2 / 2, .05, 20)
pbar = tfp.experimental.mcmc.ProgressBarReducer(100)
kernel = tfp.experimental.mcmc.WithReductions(kernel, pbar)
plt.hist(tf.reshape(tfp.mcmc.sample_chain(100, current_state=tf.ones([128]), kernel=kernel, trace_fn=None), [-1]), bins='auto')
pbar.bar.close()
99%|█████████▉| 99/100 [00:03<00:00, 27.37it/s]

png

sample_sequential_monte_carlo tekrarlanabilir örnekleme destekler

initial_state = tf.random.uniform([4096], -2., 2.)
def smc(seed):
  return tfp.experimental.mcmc.sample_sequential_monte_carlo(
    prior_log_prob_fn=lambda x: -x**2 / 2,
    likelihood_log_prob_fn=lambda x: -(x-1.)**2 / 2,
    current_state=initial_state,
    seed=seed)[1]
plt.hist(smc(seed=(12, 34)), bins='auto');plt.show()

print(smc(seed=(12, 34))[:10])
print('different:', smc(seed=(10, 20))[:10])
print('same:', smc(seed=(12, 34))[:10])

png

tf.Tensor(
[ 0.665834    0.9892149   0.7961128   1.0016634  -1.000767   -0.19461267
  1.3070581   1.127177    0.9940303   0.58239716], shape=(10,), dtype=float32)
different: tf.Tensor(
[ 1.3284367   0.4374407   1.1349089   0.4557473   0.06510283 -0.08954388
  1.1735026   0.8170528   0.12443061  0.34413314], shape=(10,), dtype=float32)
same: tf.Tensor(
[ 0.665834    0.9892149   0.7961128   1.0016634  -1.000767   -0.19461267
  1.3070581   1.127177    0.9940303   0.58239716], shape=(10,), dtype=float32)

Varyans, kovaryans, Rhat akış hesaplamaları eklendi

Not bunlara arabirimler biraz değişti tfp-nightly .

def cov_to_ellipse(t, cov, mean):
  """Draw a one standard deviation ellipse from the mean, according to cov."""
  diag = tf.linalg.diag_part(cov)
  a = 0.5 * tf.reduce_sum(diag)
  b = tf.sqrt(0.25 * (diag[0] - diag[1])**2 + cov[0, 1]**2)
  major = a + b
  minor = a - b
  theta = tf.math.atan2(major - cov[0, 0], cov[0, 1])
  x = (tf.sqrt(major) * tf.cos(theta) * tf.cos(t) -
       tf.sqrt(minor) * tf.sin(theta) * tf.sin(t))
  y = (tf.sqrt(major) * tf.sin(theta) * tf.cos(t) +
       tf.sqrt(minor) * tf.cos(theta) * tf.sin(t))
  return x + mean[0], y + mean[1]

fig, axes = plt.subplots(nrows=4, ncols=5, figsize=(14, 8), 
                         sharex=True, sharey=True, constrained_layout=True)
t = tf.linspace(0., 2 * np.pi, 200)
tot = 10
cov = 0.1 * tf.eye(2) + 0.9 * tf.ones([2, 2])
mvn = tfd.MultivariateNormalTriL(loc=[1., 2.],
                                 scale_tril=tf.linalg.cholesky(cov))

for ax in axes.ravel():
  rv = tfp.experimental.stats.RunningCovariance(
      num_samples=0., mean=tf.zeros(2), sum_squared_residuals=tf.zeros((2, 2)),
      event_ndims=1)

  for idx, x in enumerate(mvn.sample(tot)):
    rv = rv.update(x)
    ax.plot(*cov_to_ellipse(t, rv.covariance(), rv.mean),
            color='k', alpha=(idx + 1) / tot)
  ax.plot(*cov_to_ellipse(t, mvn.covariance(), mvn.mean()), 'r')
fig.suptitle("Twenty tries to approximate the red covariance with 10 draws");

png

Matematik, istatistikler

Bessel fonksiyonları: ive, kve, log-ive

xs = tf.linspace(0.5, 20., 100)
ys = tfp.math.bessel_ive([[0.5], [1.], [np.pi], [4.]], xs)
zs = tfp.math.bessel_kve([[0.5], [1.], [2.], [np.pi]], xs)

for i in range(4):
  plt.plot(xs, ys[i])
plt.show()

for i in range(4):
  plt.plot(xs, zs[i])
plt.show()

png

png

İsteğe bağlı weights için arg tfp.stats.histogram

edges = tf.linspace(-4., 4, 31)
samps = tfd.TruncatedNormal(0, 1, -4, 4).sample(100_000, seed=(123, 456))
_, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 3))
ax1.bar(edges[:-1], tfp.stats.histogram(samps, edges))
ax1.set_title('samples histogram')
ax2.bar(edges[:-1], tfp.stats.histogram(samps, edges, weights=1 / tfd.Normal(0, 1).prob(samps)))
ax2.set_title('samples, weighted by inverse p(sample)');

png

tfp.math.erfcinv

x = tf.linspace(-3., 3., 10)
y = tf.math.erfc(x)
z = tfp.math.erfcinv(y)
print(x)
print(z)
tf.Tensor(
[-3.         -2.3333333  -1.6666666  -1.         -0.33333325  0.3333335

  1.          1.666667    2.3333335   3.        ], shape=(10,), dtype=float32)
tf.Tensor(
[-3.0002644  -2.3333426  -1.6666666  -0.9999997  -0.3333332   0.33333346
  0.9999999   1.6666667   2.3333335   3.0000002 ], shape=(10,), dtype=float32)