Python ML & Data Science
-
Google
- Oakland CA
- http://www.vanderplas.com
- @jakevdp
Block or Report
Block or report jakevdp
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
PythonDataScienceHandbook Public
Python Data Science Handbook: full text in Jupyter Notebooks
-
-
WhirlwindTourOfPython Public
The Jupyter Notebooks behind my OReilly report, "A Whirlwind Tour of Python"
-
1,520 contributions in the last year
Less
More
Contribution activity
November 2022
Created 30 commits in 2 repositories
Created a pull request in google/jax that received 4 comments
jnp.split: use dynamic rather than static slices for speed
Fixes #12999 This is very similar to the issue & fix in #12219 main branch: In [1]: import jax.numpy as jnp ...: x = jnp.ones(5000) ...: %time _ = jnp
+4
−2
•
4
comments
Opened 26 other pull requests in 1 repository
google/jax
25
merged
1
open
- CI: bump timeout for pre-commit
- [sparse] fix bug in sparse.eye under JIT
- x64 JEP: initial draft
- [sparse] avoid re-indexing for linear unary ops
- jaxlib: fix mlir_hlo build rule
- [sparse] refactor with sparse.test_util
- CI: update flake8 pre-commit URL
- [sparse] implement iter() of BCOO
- [sparse] support dense dimensions in bcoo_reshape
- [sparse] streamline sparse rules for standard primitives
- [sparse] improve testing framework
- [sparse] support strides in bcoo_slice
- [sparse] bcoo_dynamic_slice: remove unnecessary padding from output
- [sparse] fix shape bug in bcoo_transpose
- [sparse] fix bcoo_reshape under jit
- [sparse] add bcoo_squeeze function
- [sparse] add support for BCOO.astype method
- [sparse] add bcoo_gather & support for sparse indexing
- [sparse] fix bcoo_reshape when n_sparse=0
- Remove whitespace to fix flake8
- CI: add absl-py to docs/requirements.txt
- [typing] update jaxlib & remove unnecessary ignore
- CI: set explicit permissions for ci-build action
- add jet rule for dynamic_update_slice
- BUG: fix jet rule for dynamic_slice
- Some pull requests not shown.
Reviewed 25 pull requests in 2 repositories
google/jax
24 pull requests
-
Added more accurate version of the
betalnfunction. -
Add
jax.scipy.linalg.toeplitz. - Weaken register_keypaths annotation
- Implementation of scipy.stats bernoulli.cdf & bernoulli.ppf
- Pin IPython version in docs build to avoid RTD warning
- [sparse] avoid re-indexing for linear unary ops
- Update TensorBoard install instructions for profiling
- Update Github Actions workflows using Ratchet
-
Fix less-nice type annotations for e.g.
List[jax.Array] - jax.Array tutorial
- [sparse] improve testing framework
- Send message to internal chat room on Cloud TPU CI failure
-
Update changelog w/ info about deleting
jax_experimental_name_stack - Update jax.numpy.lcm function to match with np.lcm in case of overflow
- Consolidate links in JAX documentation and rearrange on usage
- PRNGKeyArray is now a virtual subclass of ndarray
- [sparse] BCSR batching rule.
- Fix the docs build
- [sparse] add bcoo_gather & support for sparse indexing
- [sparse] Add BCSR from_scipy_sparse.
- Add support for the hermitian option on jnp.linalg.pinv.
- Avoid forming identity matrix in SVD JVP.
- jnp.split: use dynamic rather than static slices for speed
- BUG: fix jet rule for dynamic_slice
jax-ml/jax-triton
1 pull request
Opened 1 issue in 1 repository
google/jax
1
closed
Answered 11 discussions in 1 repository
google/jax
google/jax
-
How to use `dtype` in `sqrt`?
This contribution was made on Nov 27
-
Hessian with `jax.pure_callback` for custom JVP/custom VJP
This contribution was made on Nov 17
-
Transposition not supported for `custom_jvp` with `pure_callback`
This contribution was made on Nov 17
-
vmap + jacfwd error when attempting to compute higher order derivatives of a 1D at an array of values
This contribution was made on Nov 17
-
Using `jax.scipy.sparse.linalg` with a custom operator leads to `jax.interpreters.partial_eval.DynamicJaxprTracer` being passed to the operator instead of `DeviceArray`
This contribution was made on Nov 16
-
jitted-functions slow performance due to copying of arrays
This contribution was made on Nov 14
-
Why jvp imposes primals and tangents to be on the same dtype?
This contribution was made on Nov 13
-
JIT compilation for dynamic results
This contribution was made on Nov 11
-
Jax equivalent of Scipy.constants h, c, k
This contribution was made on Nov 9
-
`custom_vjp` with variable `nondiff_argnums`
This contribution was made on Nov 7
-
Jax's internal notion of a function
This contribution was made on Nov 6





