[BUG] load_classification function not working and time-series classification's website down

[fmorenopino]

Describe the bug

Hi, I think that the website https://www.timeseriesclassification.com/index.php has been down for a couple of days now. Further, on python, I cannot use the function load_classification (from aeon.datasets import load_classification) to download non-previously used datasets (for example, X, Y = load_classification("PhonemeSpectra") is not working).

Steps/Code to reproduce the bug

from aeon.datasets import load_classification
X, Y = load_classification("PhonemeSpectra")

Expected results

No error is thrown

Actual results

Traceback (most recent call last):
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 1348, in do_open
    h.request(req.get_method(), req.selector, req.data, headers,
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 1283, in request
    self._send_request(method, url, body, headers, encode_chunked)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 1329, in _send_request
    self.endheaders(body, encode_chunked=encode_chunked)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 1278, in endheaders
    self._send_output(message_body, encode_chunked=encode_chunked)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 1038, in _send_output
    self.send(msg)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 976, in send
    self.connect()
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 1448, in connect
    super().connect()
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/http/client.py", line 942, in connect
    self.sock = self._create_connection(
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/socket.py", line 824, in create_connection
    for res in getaddrinfo(host, port, 0, SOCK_STREAM):
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/socket.py", line 955, in getaddrinfo
    for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
socket.gaierror: [Errno -5] No address associated with hostname

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/nfs/home/fernandom/.local/lib/python3.10/site-packages/aeon/datasets/_data_loaders.py", line 1359, in load_classification
    raise e
  File "/nfs/home/fernandom/.local/lib/python3.10/site-packages/aeon/datasets/_data_loaders.py", line 1354, in load_classification
    response = urlopen(req, timeout=60)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 216, in urlopen
    return opener.open(url, data, timeout)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 519, in open
    response = self._open(req, data)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 536, in _open
    result = self._call_chain(self.handle_open, protocol, protocol +
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 496, in _call_chain
    result = func(*args)
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 1391, in https_open
    return self.do_open(http.client.HTTPSConnection, req,
  File "/nfs/home/fernandom/miniconda3/envs/contrastive/lib/python3.10/urllib/request.py", line 1351, in do_open
    raise URLError(err)
urllib.error.URLError: <urlopen error [Errno -5] No address associated with hostname>

Versions

System: python: 3.10.13 | packaged by conda-forge | (main, Dec 23 2023, 15:36:39) [GCC 12.3.0] executable: /nfs/home/fernandom/miniconda3/envs/contrastive/bin/python machine: Linux-5.15.0-101-generic-x86_64-with-glibc2.35

Python dependencies: pip: 23.3.2 setuptools: 68.0.0 scikit-learn: 1.3.0 aeon: 0.8.1 statsmodels: None numpy: 1.24.3 scipy: 1.11.1 pandas: 2.0.3 matplotlib: 3.7.2 joblib: 1.3.1 numba: 0.57.1 pmdarima: None tsfresh: None

[TonyBagnall]

hi, yes the website is down at the moment, sorry about that. We are switching over accounts for the running of timeseriesclassification.com and it is taking longer than I hoped. Should be up and running by the end of the week.

[fmorenopino]

Thanks @TonyBagnall. Is there any alternative I can use to 'load_classification' in python besides manually downloading all the datasets?

[TonyBagnall]

not to my knowledge, but there might be other locations. If you download the datasets, load_classification will load from disk if you give it the location, or use load_from_tsfile. I'll post here when its back up, but I need to rebuild all the databases and upload all the files even once they have done the switch and its just me doing tsc.com in my spare time really.

[TonyBagnall]

UK2.net not coming out of this well btw!

[fmorenopino]

Thanks for the feedback, @TonyBagnall. I will give it a try to load them locally in the meantime. Thanks again.

[TonyBagnall]

we can give you the datasets if you need them, just email us

[MatthewMiddlehurst]

We should keep this open while the website remains down IMO

[fmorenopino]

Hi, is there any update on this? Thanks again.

[lmriccardo]

It seems that the site is still down

[MatthewMiddlehurst]

We are hoping this should be resolved soon, but as mentioned Tony can provide the datasets if you email him at [email protected].

[TonyBagnall]

update, our ISP are still being hopeless, so we are backing up the data onto zenodo and will release a patch to use zenodo as backup

[TonyBagnall]

@lmriccardo @fmorenopino Everything should now work. FYI we have also uploaded all the data to zenodo and if timeseriesclassification.com is down, load_classifier (and shortly load_regression) go there instead. Any problems, let me know, otherwise I will close this issue

[fmorenopino]

It looks like everything is working fine. Thank you so much!

[TonyBagnall]

going to close this now, all should work (and be much faster!)