tensorforce.exception.TensorforceError: Invalid value for variable argument spec: TensorSpec

[waynezw0618]

Hello
I am using tensorforce 0.6.0 in DRLinFluids package. I can run tutorial case in DRLinFluild on the cluster. I slightly change the openfoam case part in DRLinfluid from a cylinder of two jet variable to one. then the initialization error arise. as :

WARNING:root:No min_value bound specified for state. OpenFOAM_init running time： 45.16 s OpenFOAM_init running time： 39.83 s OpenFOAM_init running time： 40.02 s OpenFOAM_init running time： 40.01 s OpenFOAM_init running time： 39.02 s Traceback (most recent call last): File "DRLinFluids_cavity/launch_multiprocessing_traning_cavity.py", line 115, in <module> agent = Agent.create( File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/agents/agent.py", line 131, in create return Agent.create(agent=agent, environment=environment, **kwargs) File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/agents/agent.py", line 104, in create return Agent.create(agent=agent, environment=environment) File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/agents/agent.py", line 71, in create agent.initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/agents/agent.py", line 263, in initialize self.model.initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/models/tensorforce.py", line 503, in initialize super().initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/models/model.py", line 264, in initialize super().initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/module.py", line 270, in initialize module.initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/module.py", line 270, in initialize module.initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/module.py", line 270, in initialize module.initialize() File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/layers/dense.py", line 84, in initia lize self.weights = self.variable( File "/work/home/acsm7tkef4/miniconda3/envs/py38/lib/python3.8/site-packages/tensorforce/core/module.py", line 510, in variable raise TensorforceError.value( tensorforce.exception.TensorforceError: Invalid value for variable argument spec: TensorSpec(type=float, shape=(0, 1024)) underspeci fied.

I also made comparison for the original setup and the one I modified.

'''

base) [acsm7tkef4@login06 acousticCavity]$ diff DRLinFluids_cavity/launch_multiprocessing_traning_cavity.py ../square2D_2112/DRLinFluids_square/launch_multiprocessing_traning_square.py
5c5
< import envobject_cavity

import envobject_square
26c26
< 'num_processor': 16,

---

'num_processor': 4,

28c28
< 'cfd_init_time': 0.05,

'cfd_init_time': 0.005,

35c35
< 'actuator': {

    'JET4': {

46c46
< 'actuator': {

    'JET4': {

57c57
< 'actuator': {

    'JET4': {

72c72
< 'interaction_period': 0.3,

'interaction_period': 0.025,

74c74
< 'writeInterval': 0.3,

'writeInterval': 0.025,

91c91
< env = envobject_cavity.cavityNoise2D(

env = envobject_square.FlowAroundSquare2D(

'''
can you please let me know what I can do ?

Thanks
Wei

[moi567]

Have you solved this problem?