diff --git a/examples/m5.ipynb b/examples/m5.ipynb index dca854e..7c80692 100644 --- a/examples/m5.ipynb +++ b/examples/m5.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 17, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -44,17 +44,17 @@ "metadata": {}, "outputs": [], "source": [ - "dataset = get_dataset(\"m5\", regenerate=False)" + "dataset = get_dataset(\"m4_hourly\", regenerate=False)" ] }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 6, "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -77,12 +77,12 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 7, "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -113,8 +113,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "Recommended prediction horizon: 28\n", - "Frequency of the time series: D\n" + "Recommended prediction horizon: 48\n", + "Frequency of the time series: H\n" ] } ], @@ -130,34 +130,34 @@ "outputs": [], "source": [ "from pts.model.deepar import DeepAREstimator\n", - "from pts.modules import ZeroInflatedNegativeBinomialOutput\n", + "from pts.modules import ZeroInflatedNegativeBinomialOutput, PiecewiseLinearOutput\n", "from pts import Trainer" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "estimator = DeepAREstimator(\n", - " distr_output=ZeroInflatedNegativeBinomialOutput(),\n", + " distr_output=PiecewiseLinearOutput(num_pieces=10),\n", " cell_type='GRU',\n", - " input_size=72,\n", + " input_size=48,\n", " num_cells=64,\n", " num_layers=3,\n", " dropout_rate=0.2,\n", - " use_feat_dynamic_real=True,\n", - " use_feat_static_cat=True,\n", - " cardinality=[int(cat_feat_info.cardinality) for cat_feat_info in dataset.metadata.feat_static_cat],\n", - " embedding_dimension = [4, 4, 4, 4, 16],\n", + " use_feat_dynamic_real=False,\n", + " use_feat_static_cat=False,\n", + " #cardinality=[int(cat_feat_info.cardinality) for cat_feat_info in dataset.metadata.feat_static_cat],\n", + " #embedding_dimension = [4, 4, 4, 4, 16],\n", " prediction_length=dataset.metadata.prediction_length,\n", " context_length=dataset.metadata.prediction_length*2,\n", " freq=dataset.metadata.freq,\n", " scaling=True,\n", " trainer=Trainer(device=device,\n", " epochs=50,\n", - " learning_rate=1e-3,\n", + " learning_rate=1e-5,\n", " num_batches_per_epoch=120,\n", " batch_size=256,\n", " num_workers=8,\n", @@ -168,63 +168,14 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "119it [00:29, 4.09it/s, avg_epoch_loss=1.17, epoch=0]\n", - "119it [00:31, 3.81it/s, avg_epoch_loss=1.14, epoch=1]\n", - "119it [00:27, 4.29it/s, avg_epoch_loss=1.12, epoch=2]\n", - "119it [00:29, 4.06it/s, avg_epoch_loss=1.11, epoch=3]\n", - "119it [00:28, 4.17it/s, avg_epoch_loss=1.1, epoch=4]\n", - "119it [00:28, 4.14it/s, avg_epoch_loss=1.1, epoch=5]\n", - "119it [00:30, 3.93it/s, avg_epoch_loss=1.1, epoch=6] \n", - "119it [00:27, 4.27it/s, avg_epoch_loss=1.11, epoch=7]\n", - "119it [00:28, 4.16it/s, avg_epoch_loss=1.09, epoch=8]\n", - "119it [00:29, 4.09it/s, avg_epoch_loss=1.11, epoch=9]\n", - "119it [00:27, 4.40it/s, avg_epoch_loss=1.1, epoch=10]\n", - "119it [00:28, 4.23it/s, avg_epoch_loss=1.1, epoch=11]\n", - "119it [00:28, 4.12it/s, avg_epoch_loss=1.1, epoch=12] \n", - "119it [00:29, 4.06it/s, avg_epoch_loss=1.1, epoch=13]\n", - "119it [00:29, 4.10it/s, avg_epoch_loss=1.11, epoch=14]\n", - "119it [00:28, 4.24it/s, avg_epoch_loss=1.1, epoch=15] \n", - "119it [00:30, 3.95it/s, avg_epoch_loss=1.1, epoch=16]\n", - "119it [00:27, 4.28it/s, avg_epoch_loss=1.09, epoch=17]\n", - "119it [00:27, 4.26it/s, avg_epoch_loss=1.1, epoch=18]\n", - "119it [00:29, 4.07it/s, avg_epoch_loss=1.1, epoch=19] \n", - "119it [00:29, 3.98it/s, avg_epoch_loss=1.09, epoch=20]\n", - "119it [00:27, 4.33it/s, avg_epoch_loss=1.1, epoch=21]\n", - "119it [00:29, 4.08it/s, avg_epoch_loss=1.09, epoch=22]\n", - "119it [00:29, 4.09it/s, avg_epoch_loss=1.09, epoch=23]\n", - "119it [00:28, 4.22it/s, avg_epoch_loss=1.09, epoch=24]\n", - "119it [00:27, 4.26it/s, avg_epoch_loss=1.09, epoch=25]\n", - "119it [00:31, 3.81it/s, avg_epoch_loss=1.1, epoch=26]\n", - "119it [00:31, 3.73it/s, avg_epoch_loss=1.09, epoch=27]\n", - "119it [00:27, 4.32it/s, avg_epoch_loss=1.08, epoch=28]\n", - "119it [00:28, 4.14it/s, avg_epoch_loss=1.09, epoch=29]\n", - "119it [00:30, 3.87it/s, avg_epoch_loss=1.08, epoch=30]\n", - "119it [00:28, 4.19it/s, avg_epoch_loss=1.09, epoch=31]\n", - "119it [00:28, 4.17it/s, avg_epoch_loss=1.08, epoch=32]\n", - "119it [00:29, 4.09it/s, avg_epoch_loss=1.1, epoch=33] \n", - "119it [00:27, 4.39it/s, avg_epoch_loss=1.09, epoch=34]\n", - "119it [00:28, 4.21it/s, avg_epoch_loss=1.09, epoch=35]\n", - "119it [00:28, 4.16it/s, avg_epoch_loss=1.09, epoch=36]\n", - "119it [00:27, 4.31it/s, avg_epoch_loss=1.08, epoch=37]\n", - "119it [00:29, 4.07it/s, avg_epoch_loss=1.09, epoch=38]\n", - "119it [00:28, 4.19it/s, avg_epoch_loss=1.09, epoch=39]\n", - "119it [00:29, 4.06it/s, avg_epoch_loss=1.09, epoch=40]\n", - "119it [00:28, 4.14it/s, avg_epoch_loss=1.08, epoch=41]\n", - "119it [00:28, 4.16it/s, avg_epoch_loss=1.09, epoch=42]\n", - "119it [00:27, 4.25it/s, avg_epoch_loss=1.09, epoch=43]\n", - "119it [00:27, 4.26it/s, avg_epoch_loss=1.1, epoch=44]\n", - "119it [00:26, 4.41it/s, avg_epoch_loss=1.09, epoch=45]\n", - "119it [00:27, 4.25it/s, avg_epoch_loss=1.08, epoch=46]\n", - "119it [00:28, 4.20it/s, avg_epoch_loss=1.09, epoch=47]\n", - "119it [00:30, 3.92it/s, avg_epoch_loss=1.09, epoch=48]\n", - "119it [00:30, 3.96it/s, avg_epoch_loss=1.09, epoch=49]\n" + "21it [00:16, 1.45it/s, avg_epoch_loss=nan, epoch=0] " ] } ], @@ -390,7 +341,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.7.4" + "version": "3.8.5" } }, "nbformat": 4, diff --git a/pts/distributions/__init__.py b/pts/distributions/__init__.py index d2f592c..b369e84 100644 --- a/pts/distributions/__init__.py +++ b/pts/distributions/__init__.py @@ -4,3 +4,4 @@ from .zero_inflated import ( ZeroInflatedPoisson, ZeroInflatedNegativeBinomial, ) +from .piecewise_linear import PiecewiseLinear, TransformedPiecewiseLinear diff --git a/pts/distributions/piecewise_linear.py b/pts/distributions/piecewise_linear.py new file mode 100644 index 0000000..8f53ceb --- /dev/null +++ b/pts/distributions/piecewise_linear.py @@ -0,0 +1,143 @@ +# Copyright 2018 Amazon.com, Inc. or its affiliates. