jakevdp
Follow
jakevdp
Follow
Achievements
Achievements
Highlights
- 133 discussions answered
Pinned
1,497 contributions in the last year
Less
More
Contribution activity
January 2022
Created 16 commits in 3 repositories
Created 1 repository
Created a pull request in google/jax that received 10 comments
Add lax.searchsorted_p primitive
Make searchsorted a lax primitive, primarily to allow switching to a more efficient implementation on accelerators.
Also addresses one of the TODOs…
+308
−27
•
10
comments
Opened 15 other pull requests in 1 repository
google/jax
3
open
11
merged
1
closed
- jnp.[nan]argmin/max: implement keepdims
- jax.numpy: add where and initial arguments to nan reductions
- jnp.angle: support deg keyword
- Make array_copy a primitive
- [sparse] add bcoo_sort_indices
- [sparse]: change call signature of coo primitives
- einsum: clarify use of precision.
- jnp.unique: properly handle NaN values
- lax.sort: regularize handling of -0.0 and -NaN
- [sparse] Generalize BCOO metadata
- searchsorted: use correct ordering for complex inputs
- jnp.searchsorted: properly handle NaNs
- [sparse] add getitem stub in JAXSparse
- Update sphinx-related packages
- lax.dynamic_update_slice: fix batching rule
Reviewed 17 pull requests in 2 repositories
google/jax
15 pull requests
- Add more documentation for buffer donation
- jnp.[nan]argmin/max: implement keepdims
-
Fix
barrier_waitexample - Add lax.searchsorted_p primitive
- Correct lax.while_loop type annotation
- [sparse] add bcoo_sort_indices
- Add ensure_compile_time_eval docstring to docs
- Implemented mode function
- lax.sort: regularize handling of -0.0 and -NaN
- Implement SciPy's RegularGridInterpolator
- Fix DeviceArray class reference
- Updated broken links and cell outputs in thinking_in_jax.ipynb to reflect the current JAX behavior
- Fix a type issue in odeint
- Implemented mode function
- Fix test failures under Numpy 1.22.
altair-viz/altair_viewer
2 pull requests
Created an issue in google/jax that received 2 comments
jnp.unique does not properly handle NaN values
Example: import numpy as np import jax.numpy as jnp x = np.array([0, 1, np.nan, np.nan]) print(np.unique(x)) # [ 0., 1., nan]) print(jnp.unique(x)) [
2
comments