"Invalid bucket name" error when downloading OPENAI data

[lixiaoyun0239]

Hi, I'm trying to test the benchmark code and run into the following error.

=====================================

Code:
from vectordb_bench.backend.dataset import Dataset, DatasetManager
import numpy as np

size = 500000
data_name = 'openai'

data = Dataset.OPENAI.manager(size)
data.prepare(check=False)

====================================
data.prepare(check=False) raises the following error:

2024-03-01 00:37:05,316 | INFO: local file: \tmp\vectordb_bench\dataset\openai\openai_medium_500k\test.parquet not match with remote: assets.zilliz.com\benchmark\openai_medium_500k\test.parquet; add to downloading list (data_source.py:136) (4644)
2024-03-01 00:37:05,316 | INFO: local file: \tmp\vectordb_bench\dataset\openai\openai_medium_500k\neighbors.parquet not match with remote: assets.zilliz.com\benchmark\openai_medium_500k\neighbors.parquet; add to downloading list (data_source.py:136) (4644)
2024-03-01 00:37:05,316 | INFO: local file: \tmp\vectordb_bench\dataset\openai\openai_medium_500k\neighbors_tail_1p.parquet not match with remote: assets.zilliz.com\benchmark\openai_medium_500k\neighbors_tail_1p.parquet; add to downloading list (data_source.py:136) (4644)
2024-03-01 00:37:05,316 | INFO: local file: \tmp\vectordb_bench\dataset\openai\openai_medium_500k\neighbors_head_1p.parquet not match with remote: assets.zilliz.com\benchmark\openai_medium_500k\neighbors_head_1p.parquet; add to downloading list (data_source.py:136) (4644)
2024-03-01 00:37:05,316 | INFO: local file: \tmp\vectordb_bench\dataset\openai\openai_medium_500k\shuffle_train.parquet not match with remote: assets.zilliz.com\benchmark\openai_medium_500k\shuffle_train.parquet; add to downloading list (data_source.py:136) (4644)
2024-03-01 00:37:05,316 | INFO: Start to downloading files, total count: 5 (data_source.py:142) (4644)
0%| | 0/5 [00:00<?, ?it/s]
Traceback (most recent call last):

File ~\anaconda3\envs\env1\Lib\site-packages\spyder_kernels\py3compat.py:356 in compat_exec
exec(code, globals, locals)

File c:\users\bdfzl\desktop\env1\vdb_bench.py:9
data.prepare(check=False)

File ~\anaconda3\envs\env1\Lib\site-packages\vectordb_bench\backend\dataset.py:202 in prepare
source.reader().read(

File ~\anaconda3\envs\env1\Lib\site-packages\vectordb_bench\backend\data_source.py:145 in read
self.fs.download(s3_file, local_ds_root.as_posix())

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\spec.py:1534 in download
return self.get(rpath, lpath, recursive=recursive, **kwargs)

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\asyn.py:118 in wrapper
return sync(self.loop, func, *args, **kwargs)

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\asyn.py:103 in sync
raise return_result

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\asyn.py:56 in _runner
result[0] = await coro

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\asyn.py:650 in _get
return await _run_coros_in_chunks(

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\asyn.py:254 in _run_coros_in_chunks
await asyncio.gather(*chunk, return_exceptions=return_exceptions),

File ~\anaconda3\envs\env1\Lib\asyncio\tasks.py:452 in wait_for
return await fut

File ~\anaconda3\envs\env1\Lib\site-packages\fsspec\callbacks.py:81 in func
return await fn(path1, path2, callback=child, **kwargs)

File ~\anaconda3\envs\env1\Lib\site-packages\s3fs\core.py:1224 in _get_file
body, content_length = await _open_file(range=0)

File ~\anaconda3\envs\env1\Lib\site-packages\s3fs\core.py:1215 in _open_file
resp = await self._call_s3(

File ~\anaconda3\envs\env1\Lib\site-packages\s3fs\core.py:348 in _call_s3
return await _error_wrapper(

File ~\anaconda3\envs\env1\Lib\site-packages\s3fs\core.py:140 in _error_wrapper
raise err

File ~\anaconda3\envs\env1\Lib\site-packages\s3fs\core.py:113 in _error_wrapper
return await func(*args, **kwargs)

File ~\anaconda3\envs\env1\Lib\site-packages\aiobotocore\client.py:345 in _make_api_call
api_params = await self._emit_api_params(

File ~\anaconda3\envs\env1\Lib\site-packages\aiobotocore\client.py:475 in _emit_api_params
await self.meta.events.emit(

File ~\anaconda3\envs\env1\Lib\site-packages\aiobotocore\hooks.py:66 in _emit
response = await resolve_awaitable(handler(**kwargs))

File ~\anaconda3\envs\env1\Lib\site-packages\botocore\handlers.py:288 in validate_bucket_name
raise ParamValidationError(report=error_msg)

ParamValidationError: Parameter validation failed:
Invalid bucket name "assets.zilliz.com\benchmark\openai_medium_500k\test.parquet": Bucket name must match the regex "^[a-zA-Z0-9.-_]{1,255}$" or be an ARN matching the regex "^arn:(aws).:(s3|s3-object-lambda):[a-z-0-9]:[0-9]{12}:accesspoint[/:][a-zA-Z0-9-.]{1,63}$|^arn:(aws).*:s3-outposts:[a-z-0-9]+:[0-9]{12}:outpost[/:][a-zA-Z0-9-]{1,63}[/:]accesspoint[/:][a-zA-Z0-9-]{1,63}$"

===========================

Operating system: Windows 11, Anaconda env with Python 3.11

=================== ========
Installation:
pip install vectordb-bench[all]

Thank you for the help.

[XuanYang-cn]

In windows, dataset source should be consistent with Linux like xxxxx/benchmark/openai_medium_500k/test.parquet, fixing it

[XuanYang-cn]

0.0.7 released, should be fixed