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"). +# You may not use this file except in compliance with the License. +# A copy of the License is located at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# or in the "license" file accompanying this file. This file 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. + +import torch +import torch.nn.functional as F +from torch.distributions import ( + constraints, + NegativeBinomial, + Poisson, + Distribution, + TransformedDistribution, + AffineTransform, +) +from torch.distributions.utils import broadcast_all, lazy_property + +from .utils import broadcast_shape + + +class PiecewiseLinear(Distribution): + def __init__(self, gamma, slopes, knot_spacings, validate_args=None): + self.gamma = gamma + self.slopes = slopes + self.knot_spacings = knot_spacings + + self.b, self.knot_positions = PiecewiseLinear._to_orig_params( + slopes=slopes, knot_spacings=knot_spacings + ) + super(PiecewiseLinear, self).__init__( + batch_shape=self.gamma.shape, validate_args=validate_args + ) + + @staticmethod + def _to_orig_params(slopes, knot_spacings): + # b: the difference between slopes of consecutive pieces + b = slopes[..., 1:] - slopes[..., 0:-1] + + # Add slope of first piece to b: b_0 = m_0 + m_0 = slopes[..., 0:1] + b = torch.cat((m_0, b), dim=-1) + + # The actual position of the knots is obtained by cumulative sum of + # the knot spacings. The first knot position is always 0 for quantile + # functions. + knot_positions = torch.cumsum(knot_spacings, dim=-1) - knot_spacings + + return b, knot_positions + + @torch.no_grad() + def sample(self, sample_shape=torch.Size()): + shape = self._extended_shape(sample_shape) + u = torch.rand_like(self.gamma.expand(shape)) + + sample = self.quantile(u) + + if len(sample_shape) == 0: + sample = sample.squeeze(0) + + return sample + + def quantile(self, level): + return self.quantile_internal(level, dim=0) + + def quantile_internal(self, x, dim=None): + if dim is not None: + gamma = self.gamma.unsqueeze(dim=dim if dim == 0 else -1) + knot_positions = self.knot_positions.unsqueeze(dim) + b = self.b.unsqueeze(dim) + else: + gamma, knot_positions, b = self.gamma, self.knot_positions, self.b + + x_minus_knots = x.unsqueeze(-1) - knot_positions + + quantile = gamma + (b * F.relu(x_minus_knots)).sum(-1) + + return quantile + + def cdf(self, x): + gamma, b, knot_positions = self.gamma, self.b, self.knot_positions + + quantiles_at_knots = self.quantile_internal(knot_positions, dim=-2) + + # Mask to nullify the terms corresponding to knots larger than l_0, + # which is the largest knot (quantile level) such that the quantile + # at l_0, s(l_0) < x. + mask = torch.le(quantiles_at_knots, x.unsqueeze(-1)) + + slope_l0 = (b * mask).sum(-1) + + # slope_l0 can be zero in which case a_tilde = 0. + # The following is to circumvent an issue where the + # backward() returns nans when slope_l0 is zero in the where + slope_l0_nz = torch.where(slope_l0 == 0.0, torch.ones_like(x), slope_l0) + + a_tilde = torch.where( + slope_l0 == 0.0, + torch.zeros_like(x), + (x - gamma + (b * knot_positions * mask).sum(-1)) / slope_l0_nz, + ) + + return torch.clamp(a_tilde, min=0.0, max=1.0) + + def crps(self, x): + gamma, b, knot_positions = self.gamma, self.b, self.knot_positions + + a_tilde = self.cdf(x) + + max_a_tilde_knots = torch.max(a_tilde.unsqueeze(-1), knot_positions) + + knots_cubed = torch.pow(knot_positions, 3.0) + coeff = ( + (1.0 - knots_cubed) / 3.0 + - knot_positions + - torch.square(max_a_tilde_knots) + + 2 * max_a_tilde_knots * knot_positions + ) + + return (2 * a_tilde - 1) * x + (1 - 2 * a_tilde) * gamma + (b * coeff).sum(-1) + + +class TransformedPiecewiseLinear(TransformedDistribution): + def __init__(self, base_distribution, transforms): + super().