Code

Devel update#18

Code

Merged

iProzd merged 55 commits intoiProzd:develiProzd/deepmd-kit:develfrom deepmodeling:develdeepmodeling/deepmd-kit:develCopy head branch name to clipboardAug 29, 2021

+4,405-2,554

Commits

Commits on Aug 10, 2021

• clean library and header dependencies (#930)

  Show description for 06d591d
  njzjz

  authored
  06d591d

• fix bug of gelu activation function (#939)

  denghuilu

  authored
  c4856ef

Commits on Aug 11, 2021

• add model compression support for models without training script (#948)

  Show description for e8a9101
  denghuilu

  authored
  e8a9101

Commits on Aug 12, 2021

• convert decay_rate to stop_lr from old inputs (#949)

  Show description for 5529a97
  njzjz

  authored
  5529a97

• add LAMMPS plugin (#945)

  Show description for 72aaa9b
  njzjz

  authored
  72aaa9b

• make native MD compatible with v2.0 (#950)

  Show description for 3c2a2bd
  njzjz

  authored
  3c2a2bd

• Cherry pick #492 and #664: create issue templates (#961)

  Show description for e345105
  

  njzjz

  and

  tuoping

  authored
  e345105

• only enable link what you use on GNU compilers (#962)

  Show description for d02416f
  njzjz

  authored
  d02416f

• Do not find protobuf for python (#963)

  Show description for 56fd343
  njzjz

  authored
  56fd343

• cherry pick #338: fix an error in stress by ase interface (#964)

  Show description for fd397bb
  

  njzjz

  and

  hsulab

  authored
  fd397bb

• cherry pick #564: add more badges (#965)

  Show description for f743bda
  njzjz

  authored
  f743bda

• speedup model compression UTs (#956)

  denghuilu

  authored
  074f3f2

Commits on Aug 13, 2021

• update type embedding doc (#967)

  yhliu918

  authored
  ee0f6e6

Commits on Aug 14, 2021

• add code comment for InputNlist struct (#968)

  Show description for 6c83c27
  denghuilu

  authored
  6c83c27

Commits on Aug 15, 2021

• [ImgBot] Optimize images (#970)

  Show description for 68577be
  

  njzjz

  and

  ImgBotApp

  authored
  68577be

Commits on Aug 16, 2021

• update contributors (#975)

  Show description for 6d8c31c
  

  amcadmus

  and

  Han Wang

  authored
  6d8c31c

• refactor docs (#952)

  Show description for 820b3ed
  

  3 peopleauthored
  820b3ed

• remove bare except and limit the try clause (#977)

  Show description for e16d75c
  njzjz

  authored
  e16d75c

Commits on Aug 17, 2021

• fix python cmake error (#976)

  denghuilu

  authored
  a1914ea

Commits on Aug 18, 2021

• add an example of nopbc and related docs (#994)

  Show description for 795a79b
  njzjz

  authored
  795a79b

Commits on Aug 19, 2021

• add dp --version (#995)

  njzjz

  authored
  4f54de0

• Add the argument tensorboard_freq to control sampling ratio during training. (#996)

  shishaochen

  authored
  cf3e7d9

Commits on Aug 20, 2021

• add sphinx plugins viewcode and intersphinx (#997)

  njzjz

  authored
  f3b08b3

• generate Python API document automatically (#998)

  Show description for 106f570
  njzjz

  authored
  106f570

• unpin tensorflow version (#999)

  Show description for f2a7303
  njzjz

  authored
  f2a7303

Commits on Aug 21, 2021

• fix type hints and add doc for exclude_types (#1005)

  Show description for 475b3d7
  njzjz

  authored
  475b3d7

• add a note for DP-GUI (#1004)

  njzjz

  authored
  346fab6

• fix a bug in setup.py (#1002)

  Show description for 532e84b
  njzjz

  authored
  532e84b

• fix incorrect type hints and docstrings (#1001)

  Show description for 16b81f4
  njzjz

  authored
  16b81f4

• add model compression training support for deepmd-kit (#1000)

  Show description for 472d7c4
  denghuilu

  authored
  472d7c4

Commits on Aug 22, 2021

• add examples to docstings (#1010)

  Show description for e9edec6
  njzjz

  authored
  e9edec6

• rm dp transfer from the model compression workflow (#1008)

  denghuilu

  authored
  1306c32

• bump sphinx_rtd_theme to >=1.0.0rc1 (#1014)

  Show description for 965d91d
  njzjz

  authored
  965d91d

• show citation in the GitHub (#1012)

  Show description for f7a5b08
  njzjz

  authored
  f7a5b08

• move parameters docstring from constructor method to class itself (#1011)

  Show description for d42ca99
  njzjz

  authored
  d42ca99

Commits on Aug 23, 2021

• give a clear message if model.get_ntypes()<data.get_ntypes() (#1016)

  Show description for a1a97fe
  njzjz

  authored
  a1a97fe

• Instantiate RunOptions first when training. (#1019)

  Show description for d46469b
  shishaochen

  authored
  d46469b

• add docstring to DescrptSeA (#1017)

  Show description for 297ce70
  njzjz

  authored
  297ce70

• add InputNlist into API doc (#1009)

  Show description for 29546c1
  

  njzjz

  and

  denghuilu

  authored
  29546c1

Commits on Aug 24, 2021

• Update easy-install.md (#1021)

  caic99

  authored
  93c140f

• update doc for model compression (#1022)

  denghuilu

  authored
  c0874f0

Commits on Aug 25, 2021

• add docstring for EnerFitting (#1024)

  Show description for fd1f371
  

  njzjz

  and

  amcadmus

  authored
  fd1f371

• Use a local Numpy random generator. (#1030)

  Show description for 8048e77
  shishaochen

  authored
  8048e77

• save checkpoint files with step and keep recent files (#1031)

  Show description for 96bc025
  njzjz

  authored
  96bc025

• Avoid the same batch sequence among workers. (#1032)

  shishaochen

  authored
  0004484

• get library extension suffix from built-in method (#1036)

  njzjz

  authored
  f35afe0

• speedup scan_nlist kernel (#1028)

  Show description for 602760e
  denghuilu

  authored
  602760e

Commits on Aug 26, 2021

• use TF's built-in method to get numpy dtype (#1035)

  Show description for ed20998
  njzjz

  authored
  ed20998

• clean up the unused including files of device source files (#1037)

  denghuilu

  authored
  b51f981

• Fix typo: CMAKE_COMPILER_IS_GNUCXX (#1038)

  Show description for 8bbe565
  caic99

  authored
  8bbe565

Commits on Aug 27, 2021

• decouple activation function's type from model compression's process in SE_A, now tanh & gelu is both available. (#1020)

  Show description for 5ab5fa1
  

  liangadam

  and

  HLA

  authored
  5ab5fa1

• fix model compression error (#1043)

  Show description for 37af1d1
  denghuilu

  authored
  37af1d1

Commits on Aug 28, 2021

• change to almostEqual in UT to avoid difference at FP precision (#1048)

  Show description for a7e4760
  

  amcadmus

  and

  Han Wang

  authored
  a7e4760

• fix hybrid descriptor training error (#1052)

  Show description for 5f8989b
  njzjz

  authored
  5f8989b

Commits on Aug 29, 2021

• add compatibility to dp freeze (#1055)

  Show description for 1dd6e97
  njzjz

  authored
  1dd6e97