diff --git a/benchmark/log/default/ppo/Mar21-203336/events.out.tfevents.1711024419.cncl4576.1281258.0 b/benchmark/log/default/ppo/Mar21-203336/events.out.tfevents.1711024419.cncl4576.1281258.0 deleted file mode 100644 index 6816c8a..0000000 Binary files a/benchmark/log/default/ppo/Mar21-203336/events.out.tfevents.1711024419.cncl4576.1281258.0 and /dev/null differ diff --git a/benchmark/log/default/ppo/Mar21-203336/policy.pth b/benchmark/log/default/ppo/Mar21-203336/policy.pth deleted file mode 100644 index 3c3be1a..0000000 Binary files a/benchmark/log/default/ppo/Mar21-203336/policy.pth and /dev/null differ diff --git a/diffusion/__pycache__/model.cpython-38.pyc b/diffusion/__pycache__/model.cpython-38.pyc index 3e12070..295bf34 100644 Binary files a/diffusion/__pycache__/model.cpython-38.pyc and b/diffusion/__pycache__/model.cpython-38.pyc differ diff --git a/diffusion/model.py b/diffusion/model.py index a8e316e..ee167f5 100644 --- a/diffusion/model.py +++ b/diffusion/model.py @@ -39,8 +39,7 @@ def forward(self, x, time, state): t = self.time_mlp(time) x = torch.cat([x, t, processed_state], dim=1) x = self.mid_layer(x) - # x = self.final_layer(x) - # return torch.tanh(x) + x = self.final_layer(x) return x diff --git a/env/__pycache__/env.cpython-38.pyc b/env/__pycache__/env.cpython-38.pyc index 3c9d92d..b6f4fa1 100644 Binary files a/env/__pycache__/env.cpython-38.pyc and b/env/__pycache__/env.cpython-38.pyc differ diff --git a/env/__pycache__/utility.cpython-38.pyc b/env/__pycache__/utility.cpython-38.pyc index 678c194..98c1a4e 100644 Binary files a/env/__pycache__/utility.cpython-38.pyc and b/env/__pycache__/utility.cpython-38.pyc differ diff --git a/env/env.py b/env/env.py index c1151e4..ab9a953 100644 --- a/env/env.py +++ b/env/env.py @@ -38,8 +38,8 @@ def action_space(self): @property def state(self): # Provide the current state to the agent - states1 = np.random.uniform(13, 14, 50) - states2 = np.random.uniform(0, 0.1, 50) + states1 = np.random.uniform(13, 14, 5) + states2 = np.random.uniform(0, 0.1, 5) states = np.concatenate([states1, states2]) self._laststate = states return states diff --git a/env/utility.py b/env/utility.py index 3645d7c..1e07a23 100644 --- a/env/utility.py +++ b/env/utility.py @@ -56,8 +56,8 @@ def water(s, total_power): # Function to compute utility (reward) for the given state and action def CompUtility(State, Aution): actions = torch.from_numpy(Aution).float() - actions = torch.abs(actions) - # actions = torch.sigmoid(actions) + # actions = torch.abs(actions) + actions = torch.sigmoid(1.5*actions) Aution = actions.numpy() total_power = 12 diff --git a/log/default/diffusion/Mar21-203131/events.out.tfevents.1711024293.cncl4576.1276206.0 b/log/default/diffusion/Mar21-203131/events.out.tfevents.1711024293.cncl4576.1276206.0 deleted file mode 100644 index 11ad077..0000000 Binary files a/log/default/diffusion/Mar21-203131/events.out.tfevents.1711024293.cncl4576.1276206.0 and /dev/null differ diff --git a/log/default/diffusion/Mar21-203131/policy.pth b/log/default/diffusion/Mar21-203131/policy.pth deleted file mode 100644 index d41f401..0000000 Binary files a/log/default/diffusion/Mar21-203131/policy.pth and /dev/null differ diff --git a/log/default/diffusion/Mar21-205055/events.out.tfevents.1711025458.cncl4576.1321880.0 b/log/default/diffusion/Mar21-205055/events.out.tfevents.1711025458.cncl4576.1321880.0 new file mode 100644 index 0000000..c95bf32 Binary files /dev/null and b/log/default/diffusion/Mar21-205055/events.out.tfevents.1711025458.cncl4576.1321880.0 differ