__init__(base_distribution, transforms) + + def crps(self, x): + scale = 1.0 + + for transform in reversed(self.transforms): + assert isinstance(transform, AffineTransform), "Not an AffineTransform" + x = transform.inv(x) + scale *= transform.scale + + p = self.base_dist.crps(x) + return p * scale diff --git a/pts/modules/__init__.py b/pts/modules/__init__.py index fa20d1f..b654665 100644 --- a/pts/modules/__init__.py +++ b/pts/modules/__init__.py @@ -7,6 +7,7 @@ from .distribution_output import ( BetaOutput, PoissonOutput, ZeroInflatedPoissonOutput, + PiecewiseLinearOutput, NegativeBinomialOutput, ZeroInflatedNegativeBinomialOutput, NormalMixtureOutput, diff --git a/pts/modules/distribution_output.py b/pts/modules/distribution_output.py index ce61536..1f8f15c 100644 --- a/pts/modules/distribution_output.py +++ b/pts/modules/distribution_output.py @@ -22,7 +22,12 @@ from torch.distributions import ( Poisson, ) -from pts.distributions import ZeroInflatedPoisson, ZeroInflatedNegativeBinomial +from pts.distributions import ( + ZeroInflatedPoisson, + ZeroInflatedNegativeBinomial, + PiecewiseLinear, + TransformedPiecewiseLinear, +) from pts.core.component import validated from .lambda_layer import LambdaLayer @@ -333,6 +338,45 @@ class StudentTMixtureOutput(DistributionOutput): return () +class PiecewiseLinearOutput(DistributionOutput): + distr_cls: type = PiecewiseLinear + + @validated() + def __init__(self, num_pieces: int) -> None: + super().__init__(self) + assert ( + isinstance(num_pieces, int) and num_pieces > 1 + ), "num_pieces should be an integer larger than 1" + + self.num_pieces = num_pieces + self.args_dim = {"gamma": 1, "slopes": num_pieces, "knot_spacings": num_pieces} + + @classmethod + def domain_map(cls, gamma, slopes, knot_spacings): + # slopes of the pieces are non-negative + slopes_proj = F.softplus(slopes) + 1e-4 + + # the spacing between the knots should be in [0, 1] and sum to 1 + knot_spacings_proj = torch.softmax(knot_spacings, dim=-1) + + return gamma.squeeze(axis=-1), slopes_proj, knot_spacings_proj + + def distribution( + self, distr_args, scale: Optional[torch.Tensor] = None, + ) -> PiecewiseLinear: + if scale is None: + return self.distr_cls(*distr_args) + else: + distr = self.distr_cls(*distr_args) + return TransformedPiecewiseLinear( + distr, [AffineTransform(loc=0, scale=scale)] + ) + + @property + def event_shape(self) -> Tuple: + return () + + class NormalMixtureOutput(DistributionOutput): @validated() def __init__(self, components: int = 1) -> None: diff --git a/test/distributions/test_piecewise_linear.py b/test/distributions/test_piecewise_linear.py new file mode 100644 index 0000000..de1e1f3 --- /dev/null +++ b/test/distributions/test_piecewise_linear.py @@ -0,0 +1,209 @@ +# Copyright 2018 Amazon.com, Inc. or its affiliates. