buffer.dump(f) method not working (Latest d3rlpy v2.0.2 installed)

[HateBunnyPlzzz]

When running the code example in the newest documentation of Data Collection in tutorial section, the buffer.dump(f) method in the loop is throwing of UnsupportedOperation: read error

To reproduce: run the following code

# setup algorithm
dqn = d3rlpy.algos.DQNConfig().create()

# prepare experience replay buffer
buffer = d3rlpy.dataset.create_fifo_replay_buffer(limit=100000, env=env)

# prepare exploration strategy if necessary
explorer = d3rlpy.algos.ConstantEpsilonGreedy(0.3)

# start data collection
dqn.fit_online(env, buffer, explorer, n_steps=100000)

# save ReplayBuffer
with open("replay_dataset.h5", "wb") as f:
  buffer.dump(f)

Expected behavior
Expected to dump the replay_buffer.h5 file successfully

for more details here's the full error code below

UnsupportedOperation                      Traceback (most recent call last)
Cell In [11], line 2
      1 with open("replay_dataset.h5", "wb") as f:
----> 2   buffer.dump(f)

File ~/anaconda3/envs/ICRL/lib/python3.10/site-packages/d3rlpy/dataset/replay_buffer.py:245, in ReplayBuffer.dump(self, f)
    234 def dump(self, f: BinaryIO) -> None:
    235     """Dumps buffer data.
    236 
    237     .. code-block:: python
   (...)
    243         f: IO object to write to.
    244     """
--> 245     dump(self._buffer.episodes, f)

File ~/anaconda3/envs/ICRL/lib/python3.10/site-packages/d3rlpy/dataset/io.py:25, in dump(episodes, f)
     23 keys = list(serializedData[0].keys())
     24 with h5py.File(f, "w") as h5:
---> 25     h5.create_dataset("columns", data=keys)
     26     h5.create_dataset("total_size", data=len(episodes))
     27     for key in keys:

File ~/anaconda3/envs/ICRL/lib/python3.10/site-packages/h5py/_hl/group.py:183, in Group.create_dataset(self, name, shape, dtype, data, **kwds)
    180         parent_path, name = name.rsplit(b'/', 1)
...
File h5py/_proxy.pyx:146, in h5py._proxy.dset_rw()

File h5py/h5fd.pyx:163, in h5py.h5fd.H5FD_fileobj_read()

UnsupportedOperation: read

[takuseno]

@HateBunnyPlzzz Thank you for reporting this! This has been fixed at this commit: cb9ca69 . I'll release a patch that includes this fix tomorrow.

[takuseno]

The latest patch has been released.
https://github.com/takuseno/d3rlpy/releases/tag/v2.0.4