From 81e6e1034485685bb117fa10731a085a2bd83260 Mon Sep 17 00:00:00 2001
From: Erik Nygren <erik.nygren@sbb.ch>
Date: Sun, 1 Sep 2019 20:58:52 -0400
Subject: [PATCH] fixed reward bug
---
torch_training/render_agent_behavior.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/torch_training/render_agent_behavior.py b/torch_training/render_agent_behavior.py
index aabd457..589d610 100644
--- a/torch_training/render_agent_behavior.py
+++ b/torch_training/render_agent_behavior.py
@@ -48,8 +48,8 @@ stochastic_data = {'prop_malfunction': 0.0, # Percentage of defective agents
TreeObservation = TreeObsForRailEnv(max_depth=2)
# Different agent types (trains) with different speeds.
-speed_ration_map = {1.: 1., # Fast passenger train
- 1. / 2.: 0.0, # Fast freight train
+speed_ration_map = {1.: 0., # Fast passenger train
+ 1. / 2.: 1.0, # Fast freight train
1. / 3.: 0.0, # Slow commuter train
1. / 4.: 0.0} # Slow freight train
@@ -102,7 +102,7 @@ action_prob = [0] * action_size
agent_obs = [None] * env.get_num_agents()
agent_next_obs = [None] * env.get_num_agents()
agent = Agent(state_size, action_size, "FC", 0)
-with path(torch_training.Nets, "navigator_checkpoint9600.pth") as file_in:
+with path(torch_training.Nets, "navigator_checkpoint10700.pth") as file_in:
agent.qnetwork_local.load_state_dict(torch.load(file_in))
record_images = False
--
GitLab