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"). +# You may not use this file except in compliance with the License. +# A copy of the License is located at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# or in the "license" file accompanying this file. This file 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. + +from typing import Tuple, List +import pytest + +import torch +import numpy as np + +from pts.distributions import PiecewiseLinear +from pts.modules import PiecewiseLinearOutput + + +def empirical_cdf( + samples: np.ndarray, num_bins: int = 100 +) -> Tuple[np.ndarray, np.ndarray]: + """ + Calculate the empirical cdf from the given samples. + Parameters + ---------- + samples + Tensor of samples of shape (num_samples, batch_shape) + Returns + ------- + Tensor + Empirically calculated cdf values. shape (num_bins, batch_shape) + Tensor + Bin edges corresponding to the cdf values. shape (num_bins + 1, batch_shape) + """ + + # calculate histogram separately for each dimension in the batch size + cdfs = [] + edges = [] + + batch_shape = samples.shape[1:] + agg_batch_dim = np.prod(batch_shape, dtype=np.int) + + samples = samples.reshape((samples.shape[0], -1)) + + for i in range(agg_batch_dim): + s = samples[:, i] + bins = np.linspace(s.min(), s.max(), num_bins + 1) + hist, edge = np.histogram(s, bins=bins) + cdfs.append(np.cumsum(hist / len(s))) + edges.append(edge) + + empirical_cdf = np.stack(cdfs, axis=-1).reshape(num_bins, *batch_shape) + edges = np.stack(edges, axis=-1).reshape(num_bins + 1, *batch_shape) + return empirical_cdf, edges + + +@pytest.mark.parametrize( + "distr, target, expected_target_cdf, expected_target_crps", + [ + ( + PiecewiseLinear( + gamma=torch.ones(size=(1,)), + slopes=torch.Tensor([2, 3, 1]).reshape(shape=(1, 3)), + knot_spacings=torch.Tensor([0.3, 0.4, 0.3]).reshape(shape=(1, 3)), + ), + [2.2], + [0.5], + [0.223000], + ), + ( + PiecewiseLinear( + gamma=torch.ones(size=(2,)), + slopes=torch.Tensor([[1, 1], [1, 2]]).reshape(shape=(2, 2)), + knot_spacings=torch.Tensor([[0.4, 0.6], [0.4, 0.6]]).reshape( + shape=(2, 2) + ), + ), + [1.5, 1.6], + [0.5, 0.5], + [0.083333, 0.145333], + ), + ], +) +def test_values( + distr: PiecewiseLinear, + target: List[float], + expected_target_cdf: List[float], + expected_target_crps: List[float], +): + target = torch.Tensor(target).reshape(shape=(len(target),)) + expected_target_cdf = np.array(expected_target_cdf).reshape( + (len(expected_target_cdf),) + ) + expected_target_crps = np.array(expected_target_crps).reshape( + (len(expected_target_crps),) + ) + + assert all(np.isclose(distr.cdf(target).numpy(), expected_target_cdf)) + assert all(np.isclose(distr.crps(target).numpy(), expected_target_crps)) + + # compare with empirical cdf from samples + num_samples = 100_000 + samples = distr.sample((num_samples,)).numpy() + assert np.isfinite(samples).all() + + emp_cdf, edges = empirical_cdf(samples) + calc_cdf = distr.cdf(torch.Tensor(edges)).numpy() + assert np.allclose(calc_cdf[1:, :], emp_cdf, atol=1e-2) + + +@pytest.mark.parametrize( + "batch_shape, num_pieces, num_samples", + [((3, 4, 5), 10, 100), ((1,), 2, 1), ((10,), 10, 10), ((10, 5), 2, 1)], +) +def test_shapes(batch_shape: Tuple, num_pieces: int, num_samples: int): + gamma = torch.ones(size=(*batch_shape,)) + slopes = torch.ones(size=(*batch_shape, num_pieces)) # all positive + knot_spacings = ( + torch.ones(size=(*batch_shape, num_pieces)) / num_pieces + ) # positive and sum to 1 + target = torch.ones(size=batch_shape) # shape of gamma + + distr = PiecewiseLinear(gamma=gamma, slopes=slopes, knot_spacings=knot_spacings) + + # assert that the parameters and target have proper shapes + assert gamma.shape == target.shape + assert knot_spacings.shape == slopes.shape + assert len(gamma.shape) + 1 == len(knot_spacings.shape) + + # assert that batch_shape is computed properly + assert distr.batch_shape == batch_shape + + # assert that shapes of original parameters are correct + assert distr.b.shape == slopes.shape + assert distr.knot_positions.shape == knot_spacings.shape + + # assert that the shape of crps is correct + assert distr.crps(target).shape == batch_shape + + # assert that the quantile shape is correct when computing the + # quantile values at knot positions - used for a_tilde + assert distr.quantile_internal(knot_spacings, dim=-2).shape == ( + *batch_shape, + num_pieces, + ) + + # assert that the samples and the quantile values shape when num_samples + # is None is correct + samples = distr.sample() + assert samples.shape == batch_shape + assert distr.quantile_internal(samples).shape == batch_shape + + # assert that the samples and the quantile values shape when num_samples + # is not None is correct + samples = distr.sample((num_samples,)) + assert samples.shape == (num_samples, *batch_shape) + assert distr.quantile_internal(samples, dim=0).shape == (num_samples, *batch_shape,) + + +def test_simple_symmetric(): + gamma = torch.Tensor([-1.0]) + slopes = torch.Tensor([[2.0, 2.0]]) + knot_spacings = torch.Tensor([[0.5, 0.5]]) + + distr = PiecewiseLinear(gamma=gamma, slopes=slopes, knot_spacings=knot_spacings) + + assert distr.cdf(torch.Tensor([-2.0])).numpy().item() == 0.0 + assert distr.cdf(torch.Tensor([+2.0])).numpy().item() == 1.0 + + expected_crps = np.array([1.0 + 2.0 / 3.0]) + + assert np.allclose(distr.crps(torch.Tensor([-2.0])).numpy(), expected_crps) + + assert np.allclose(distr.crps(torch.Tensor([2.0])).numpy(), expected_crps) + + +def test_robustness(): + distr_out = PiecewiseLinearOutput(num_pieces=10) + args_proj = distr_out.get_args_proj(in_features=30) + + net_out = torch.normal(mean=0.0, size=(1000, 30), std=1e2) + gamma, slopes, knot_spacings = args_proj(net_out) + distr = distr_out.distribution((gamma, slopes, knot_spacings)) + + # compute the 1-quantile (the 0-quantile is gamma) + sup_support = gamma + (slopes * knot_spacings).sum(-1) + + assert torch.le(gamma, sup_support).numpy().all() + + width = sup_support - gamma + x = torch.from_numpy( + np.random.uniform( + low=(gamma - width).detach().numpy(), + high=(sup_support + width).detach().numpy(), + ).astype(np.float32), + ) + + # check that 0 < cdf < 1 + cdf_x = distr.cdf(x) + assert torch.min(cdf_x).item() >= 0.0 and torch.max(cdf_x).item() <= 1.0 + + # check that 0 <= crps + crps_x = distr.crps(x) + assert torch.min(crps_x).item() >= 0.0 diff --git a/test/distributions/test_zero_inflated.py b/test/distributions/test_zero_inflated.py