diff --git a/algos_tf14/a2c_discrete.py b/algos_tf14/a2c_discrete.py index 4b6beb08..a2433860 100644 --- a/algos_tf14/a2c_discrete.py +++ b/algos_tf14/a2c_discrete.py @@ -349,7 +349,7 @@ def train(self): last_mean_rewards = -100500 play_time = 0 epoch_num = 0 - max_epochs = self.config.get('max_epochs', 1e6) + max_epochs = self.config.get('max_epochs', 1e4) start_time = time.time() total_time = 0 @@ -357,6 +357,7 @@ def train(self): while True: play_time_start = time.time() epoch_num = self.update_epoch() + print(epoch_num) frame += batch_size_envs obses, returns, dones, actions, values, neglogpacs, lstm_states, _ = self.play_steps() advantages = returns - values diff --git a/launch_servers.sh b/launch_servers.sh index 7f54cc1e..277930f6 100755 --- a/launch_servers.sh +++ b/launch_servers.sh @@ -19,3 +19,13 @@ for i in {1..3}; do # ./run_servers.sh ${i} python3 tf14_runner.py --file=whirl_baselines/vdn_MMM2 with name=vdn_MMM2_tf label=vdn_MMM2_tf & # ./run_servers.sh ${i} python3 tf14_runner.py --file=whirl_baselines/3s_vs_5z with name=3s_vs_5z_tf label=3s_vs_5z_tf & done +for i in {1..15}; do ./run_servers.sh 0 python3 tf14_runner.py --file=whirl_baselines/3s5z_vs_3s6z with name=3s5z_vs_3s6z_tf_v1_4.10_r4 label=3s5z_vs_3s6z_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 1 python3 tf14_runner.py --file=whirl_baselines/6h_vs_8z with name=6h_vs_8z_tf_v1_4.10_r4 label=6h_vs_8z_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 2 python3 tf14_runner.py --file=whirl_baselines/corridor with name=corridor_tf_v1_4.10_r4 label=corridor_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 3 python3 tf14_runner.py --file=whirl_baselines/MMM2 with name=MMM2_tf_v1_4.10_r4 label=MMM2_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 4 python3 tf14_runner.py --file=whirl_baselines/27m_vs_30m with name=27m_vs_30m_tf_v1_4.10_r4 label=27m_vs_30m_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 5 python3 tf14_runner.py --file=whirl_baselines/3s_vs_5z with name=3s_vs_5z_tf_v1_4.10_r4 label=3s_vs_5z_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 6 python3 tf14_runner.py --file=whirl_baselines/5m_vs_6m with name=5m_vs_6m_tf_v1_4.10_r4 label=5m_vs_6m_tf_v1_4.10_r4 ; done +for i in {1..15}; do ./run_servers.sh 7 python3 tf14_runner.py --file=whirl_baselines/6h_vs_8z with name=6h_vs_8z_tf_v1_4.10_r4 label=6h_vs_8z_tf_v1_4.10_r4 ; done + +