Upload sb3/hyperpars/ppo-caribou.yml with huggingface_hub
Browse files
sb3/hyperpars/ppo-caribou.yml
CHANGED
|
@@ -17,7 +17,7 @@ algo_config:
|
|
| 17 |
tensorboard_log: "../../../logs"
|
| 18 |
#
|
| 19 |
policy: 'MlpPolicy'
|
| 20 |
-
policy_kwargs: "dict(net_arch=[
|
| 21 |
# policy_kwargs: "dict(net_arch=[256, 128])"
|
| 22 |
# batch_size: 512
|
| 23 |
# gamma: 0.9999
|
|
@@ -40,11 +40,12 @@ env_id: "CaribouScipy"
|
|
| 40 |
#this is what is passed into the creation of the environment, could pass diff param values here to change the defaults (instead of messing with the params defined in caribou_ode.py)
|
| 41 |
config:
|
| 42 |
Tmax: 800
|
|
|
|
| 43 |
n_envs: 12
|
| 44 |
|
| 45 |
tensorboard: "/home/rstudio/logs"
|
| 46 |
#total_timesteps: 50000
|
| 47 |
-
total_timesteps:
|
| 48 |
use_sde: True
|
| 49 |
repo: "boettiger-lab/rl4eco"
|
| 50 |
save_path: "../saved_agents"
|
|
|
|
| 17 |
tensorboard_log: "../../../logs"
|
| 18 |
#
|
| 19 |
policy: 'MlpPolicy'
|
| 20 |
+
policy_kwargs: "dict(net_arch=[64, 32])"
|
| 21 |
# policy_kwargs: "dict(net_arch=[256, 128])"
|
| 22 |
# batch_size: 512
|
| 23 |
# gamma: 0.9999
|
|
|
|
| 40 |
#this is what is passed into the creation of the environment, could pass diff param values here to change the defaults (instead of messing with the params defined in caribou_ode.py)
|
| 41 |
config:
|
| 42 |
Tmax: 800
|
| 43 |
+
budget: 50
|
| 44 |
n_envs: 12
|
| 45 |
|
| 46 |
tensorboard: "/home/rstudio/logs"
|
| 47 |
#total_timesteps: 50000
|
| 48 |
+
total_timesteps: 10000000
|
| 49 |
use_sde: True
|
| 50 |
repo: "boettiger-lab/rl4eco"
|
| 51 |
save_path: "../saved_agents"
|