Skip to content

support the custom training times for each frame#2208

Merged
wanghan-iapcm merged 7 commits into
deepmodeling:develfrom
njzjz:load_data_with_weight
Jan 2, 2023
Merged

support the custom training times for each frame#2208
wanghan-iapcm merged 7 commits into
deepmodeling:develfrom
njzjz:load_data_with_weight

Conversation

@njzjz

@njzjz njzjz commented Dec 30, 2022

Copy link
Copy Markdown
Member

Before, sys_probs only specifies the probability of each system. This patch allows numb_copy.npy (int) to specify each frame is copied by the "numb_copy" (int) times. The default is 1 for all frames, which keeps the same behavior as before.

Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
@codecov-commenter

codecov-commenter commented Dec 30, 2022

Copy link
Copy Markdown

Codecov Report

Base: 74.86% // Head: 74.92% // Increases project coverage by +0.06% 🎉

Coverage data is based on head (ac35c58) compared to base (41180ca).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files
@@            Coverage Diff             @@
##            devel    #2208      +/-   ##
==========================================
+ Coverage   74.86%   74.92%   +0.06%     
==========================================
  Files         203      203              
  Lines       20096    20104       +8     
  Branches     1436     1438       +2     
==========================================
+ Hits        15044    15063      +19     
+ Misses       4146     4134      -12     
- Partials      906      907       +1     
Impacted Files Coverage Δ
deepmd/common.py 78.18% <ø> (ø)
deepmd/utils/data.py 93.64% <100.00%> (+0.40%) ⬆️
source/api_cc/src/DeepPot.cc 66.35% <0.00%> (-0.76%) ⬇️
source/lib/include/ComputeDescriptor.h 76.65% <0.00%> (-0.23%) ⬇️
source/api_c/include/deepmd.hpp 93.75% <0.00%> (+0.11%) ⬆️
source/api_c/src/c_api.cc 81.22% <0.00%> (+0.23%) ⬆️
source/api_cc/src/DataModifier.cc 80.68% <0.00%> (+0.27%) ⬆️
source/api_cc/src/common.cc 77.62% <0.00%> (+0.81%) ⬆️
source/api_cc/src/DeepTensor.cc 84.29% <0.00%> (+3.81%) ⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@wanghan-iapcm wanghan-iapcm left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I suggest not naming it as "probability", because a probability should be summed to 1.

"numb_copy" may be a better name, i.e. each frame is copied by the "numb_copy" (int) times. This is the behavior of your PR.

@njzjz njzjz changed the title support the custom probability for frames support the custom training times for each frame Dec 31, 2022
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
@wanghan-iapcm wanghan-iapcm merged commit 7a97c95 into deepmodeling:devel Jan 2, 2023
mingzhong15 pushed a commit to mingzhong15/deepmd-kit that referenced this pull request Jan 15, 2023
Before, `sys_probs` only specifies the probability of each system. This
patch allows `numb_copy.npy` (`int`) to specify each frame is copied by
the "numb_copy" (int) times. The default is 1 for all frames, which
keeps the same behavior as before.

Signed-off-by: Jinzhe Zeng <jinzhe.zeng@rutgers.edu>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants