Still encounter this error: Could not find trained model in model_dir: ./output/sa_nn1.

[ZhaoMingYang-tju]

Hi, i have seen the privious issue about the error ValueError: Could not find trained model in model_dir: ./output/sa_nn1. But it seems that your latest modify doesn't work. I still encounter this error when i use the latest code. Can you check it? here's the traceback :

Traceback (most recent call last): File "E:/learning-drl/magnet-1/magnet/main.py", line 87, in main() File "E:/learning-drl/magnet-1/magnet/main.py", line 49, in main ddpg_agent.train_transformer(sess, env) File "E:\learning-drl\magnet-1\magnet\models\ddpg_agent.py", line 150, in train_transformer next_state, reward, terminal, info = self.env.step(action) File "E:\learning-drl\magnet-1\magnet\env_processing\env_wrapper.py", line 47, in step all_actions = self.gym.act(obs) File "D:\Anaconda3\envs\tensorflow-gpu\lib\site-packages\pommerman\envs\v0.py", line 137, in act return self.model.act(agents, obs, self.action_space) File "D:\Anaconda3\envs\tensorflow-gpu\lib\site-packages\pommerman\forward_model.py", line 122, in act ret.append(act_ex_communication(agent)) File "D:\Anaconda3\envs\tensorflow-gpu\lib\site-packages\pommerman\forward_model.py", line 101, in act_ex_communication return agent.act(obs[agent.agent_id], action_space=action_space) File "E:\learning-drl\magnet-1\magnet\models\ddpg_agent.py", line 99, in act input_to_ddpg = self.input_to_ddpg(prev_state, curr_state) File "E:\learning-drl\magnet-1\magnet\models\ddpg_agent.py", line 84, in input_to_ddpg graph_predictions = np.asmatrix(list(itertools.islice(y_generator, prev_state.shape[0]))[0]['graph']) File "D:\Anaconda3\envs\tensorflow-gpu\lib\site-packages\tensorflow\python\estimator\estimator.py", line 488, in predict self._model_dir)) ValueError: Could not find trained model in model_dir: ./output/sa_nn1.

[yunzelu]

Hi. Is it solved now?

[Amanda2024]

Hi, how can I solve it ?