Closed Sharad24 closed 4 years ago
This pull request introduces 5 alerts when merging c55e8341f9b77e4709f4758cf30372c6ca60b710 into dc2d58bdade1c5239ccc0925aeda78d1dc59e972 - view on LGTM.com
new alerts:
I reckon it might be better to shift all env wrappers to torch now. And keeping only .numpy()
for base env (i.e., just before a function which requires a numpy array e.g. step())
Unless there's a limitation to torch not working with a particular wrapper that is. @sampreet-arthi Any comments?
I reckon it might be better to shift all env wrappers to torch now. And keeping only .numpy() for base env (i.e., just before a function which requires a numpy array e.g. step())
Would be a good idea imo. This might help increase speed a lot. I don't think there's any environment which will not work with any env.
This pull request introduces 22 alerts when merging 8b92fb82b72e9c9f27dded95355432faf952b4b3 into dc2d58bdade1c5239ccc0925aeda78d1dc59e972 - view on LGTM.com
new alerts:
Will merge this today. Most likely done here, will take a pass once again though.
This pull request introduces 24 alerts when merging ab09539e2f674576d2a2900febe76f9d4b2c0b71 into e93cdcc2527b94f3815c25aa0fe9e11a0903822b - view on LGTM.com
new alerts:
This pull request introduces 5 alerts when merging d7730f87be880f9cd0921edc8e188b72c78e72de into c20464fc065cf15b7da451f15cfb157e984dcd09 - view on LGTM.com
new alerts:
Merging #294 into master will increase coverage by
1.24%
. The diff coverage is97.59%
.
@@ Coverage Diff @@
## master #294 +/- ##
==========================================
+ Coverage 89.63% 90.87% +1.24%
==========================================
Files 87 88 +1
Lines 3752 3705 -47
==========================================
+ Hits 3363 3367 +4
+ Misses 389 338 -51
Impacted Files | Coverage Δ | |
---|---|---|
genrl/agents/deep/td3/td3.py | 92.72% <ø> (-0.13%) |
:arrow_down: |
genrl/agents/deep/vpg/vpg.py | 94.33% <ø> (-0.21%) |
:arrow_down: |
genrl/trainers/onpolicy.py | 92.00% <ø> (ø) |
|
genrl/utils/utils.py | 94.25% <77.77%> (-2.09%) |
:arrow_down: |
genrl/core/rollout_storage.py | 91.53% <94.44%> (+0.06%) |
:arrow_up: |
genrl/environments/torch.py | 95.00% <95.00%> (ø) |
|
...nrl/agents/bandits/contextual/common/base_model.py | 100.00% <100.00%> (ø) |
|
genrl/agents/bandits/contextual/common/bayesian.py | 92.30% <100.00%> (ø) |
|
genrl/agents/bandits/contextual/common/neural.py | 95.23% <100.00%> (ø) |
|
genrl/agents/deep/a2c/a2c.py | 93.24% <100.00%> (-0.18%) |
:arrow_down: |
... and 23 more |
Checked all agents. Training properly. Everything working smoothly otherwise.
This pull request introduces 15 alerts when merging 8bae9a8dd4b3518cecc629a24ccddd59cd770f02 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
This pull request introduces 14 alerts when merging 1b9af91b65c3b3c384ad04a4cfda4f151569ef91 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
This pull request introduces 14 alerts when merging 3735077f81ee9eca02b21d6a0b8e99accb044118 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
This pull request introduces 13 alerts when merging 4bdbf99d5d8efedb1fca5b813bfd1bfdc16d7901 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
This pull request introduces 13 alerts when merging 9e7cfe879e82a42fb7a8ba69d51130b9638b61e1 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
Just remembered that the reason we stuck with isort 4.3.20 was because we could avoid the LGTM alerts :P
Yeah this is getting really annoying. LGTM config file does not seem to do the job, # lgtm[query-id]
also is not working.
noqa
is the last attempt, otherwise switching back to 4.3.2
This pull request introduces 12 alerts when merging df410a8a51b3f9afef1440f63bc3ceda42384b55 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
This pull request introduces 4 alerts when merging 96d79bd2b953b047078c92c0d28bdf3242b76e49 into 52860956b6970620620ee37c9c94f4d65abb1d84 - view on LGTM.com
new alerts:
Need to update
ReplayBuffer
, documentation and typing.