wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-08-09 12:20:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

[rllib] Fix stats collection and some docs bugs since the refactoring (#2361)

Browse Source 
* fix

* fix pbt example

* fix

* fix

* single thread by default

* vec

* fix

* fix
This commit is contained in:
Eric Liang
2018-07-07 13:29:20 -07:00
committed by GitHub
parent 9a6e329325
commit d24f19fd1e
13 changed files with 60 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/ray/tune/examples/pbt_ppo_example.py
+9 -14
View File
@@ -52,28 +52,23 @@ if __name__ == "__main__":
"env": "Humanoid-v1",
"repeat": 8,
"config": {
"kl_coeff":
1.0,
"num_workers":
8,
"devices": ["/gpu:0"],
"kl_coeff": 1.0,
"num_workers": 8,
"num_gpus": 1,
"model": {
"free_log_std": True
},
# These params are tuned from a fixed starting value.
"lambda":
0.95,
"clip_param":
0.2,
"sgd_stepsize":
1e-4,
"lambda": 0.95,
"clip_param": 0.2,
"sgd_stepsize": 1e-4,
# These params start off randomly drawn from a set.
"num_sgd_iter":
lambda spec: random.choice([10, 20, 30]),
lambda spec: random.choice([10, 20, 30]),
"sgd_batchsize":
lambda spec: random.choice([128, 512, 2048]),
lambda spec: random.choice([128, 512, 2048]),
"timesteps_per_batch":
lambda spec: random.choice([10000, 20000, 40000])
lambda spec: random.choice([10000, 20000, 40000])
},
},
},