inference_params problem & mamba2 error: TypeError: causal_conv1d_fwd(): incompatible function arguments.

[JianKai0716]

First, I encountered the following errors during inference by replacing inference_params=None with other settings.

from mamba_ssm.utils.generation import InferenceParams
m = Mamba(dim_model, layer_idx=7).to(device=device)
inference_all = InferenceParams(max_seqlen=16, max_batch_size=3)
y_with_inference = m(x_all, inference_params=inference_all)

TypeError: InferenceParams.init() got an unexpected keyword argument 'max_seqlen'

Secondly, the following error showed up when I called the Mamba2 from mamba_ssm

from mamba_ssm import Mamba2
self.mamba_1 = Mamba2(d_model=7, d_state=16, d_conv=3, expand=2, headdim=7).to("cuda")

TypeError: causal_conv1d_fwd(): incompatible function arguments. The following argument types are supported: 1. (arg0: at::Tensor, arg1: at::Tensor, arg2: Optional[at::Tensor], arg3: Optional[at::Tensor], arg4: bool) -> at::Tensor
Invoked with: tensor([[[-26.1522, -15.0262, -14.7450],
[ 2.7521, -1.2708, -4.2284],
[-19.3438, -14.0407, -6.7787],
...,

I would appreciate it if someone could help me resolve the confusion above.