Bounds Error when following tutorial of ReinforcementLearning.jl

Hello community,

I have been trying to implement a basic Q-learning algorithm following the tutorial of the ReinforcementLearning.jl. The main issue comes when, instead of giving the agent a RandomPolicy() as policy, I set them with the QBasedPolicy developed in the tutorial. It gives me the following error:

BoundsError: attempt to access 2×7 Matrix{Float64} at index [0, 1]

For reference, my code is as follows:

using Revise
using ReinforcementLearning
using Flux
using ReinforcementLearningTrajectories

#create enviroment
env = RandomWalk1D()
#create policy
S, A = state_space(env), action_space(env)
NS, NA = length(S), length(A)

#basic policy learner (with epsilon greedy explorer). Implementation of q learner
policy = QBasedPolicy(
        learner = TDLearner(
                n_state = NS, 
                n_action = NA,
        explorer = EpsilonGreedyExplorer(0.1) 

#define a trajectory
trajectory = Trajectory(
        state = Int64 => (),
        action = Int64 => (),
        reward = Float64 => (),
        terminal = Bool => (),

agent = Agent(
        policy = policy,
        trajectory = trajectory
run(agent, env, StopAfterNEpisodes(10), TotalRewardPerEpisode())