issue whille executing the example notebook

[e-dinesh]

After executing the third cell, the following error pops-up saying "TypeError: run() takes 5 positional arguments but 8 were given".
Any leads please ?

---

Runner defined DONE!

Episodes: 0%| | 0/1000 [00:00, reward=0.00, ts/ep=0, sec/ep=0.00, ms/ts=0.0, agent=0.0%]

/home/dinesh/Documents/DRLinFluids/examples/active_flow_control/env01/0.005/U
/home/dinesh/Documents/DRLinFluids/examples/active_flow_control/env01/0.005/U
/home/dinesh/Documents/DRLinFluids/examples/active_flow_control/env01/0.005/U

---

TypeError Traceback (most recent call last)
Cell In [3], line 11
8 print('Runner defined DONE!')
10 # runner.run(episodes=500, max_episode_timesteps=80)
---> 11 runner.run(
12 num_episodes=num_episodes,
13 save_best_agent ='best_model'
14 )
15 runner.close()
17 for environment in environments:

File ~/local/anaconda3/envs/drlinfluids/lib/python3.8/site-packages/tensorforce/execution/runner.py:516, in Runner.run(self, num_episodes, num_timesteps, num_updates, batch_agent_calls, sync_timesteps, sync_episodes, num_sleep_secs, callback, callback_episode_frequency, callback_timestep_frequency, use_tqdm, mean_horizon, evaluation, save_best_agent, evaluation_callback)
512 break
514 else:
515 # Check whether environment is ready, otherwise continue
--> 516 observation = self.environments[n].receive_execute()
517 if observation is None:
518 self.terminals[n] = self.prev_terminals[n]

File ~/local/anaconda3/envs/drlinfluids/lib/python3.8/site-packages/tensorforce/environments/environment.py:328, in Environment.receive_execute(self)
326 self._expect_receive = None
327 assert self._actions is not None
--> 328 states, terminal, reward = self.execute(actions=self._actions)
329 self._actions = None
330 return states, int(terminal), reward

File ~/local/anaconda3/envs/drlinfluids/lib/python3.8/site-packages/tensorforce/environments/environment.py:380, in EnvironmentWrapper.execute(self, actions)
376 raise TensorforceError(
377 message="An environment episode has to be initialized by calling reset() first."
378 )
379 assert self._max_episode_timesteps is None or self._timestep < self._max_episode_timesteps
--> 380 states, terminal, reward = self._environment.execute(actions=actions)
381 if isinstance(states, dict):
382 states = states.copy()

Cell In [1], line 99, in FlowAroundCylinder2D.execute(self, actions)
96 self.foam_params['verbose'] = True
98 simulation_start_time = time()
---> 99 drlinfluids.runner.run(
100 self.foam_root_path,
101 self.foam_params,self.agent_params['interaction_period'], self.agent_params['purgeWrite_numbers'],self.agent_params['writeInterval'],
102 self.agent_params['deltaT'],
103 start_time_float, end_time_float
104 )
105 simulation_end_time = time()
107 self.probe_velocity = utils.read_foam_file(
108 self.foam_root_path + f'/postProcessing/probes/{self.start_time_filename}/U',
109 dimension=self.foam_params['num_dimension']
110 )

TypeError: run() takes 5 positional arguments but 8 were given

[venturi123]

Hi @e-dinesh, we will do a quick test about your issue, and give you feedback soon.

[1900360]

Hi @e-dinesh !
Which Tensorforce version did you use? I suggest Tensorforce==0.6.0:)

[jerabaul29]

Maybe this is a good indication that providing a conda environment file would be useful? See creating an environment.yml file at https://carpentries-incubator.github.io/introduction-to-conda-for-data-scientists/04-sharing-environments/index.html . Otherwise, maybe it is better to use the container (docker or singularity) to run the notebook, to make sure all versions are exactly the same as expected? :)