Merge pull request #1245 from NeilGirdhar:master

PiperOrigin-RevId: 380851764

Author: tensorflower-gardener

spinoffs/inference_gym/inference_gym/internal/datasets/synthetic_item_response_theory.py

@@ -33,7 +33,7 @@
 question_ids = np.arange(100)
 
 student_ids, question_ids = np.meshgrid(student_ids, question_ids)
-mask = rng.binomial(1, 0.75, size=student_ids.shape).astype(np.bool)
+mask = rng.binomial(1, 0.75, size=student_ids.shape).astype(np.bool_)
 
 student_ids = student_ids[mask]
 question_ids = question_ids[mask]

spinoffs/inference_gym/inference_gym/targets/item_response_theory.py

@@ -252,7 +252,7 @@ def _sparse_to_dense(self, student_ids, question_ids, correct):
     # tensor-valued datasets. Blocked by JAX/Numpy not implementing scatter_nd.
     dense_y = np.zeros([self._num_students, self._num_questions], np.float32)
     dense_y[student_ids, question_ids] = correct
-    y_mask = np.zeros(dense_y.shape, np.bool)
+    y_mask = np.zeros(dense_y.shape, np.bool_)
     y_mask[student_ids, question_ids] = True
     return dense_y, y_mask
 

tensorflow_probability/python/bijectors/affine_test.py

@@ -611,8 +611,8 @@ def dynamic_run(fun, x_value, **kwargs):
           run(bijector.forward_log_det_jacobian, x, event_ndims=1))
 
   def _testScaledIdentityComplexAdjoint(self, is_dynamic):
-    shift_ = np.array(-0.5, dtype=np.complex)
-    scale_ = np.array(4 + 2j, dtype=np.complex)
+    shift_ = np.array(-0.5, dtype=np.complex128)
+    scale_ = np.array(4 + 2j, dtype=np.complex128)
     shift = tf1.placeholder_with_default(
         shift_, shape=None if is_dynamic else [])
     scale = tf1.placeholder_with_default(
@@ -622,7 +622,7 @@ def _testScaledIdentityComplexAdjoint(self, is_dynamic):
         scale_identity_multiplier=scale,
         adjoint=True,
         validate_args=True)
-    z = np.array([1., 2, 3], dtype=np.complex)
+    z = np.array([1., 2, 3], dtype=np.complex128)
     y = bijector.forward(z)
     x = bijector.inverse(z)
     inv_fwd_z = bijector.inverse(tf.identity(y))

tensorflow_probability/python/bijectors/kumaraswamy_cdf_test.py

@@ -63,7 +63,7 @@ def testBijectorConcentration1LogDetJacobianFiniteAtZero(self):
         concentration1=1., concentration0=concentration0)
     ildj = self.evaluate(
         bijector.inverse_log_det_jacobian(0., event_ndims=0))
-    self.assertAllEqual(np.ones_like(ildj, dtype=np.bool), np.isfinite(ildj))
+    self.assertAllEqual(np.ones_like(ildj, dtype=np.bool_), np.isfinite(ildj))
 
   def testScalarCongruency(self):
     bijector_test_util.assert_scalar_congruency(

tensorflow_probability/python/bijectors/weibull_cdf_test.py

@@ -59,7 +59,7 @@ def testBijectorConcentration1LogDetJacobianFiniteAtZero(self):
     scale = np.logspace(0.1, 10., num=20).astype(np.float32)
     bijector = tfb.WeibullCDF(scale, concentration=1.)
     fldj = self.evaluate(bijector.forward_log_det_jacobian(0., event_ndims=0))
-    self.assertAllEqual(np.ones_like(fldj, dtype=np.bool), np.isfinite(fldj))
+    self.assertAllEqual(np.ones_like(fldj, dtype=np.bool_), np.isfinite(fldj))
 
   def testScalarCongruency(self):
     bijector_test_util.assert_scalar_congruency(

tensorflow_probability/python/distributions/beta_test.py

@@ -153,7 +153,7 @@ def testPdfXStretchedInBroadcastWhenLowerRank(self):
   def testLogPdfOnBoundaryIsFiniteWhenAlphaIsOne(self):
     b = [[0.01, 0.1, 1., 2], [5., 10., 2., 3]]
     pdf = self.evaluate(tfd.Beta(1., b, validate_args=True).prob(0.))
-    self.assertAllEqual(np.ones_like(pdf, dtype=np.bool), np.isfinite(pdf))
+    self.assertAllEqual(np.ones_like(pdf, dtype=np.bool_), np.isfinite(pdf))
 
