fix(luyd): fix new pipeline impala in Lunarlander and Atari env

[AltmanD]

Fix new pipeline impala in lunarlander env.

[codecov[bot]]

Codecov Report

Merging #713 (a6afe66) into main (3059479) will decrease coverage by 0.15%. Report is 10 commits behind head on main. The diff coverage is 77.27%.

:exclamation: Current head a6afe66 differs from pull request most recent head 0433473. Consider uploading reports for the commit 0433473 to get more accurate results

@@            Coverage Diff             @@
##             main     #713      +/-   ##
==========================================
- Coverage   80.68%   80.53%   -0.15%     
==========================================
  Files         598      598              
  Lines       50116    50031      -85     
==========================================
- Hits        40434    40294     -140     
- Misses       9682     9737      +55     

Flag	Coverage Δ
unittests	80.53% <77.27%> (-0.15%)	:arrow_down:

Flags with carried forward coverage won't be shown. Click here to find out more.

Files Changed	Coverage Δ
ding/data/buffer/deque_buffer.py	97.02% <66.66%> (-0.96%)	:arrow_down:
ding/policy/impala.py	86.39% <78.57%> (ø)
ding/config/config.py	84.58% <100.00%> (+0.10%)	:arrow_up:

... and 69 files with indirect coverage changes

[AltmanD]

Reward curve in Lunarlander, Pong, Spaceinvaders

Lunarlander:

Pong:

Spaceinvaders: