[BUG] random.poisson doesn't accept arrays of lambda parameters

[syamajala]

Software versions

cunumeric 24.06.00

Jupyter notebook / Jupyter Lab version

No response

Expected behavior

The example below works with numpy and cupy, but seems to crash with cunumeric.

Observed behavior

The code crashes. See below for a stack trace.

Example code or instructions

import cunumeric as np

def generate_gauss_2d(p, xi, yi):
    arg = -(np.square(xi - p[1]) + np.square(yi - p[2])) / (2 * p[3] * p[3])
    y = p[0] * np.exp(arg) + p[4]

    return y


if __name__ == '__main__':

    # number of fits and fit points
    number_fits = 10000
    size_x = 12
    number_points = size_x * size_x
    number_parameters = 5

    # set input arguments

    # true parameters
    true_parameters = np.array((10, 5.5, 5.5, 3, 10), dtype=np.float32)

    # initialize random number generator
    np.random.seed(0)

    # initial parameters (relative randomized, positions relative to width)
    initial_parameters = np.tile(true_parameters, (number_fits, 1))
    initial_parameters[:, (1, 2)] += true_parameters[3] * (-0.2 + 0.4 * np.random.rand(number_fits, 2))
    initial_parameters[:, (0, 3, 4)] *= 0.8 + 0.4 * np.random.rand(number_fits, 3)

    # generate x and y values
    g = np.arange(size_x)
    yi, xi = np.meshgrid(g, g, indexing='ij')
    xi = xi.astype(np.float32)
    yi = yi.astype(np.float32)

    # generate data
    data = generate_gauss_2d(true_parameters, xi, yi)
    data = np.reshape(data, (1, number_points))
    data = np.tile(data, (number_fits, 1))

    # add Poisson noise
    data = np.random.poisson(data)
    data = data.astype(np.float32)

Stack traceback or browser console output

/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/legion_top.py:302: RuntimeWarning: cuNu$
eric has not implemented meshgrid and is falling back to canonical NumPy. You may notice significantly decreased perfo$
mance for this function call.
  exec(code, module.__dict__, module.__dict__)
TypeError: only length-1 arrays can be converted to Python scalars

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/legion_top.py", line 482, in le$
ion_python_main
[0 - 7f8aa5be1000]   28.334343 {6}{python}: python exception occurred within task:
    run_path(args[start], '__main__')
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/legion_top.py", line 302, in ru$
_path
    exec(code, module.__dict__, module.__dict__)
  File "example.py", line 54, in <module>
    data = np.random.poisson(data)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "runtime.pyx", line 633, in legate.core._lib.runtime.runtime.track_provenance.decorator.wrapper
  File "runtime.pyx", line 634, in legate.core._lib.runtime.runtime.track_provenance.decorator.wrapper
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/cunumeric/_utils/coverage.py", $ine 111, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/cunumeric/random/_random.py", l$ne 956, in poisson
    return get_static_generator().poisson(lam, size)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/cunumeric/random/_generator.py"$ line 278, in poisson
    return self.bit_generator.poisson(lam, size)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/cunumeric/random/_bitgenerator.$y", line 187, in poisson
    res._thunk.bitgenerator_poisson(
  File "/sdf/group/lcls/ds/tools/conda_envs/cunumeric-mec/lib/python3.12/site-packages/cunumeric/_thunk/eager.py", lin$ 857, in bitgenerator_poisson
    self.array.fill(np.random.poisson(lam))
ValueError: setting an array element with a sequence.

[manopapad]

I can reproduce the same issue with just

import numpy as np
import cunumeric as cn
data = cn.ones((3,4))
cn.random.poisson(data)

The underlying issue here is that cunumeric.random.poisson (and other similar random distribution APIs) only accepts a single "lambda" parameter, whereas numpy.random.poisson also accepts arrays of parameters.