Mmdetection: The pre-training model cannot be downloaded!

[zihaozhang9]

When I tried to train the SSD model using ssd300_voc.py, I received some error logs
Downloading: "https://s3.ap-northeast-2.amazonaws.com/open-mmlab/pretrain/third_party/vgg16_caffe-292e1171.pth" to /home/nnir712/.cache/torch/checkpoints/vgg16_c$ ffe-292e1171.pth Traceback (most recent call last): File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/urllib/request.py", line 1317, in do_open encode_chunked=req.has_header('Transfer-encoding')) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 1244, in request self._send_request(method, url, body, headers, encode_chunked) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 1290, in _send_request self.endheaders(body, encode_chunked=encode_chunked) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 1239, in endheaders self._send_output(message_body, encode_chunked=encode_chunked) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 1026, in _send_output self.send(msg) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 966, in send self.connect() File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 1406, in connect super().connect() File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/http/client.py", line 938, in connect (self.host,self.port), self.timeout, self.source_address) File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/socket.py", line 727, in create_connection raise err File "/home/nnir712/anaconda3/envs/open-mmlab/lib/python3.7/socket.py", line 716, in create_connection sock.connect(sa) TimeoutError: [Errno 110] Connection timed out
Then I used wget to download the file and found that it could be downloaded.
wget https://s3.ap-northeast-2.amazonaws.com/open-mmlab/pretrain/third_party/vgg16_caffe-292e1171.pth
Maybe I am in China and my Internet is not good.

[lxtGH]

Yes because you are in China :)

[kingwpf]

大兄弟，复制到迅雷下载，实测满速

[hellock]

You may also try the Aliyun mirror.

[xincyu]

maybe you can copy the downline to the firefox and downlaod the pthfile cross the GreatWall Fire by using vpn;
and set in config file: load_from = 'path/to/your_pth_file'

Please correct me if I'm wrong