pretrained-models.pytorch
pretrained-models.pytorch copied to clipboard
http://data.lip6.fr/cadene/pretrainedmodels/ has no connection
I tried to load pre-trained model "inceptionresnetv2", but it failed to connect. The command as below: pretrainedmodels.dict[inceptionresnetv2](num_classes=1000, pretrained='imagenet')
and I got the error: HTTPConnectionPool(host='data.lip6.fr', port=80): Max retries exceeded with url: /cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7fd5f2e1f0f0>: Failed to establish a new connection: [Errno 110] Connection timed out',))
I was confused for a long time. thank you.
Hi. I seems to experience the same connectivity issue (resnext50 model weights are also unavailable). It seems the whole http://data.lip6.fr/ is down.
@JasperShian @BloodAxe Yes it is down. It should come back in less than an hour.
Sorry for that :/
Thank your reply, it's available now :)
Is it closed now?
Tere is an unfortunate problem with our servers. It will come back soon.
Sorry for the inconveniences.
On Sun, May 5, 2019, 10:21 Magic-chao [email protected] wrote:
Is it closed now?
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Cadene/pretrained-models.pytorch/issues/152#issuecomment-489402692, or mute the thread https://github.com/notifications/unsubscribe-auth/ABDW6LX3XIBALUZ2QXFPSRDPT2KITANCNFSM4HJJA6RQ .
@Cadene Could you provide SE-Resnet101 architecture and weights , which is supported in the MMdnn now. I see you provided now(se_resnet101-7e38fcc6.pth) is only contain weights which can't be converted in MMdnn. Thanks!
Hi, anyone else experiencing issues with http://data.lip6.fr/? Seems like the whole site is down. :(
@prnvkbr I know why. I will try to push a fix in the next 24 hours
Thanks for your quick response, @Cadene!
I tried loading fbresnet 152 from this URL: "http://data.lip6.fr/cadene/pretrainedmodels/fbresnet152-2e20f6b4.pth". I am getting the following error.
urllib.error.URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: certificate has expired (_ssl.c:1091)>
I get the same error as @shreyan2.
urllib.error.URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: certificate has expired (_ssl.c:1131)>
@Cadene any workarounds?
download it manually to the desired dir path without ssl verification:
Something like:
wget --no-check-certificate http://data.lip6.fr/cadene/pretrainedmodels/se_resnext50_32x4d-a260b3a4.pth -P ~/.cache/torch/hub/checkpoints
您的邮件,我已收到,谢谢,祝好~