new file mode 100644 index 0000000..3609f9f --- /dev/null +++ b/test/distributions/test_zero_inflated.py @@ -0,0 +1,91 @@ +# Copyright (c) 2017-2019 Uber Technologies, Inc. +# SPDX-License-Identifier: Apache-2.0 + +import pytest +import torch + +from torch.distributions import ( + NegativeBinomial, + Normal, + Poisson, +) +from pts.distributions import ( + ZeroInflatedDistribution, + ZeroInflatedNegativeBinomial, + ZeroInflatedPoisson, + broadcast_shape, +) + +from numpy.testing import assert_allclose as assert_close + + +@pytest.mark.parametrize("gate_shape", [(), (2,), (3, 1), (3, 2)]) +@pytest.mark.parametrize("base_shape", [(), (2,), (3, 1), (3, 2)]) +def test_zid_shape(gate_shape, base_shape): + gate = torch.rand(gate_shape) + base_dist = Normal(torch.randn(base_shape), torch.randn(base_shape).exp()) + + d = ZeroInflatedDistribution(gate, base_dist) + assert d.batch_shape == broadcast_shape(gate_shape, base_shape) + assert d.support == base_dist.support + + d2 = d.expand([4, 3, 2]) + assert d2.batch_shape == (4, 3, 2) + + +@pytest.mark.parametrize("rate", [0.1, 0.5, 0.9, 1.0, 1.1, 2.0, 10.0]) +def test_zip_0_gate(rate): + # if gate is 0 ZIP is Poisson + zip_ = ZeroInflatedPoisson(torch.zeros(1), torch.tensor(rate)) + pois = Poisson(torch.tensor(rate)) + s = pois.sample((20,)) + zip_prob = zip_.log_prob(s) + pois_prob = pois.log_prob(s) + assert_close(zip_prob, pois_prob, atol=1e-06) + + +@pytest.mark.parametrize("gate", [0.0, 0.25, 0.5, 0.75, 1.0]) +@pytest.mark.parametrize("rate", [0.1, 0.5, 0.9, 1.0, 1.1, 2.0, 10.0]) +def test_zip_mean_variance(gate, rate): + num_samples = 1000000 + zip_ = ZeroInflatedPoisson(torch.tensor(gate), torch.tensor(rate)) + s = zip_.sample((num_samples,)) + expected_mean = zip_.mean + estimated_mean = s.mean() + expected_std = zip_.stddev + estimated_std = s.std() + assert_close(expected_mean, estimated_mean, atol=1e-02) + assert_close(expected_std, estimated_std, atol=1e-02) + + +@pytest.mark.parametrize("total_count", [0.1, 0.5, 0.9, 1.0, 1.1, 2.0, 10.0]) +@pytest.mark.parametrize("probs", [0.1, 0.5, 0.9]) +def test_zinb_0_gate(total_count, probs): + # if gate is 0 ZINB is NegativeBinomial + zinb_ = ZeroInflatedNegativeBinomial( + torch.zeros(1), total_count=torch.tensor(total_count), probs=torch.tensor(probs) + ) + neg_bin = NegativeBinomial(torch.tensor(total_count), probs=torch.tensor(probs)) + s = neg_bin.sample((20,)) + zinb_prob = zinb_.log_prob(s) + neg_bin_prob = neg_bin.log_prob(s) + assert_close(zinb_prob, neg_bin_prob, atol=1e-06) + + +@pytest.mark.parametrize("gate", [0.0, 0.25, 0.5, 0.75, 1.0]) +@pytest.mark.parametrize("total_count", [0.1, 0.5, 0.9, 1.0, 1.1, 2.0, 10.0]) +@pytest.mark.parametrize("logits", [-0.5, 0.5, -0.9, 1.9]) +def test_zinb_mean_variance(gate, total_count, logits): + num_samples = 1000000 + zinb_ = ZeroInflatedNegativeBinomial( + torch.tensor(gate), + total_count=torch.tensor(total_count), + logits=torch.tensor(logits), + ) + s = zinb_.sample((num_samples,)) + expected_mean = zinb_.mean + estimated_mean = s.mean() + expected_std = zinb_.stddev + estimated_std = s.std() + assert_close(expected_mean, estimated_mean, atol=1e-01) + assert_close(expected_std, estimated_std, atol=1e-1)