InferenceServer Error

I am trying to use seel-RL style centralized server for inference. I modified the IMPALA example by adding the `inference_server.InferenceServerConfig`, which is passed to `experiments.make_distributed_experiment`. I keep getting the below error when running the modified code.

### Code
Only the `main` function has been modified to be as below
```python

def main(_):
experiment_config = build_experiment_config()
if RUN_DISTRIBUTED.value:
program = experiments.make_distributed_experiment(
experiment=experiment_config,
num_actors=4,
inference_server_config=inference_server.InferenceServerConfig(batch_size=32,
update_period=1000,
timeout=datetime.timedelta(seconds=100, milliseconds=0, microseconds=0))
)
lp.launch(program, xm_resources=lp_utils.make_xm_docker_resources(program))
else:
experiments.run_experiment(experiment_config)

```

### Error

```bash
batcher = pybind.BuildBatchedHandlerWrapper(func.__name__, handler,
TypeError: BuildBatchedHandlerWrapper(): incompatible function arguments. The following argument types are supported:
1. (arg0: str, arg1: courier::HandlerInterface, arg2: int, arg3: int, arg4: int, arg5: bool) -> courier::HandlerInterface

Invoked with: 'dereference_params_and_call_handler', , 32, 2, 0.0, True
terminate called without an active exception
Fatal Python error: Aborted

Thread 0x00007f36c93a5740 (most recent call first):

```

Could you please help me run the `InferenceServer` based training?

Thanks
Kinal