Suriyaa MM

If i run it in Visual Studio, IDE process gets killed. If i run it in terminal, I am getting `killed` in return.

``` $ python classification.py Killed ``` ``` $ cat classification.py import torch import logging from torch import nn from torch.utils.data import DataLoader, random_split from torchvision.datasets import PCAM from torchvision import...

```python import jax.numpy as jnp import jax jax.config.update("jax_platform_name", "cpu") print(jax.devices()) print(jnp.linalg.svd(jnp.full((1, 3, 3), jnp.nan))) print(jnp.linalg.svd(jnp.full((2, 3, 3), jnp.nan))) ``` Output ``` [CpuDevice(id=0)] SVDResult(U=Array([[[nan, nan, nan], [nan, nan, nan], [nan, nan,...

```python import scipy.stats as stats import jax.scipy.stats as jstats import numpy as np print("scipy") print(stats.gumbel_l.logpdf(np.arange(10.0).astype(np.float32), loc=0, scale=-1)) print("jax") print(jstats.gumbel_l.logpdf(np.arange(10.0), loc=0, scale=-1)) print("scipy") print(stats.gumbel_l.logcdf(np.arange(10.0), loc=0, scale=-1)) print("jax") print(jstats.gumbel_l.logcdf(np.arange(10.0), loc=0, scale=-1)) print("scipy")...

@jakevdp can we proceed further or is there anything that you would like to change?

I’ve updated the implementation as per your comments. Please take another look when convenient, happy to iterate if needed.

I have added it based on previously implemented distributions. Not sure whether it is right or not. If there is a way to check this, then please let me know.

Hi! Just checking in, is there anything else I should do on this PR, or is it good to go?

I hope everything is fine. Let me know if there is anything else that I need to change.

Tests for Oldest Supported Numpy and CPU tests when x64 = 1 failed. Should I promote the types before returning from args_maker()? like this? ```python def args_maker(): x, loc, scale...