fixed test

tests/distributions/test_truncated.py

@@ -28,6 +28,7 @@
     Normal,
     TruncatedNormal,
     TruncatedNormalRV,
+    Uniform,
 )
 from pymc.distributions.shape_utils import change_dist_size
 from pymc.distributions.transforms import _default_transform
@@ -404,25 +405,19 @@ def test_truncated_inference():
 
 
 def test_truncated_normal_logcdf_inference():
-    N = 1
-    mu, sigma = 1, 2
-    upper_bounds = np.array([1])
-    y = np.random.normal(loc=mu, scale=sigma, size=N)
-
-    def right_censor(y, upper_bounds):
-        cy = copy(y)
-        censor_index = cy >= upper_bounds
-        cy[censor_index] = upper_bounds[censor_index]
-        return cy
+    lower = 0
+    upper = 1
 
-    cy = right_censor(y, upper_bounds)
+    rng = np.random.default_rng(260)
+    x = rng.normal(0, size=5000)
+    obs = x[np.where(~((x < lower) | (x > upper)))]
 
     with Model() as m:
-        mu = Normal("mu", 0, 1)
-        sigma = Normal("sigma", 0, 1)
+        mu = Normal("mu", 10, 1)
+        sigma = Uniform("sigma", 1, 10)
         normal_ = Normal.dist(mu, sigma)
         trunc_normal = Truncated.dist(normal_, lower=0)
-        censored = Censored("censored", trunc_normal, lower=None, upper=upper_bounds, observed=cy)
+        censored = Censored("censored", trunc_normal, lower=None, upper=upper, observed=obs)
 
     ip = m.initial_point()
     norm_logp, trunc_logp, cens_logp = m.compile_logp(sum=False)(ip)