   def testBetaMean(self):
     a = [1., 2, 3]
@@ -298,8 +298,8 @@ def testBetaCdf(self, dt, rtol):
     b = 10. * np.random.random(shape).astype(dt)
     x = np.random.random(shape).astype(dt)
     actual = self.evaluate(tfd.Beta(a, b, validate_args=True).cdf(x))
-    self.assertAllEqual(np.ones(shape, dtype=np.bool), 0. <= x)
-    self.assertAllEqual(np.ones(shape, dtype=np.bool), 1. >= x)
+    self.assertAllEqual(np.ones(shape, dtype=np.bool_), 0. <= x)
+    self.assertAllEqual(np.ones(shape, dtype=np.bool_), 1. >= x)
     self.assertAllClose(sp_stats.beta.cdf(x, a, b), actual, rtol=rtol, atol=0)
 
   def testBetaCdfBeyondSupport(self):
@@ -314,8 +314,8 @@ def testBetaLogCdf(self, dt, rtol):
     x = np.random.random(shape).astype(dt)
     actual = self.evaluate(
         tf.exp(tfd.Beta(a, b, validate_args=True).log_cdf(x)))
-    self.assertAllEqual(np.ones(shape, dtype=np.bool), 0. <= x)
-    self.assertAllEqual(np.ones(shape, dtype=np.bool), 1. >= x)
+    self.assertAllEqual(np.ones(shape, dtype=np.bool_), 0. <= x)
+    self.assertAllEqual(np.ones(shape, dtype=np.bool_), 1. >= x)
     self.assertAllClose(sp_stats.beta.cdf(x, a, b), actual, rtol=rtol, atol=0)
 
   def testBetaBetaKL(self):

tensorflow_probability/python/distributions/cauchy_test.py

@@ -40,7 +40,7 @@ def setUp(self):
 
   def assertAllFinite(self, array):
     is_finite = np.isfinite(array)
-    all_true = np.ones_like(is_finite, dtype=np.bool)
+    all_true = np.ones_like(is_finite, dtype=np.bool_)
     self.assertAllEqual(all_true, is_finite)
 
   def _testParamShapes(self, sample_shape, expected):

tensorflow_probability/python/distributions/dirichlet_test.py

@@ -76,13 +76,13 @@ def testLogPdfOnBoundaryIsFiniteWhenAlphaIsOne(self):
     dist = tfd.Dirichlet(concentration, validate_args=True)
     log_prob = self.evaluate(dist.log_prob(x))
     self.assertAllEqual(
-        np.ones_like(log_prob, dtype=np.bool), np.isfinite(log_prob))
+        np.ones_like(log_prob, dtype=np.bool_), np.isfinite(log_prob))
 
     # Test when concentration[k] = 1., and x is zero at various dimensions.
     dist = tfd.Dirichlet(10 * [1.])
     log_prob = self.evaluate(dist.log_prob(x))
     self.assertAllEqual(
-        np.ones_like(log_prob, dtype=np.bool), np.isfinite(log_prob))
+        np.ones_like(log_prob, dtype=np.bool_), np.isfinite(log_prob))
 
   def testPdfOnBoundaryWhenAlphaGreaterThanOne(self):
     dist = tfd.Dirichlet(9 * [2.])

tensorflow_probability/python/distributions/dpp_test.py

@@ -257,7 +257,7 @@ def testDppLogPDF(self, n_points):
       binary = bin(i)[2:]
       subset = [0] * n_points
       subset[-len(binary):] = [int(c) for c in binary]
-      mask = np.array(subset, np.bool)
+      mask = np.array(subset, np.bool_)
       submatrix = true_kernel[mask][:, mask]
       expected = (tf.linalg.logdet(submatrix) -
                   tf.linalg.logdet(true_kernel +

tensorflow_probability/python/distributions/gamma_gamma_test.py

@@ -108,7 +108,7 @@ def testGammaGammaLogPDFAtZero(self):
         validate_args=True)
     log_pdf = self.evaluate(gg.log_prob(0.))
     self.assertAllEqual(
-        np.ones_like(log_pdf, dtype=np.bool), np.isfinite(log_pdf))
+        np.ones_like(log_pdf, dtype=np.bool_), np.isfinite(log_pdf))
 
     gg = tfd.GammaGamma(
         concentration=2.,