manuel/ai-econ
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

entropy you silly nugget

Browse Source
This commit is contained in:
manuel
2023-01-17 16:24:45 +01:00
parent 731aad0a7b
commit bfd6d1e37b
6 changed files with 443 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main.py
+3 -3
View File
@@ -157,7 +157,7 @@ class TensorboardCallback(BaseCallback):
else:
self.metrics=self.econ.previous_episode_metrics
curr_prod=self.metrics["social/productivity"]
trend_pord=curr_prod-prev_metrics["social/productivity"]
trend_pord=curr_prod/np.maximum(0.0001,prev_metrics["social/productivity"])
self.logger.record("social/total_productivity", curr_prod)
self.logger.record("social/delta_productivity", trend_pord)
@@ -292,8 +292,8 @@ runname="run_{}".format(run_number)
model_db=[None,None] # object for storing model
model = MaskablePPO("MlpPolicy",n_steps=int(env_config['episode_length']*2),ent_coef=0.1, vf_coef=0.5 ,gamma=0.99, learning_rate=learning_rate_adj_basic,env=stackenv_basic, seed=445,verbose=1,device="cuda",tensorboard_log="./log")
model_trade=MaskablePPO("MlpPolicy",n_steps=int(env_config['episode_length']*2),ent_coef=0.1, vf_coef=0.5 ,gamma=0.99, learning_rate=learning_rate_adj_trade,env=stackenv_traid, seed=445,verbose=1,device="cuda",tensorboard_log="./log")
model = MaskablePPO("MlpPolicy",n_steps=int(env_config['episode_length']*2),ent_coef=0.001, vf_coef=0.5 ,gamma=0.99, learning_rate=learning_rate_adj_basic,env=stackenv_basic, seed=445,verbose=1,device="cuda",tensorboard_log="./log")
model_trade=MaskablePPO("MlpPolicy",n_steps=int(env_config['episode_length']*2),ent_coef=0.001, vf_coef=0.5 ,gamma=0.99, learning_rate=learning_rate_adj_trade,env=stackenv_traid, seed=445,verbose=1,device="cuda",tensorboard_log="./log")
n_agents=econ.n_agents