search

higgsfield / RL-Adventure

Pytorch Implementation of DQN / DDQN / Prioritized replay/ noisy networks/ distributional values/ Rainbow/ hierarchical RL
2.99k stars 587 forks source link

Update 4.prioritized dqn.ipynb #26

Open jsrimr opened 5 years ago

jsrimr commented 5 years ago

removed volatile parameter and added wrapper @torch.no_grad() zip(samples) -> list(zip(samples)) .data[0] -> .item() . # 2 cases

Deepakgthomas commented 2 years ago

Hi @jsrimr,

While running the prioritized dqn, I came across this error - TypeError: 'zip' object is not subscriptable

I guess your solution of zip(samples) -> list(zip(samples)) works. But the author was able to run his code without any modification and still gets results. Do you know how that was possible?