I need to upload a number of checkpoints for ConvNets (weights + optimizers, all dicts of pytorch tensors), each about 0.5Gb anonymously. I don't want to use Google Drive. I trained models on the university cluster (if it's relevant).
Where can I load these files anonymously? The files must be publicly available, but my identity must remain anonymous.
wget, you have to rename the output file and set its file extension though. I normally usewget -O new_filename https://file.io/XXXXXor just download it and rename it later. You can also increase the 14 day limit by usingw,m, oryto set expiration in weeks, months, or years. I haven't used that before but assume it'd be used ascurl -F "file=@test.txt" https://file.io/?expires=1y. Check https://www.file.io/#one for more usage details. Hope this helps – Shan S Jun 02 '20 at 13:31