This repository was archived by the owner on Nov 17, 2023. It is now read-only.

[v1.7.x] Backport some numpy features + fixes#18648

Closed

sxjscience wants to merge 44 commits intoapache:v1.7.xfrom

sxjscience:backport_pr_v1.7

Commits on Jun 30, 2020

add zero grad for npi_unique (#18080)

haojin2
authored and
sxjscience
committed

Commits on Jul 1, 2020

flatnonzero (#17690)

Yiyan66
authored and
sxjscience
committed
[Numpy] FFI for cumsum and add (#17747)

haojin2
authored and
sxjscience
committed
[Numpy] FFI: Bincount, Percentile/Quantile, All/Any (#17717)

Tommliu
authored and
sxjscience
committed
Add ffi benchmark (#17780)

authored and
sxjscience
committed
ffi wrappers for polyval, ediff1d, nan_to_num (#17832)

Alicia1529
authored and
sxjscience
committed
Fix compiler warnings in new FFI (#17718)

hzfan
authored and
sxjscience
committed
ffi invocation: expand_dims, tril, diff, broadcast_to (#17738)

Alicia1529
authored and
sxjscience
committed
[Numpy] FFI for split and svd (#17816)

authored and
sxjscience
committed
add ffi for full_like, binary (#17811)

Alicia1529
authored and
sxjscience
committed
* impl - FFI for np_where_op (#17817)

authored and
sxjscience
committed
ffi for roll/rot90 (#17861)

Tommliu
authored and
sxjscience
committed
[Numpy] allow mix integer dtypes for power/add/multiply (#17921)

JiangZhaoh
authored and
sxjscience
committed
fix true_divide (#18393)
sxjscience
committed
* FFI for np.argmax and np.argmin (#17843)

authored and
sxjscience
committed
add alias for np.__version__, np._NoValue and np.dtype (#17777)

haojin2
authored and
sxjscience
committed
[Numpy] FFI for diag/diagonal/diag_indices_from (#17789)

Tommliu
authored and
sxjscience
committed
ffi_atleast_1/2/3d (#17897)

Tommliu
authored and
sxjscience
committed
add: numpy rollaxis (#17865)

yijunc
authored and
sxjscience
committed
* impl - FFI for np einsum (#17869)

authored and
sxjscience
committed
[numpy] add op random.f (#17586)

Yiyan66
authored and
sxjscience
committed
[NumPy] add op random.laplace (#17316)

authored and
sxjscience
committed
* impl - FFi for linalg op (#17795)

authored and
sxjscience
committed
[Numpy] FFI: split and svd #17816

hzfan
authored and
sxjscience
committed
ffi random (#18051)

Yiyan66
authored and
sxjscience
committed
* impl - linalg.lstsq for cpu (#17950)

authored and
sxjscience
committed
Add np.linalg.qr (#17851)

D-Roberts
authored and
sxjscience
committed
[Numpy] FFI for linalg.qr and linalg.lstsq (#18040)

authored and
sxjscience
committed
Add np.linalg.qr backward (#18050)

D-Roberts
authored and
sxjscience
committed
ffi_array_split, v/h/dsplit (#17873)

Tommliu
authored and
sxjscience
committed
[numpy] FFI for insert \ delete \ matmul etc. (#17759)

JiangZhaoh
authored and
sxjscience
committed
* impl debug - FFI for linalg multioutput op (#17879)

authored and
sxjscience
committed
* impl - linalg matrix_rank for cpu and gpu implemented (#18020)

authored and
sxjscience
committed
[Numpy] Kron operator (#17323)

hanke580
authored and
sxjscience
committed
[Numpy] OP_interp (#17793)

Tommliu
authored and
sxjscience
committed
[numpy] add op median #17084

Yiyan66
authored and
sxjscience
committed
[Numpy] Add op fmax, fmin, fmod (#17567)

hanke580
authored and
sxjscience
committed
add: numpy op tril_indices (#17904)

yijunc
authored and
sxjscience
committed
[NumPy] Add NumPy support for triu (#17614)

authored and
sxjscience
committed
fix mixed type backward (#18250)

yijunc
authored and
sxjscience
committed
[Numpy] Add ffi for np.sum, np.std, np.var, np.average and np.histogram (#17866)

haojin2
authored and
sxjscience
committed
[numpy] FFI binary bitwise ops (#17812)

Yiyan66
authored and
sxjscience
committed
[Numpy] FFI: random.choice, take and clip (#17854)

authored and
sxjscience
committed
fix np.clip scalar input case (#17788)

xidulu
authored and
sxjscience
committed