GiganticTurtle0 , are you using the latest release or the RC?
Also, can you try with dataset.upload(output_url="/home/user/server_local_storage/clearml_training_dataset/")
(note the added '/' at the end of the line)
By adding the slash I have been able to see that indeed the dataset is stored in output_url . However, when calling finalize , I get the same error. And yes, I have installed the version corresponding to the last commit :/
GiganticTurtle0 what's the Dataset Task status?
GiganticTurtle0 , I managed to reproduce and make it work once. Let me take a look
AgitatedDove14 In the 'status.json' file I could see the 'is_dirty' flag is set to True
Well the 'state.json' file is actually removed after the exception is raised
I can't figure out what might be going on
GiganticTurtle0 , it looks like an issue with the latest RC. We're working on it to fix it 🙂
Thanks, I'd appreciate it if you let me know when it's fixed :D
GiganticTurtle0 fix was just pushed to GitHub 🙂pip install git+
AgitatedDove14 Oops, something still seems to be wrong. When trying to retrieve the dataset using get_local_copy() I get the following error:Traceback (most recent call last): File "/home/user/myproject/lab.py", line 27, in <module> print(dataset.get_local_copy()) File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/datasets/dataset.py", line 554, in get_local_copy target_folder = self._merge_datasets( File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/datasets/dataset.py", line 1342, in _merge_datasets target_base_folder = self._create_ds_target_folder( File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/datasets/dataset.py", line 1291, in _create_ds_target_folder cache.lock_cache_folder(local_folder) File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/storage/cache.py", line 248, in lock_cache_folder lock.acquire(timeout=0) File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/utilities/locks/utils.py", line 130, in acquire fh = self._get_fh() File "/home/user/.conda/envs/myenv/lib/python3.9/site-packages/clearml/utilities/locks/utils.py", line 200, in _get_fh return open(self.filename, self.mode, **self.file_open_kwargs) FileNotFoundError: [Errno 2] No such file or directory: '/home/user/.clearml/cache/storage_manager/datasets/.lock.000.ds_38e9acc8d56441999e806815abddee82.clearml'Main code is the same as above, I'm just adding dataset.get_local_copy() at the end. It seems it resolves the path with a .lock file. Weird...
Well I tried several things but none of them have worked. I'm a bit lost
GiganticTurtle0 , I tried running the same script as before and added dataset.get_local_copy() at the end and it managed to work fine. Do you have any other changes? Are you on the latest repo code?
Yes, I'm working with the latest commit. Anyway, I have tried to run dataset.get_local_copy() on another machine and it works. I have no idea why this happens. However, on the new machine get_local_copy() does not return the path I expect. If I have this code:dataset.upload( output_url="/home/user/server_local_storage/mock_storage" )I would expect the dataset to be stored under the path specified in output_url . But what I get with get_local_copy() is the following path:'/home/user/.clearml/cache/storage_manager/datasets/ds_98d1bfbbb7334f50a4113409b4d691be'
Is this usual?
But what I get with
get_local_copy()
is the following path: ...
Get local path will return an immutable copy of the dataset, by definition this will not be the "source" storing the data.
(Also notice that the dataset itself is stored in zip files, and when you get the "local-copy" you get the extracted files)
Make sense ?
Mmm but what if the dataset size is too large to be stored in the .cache path? It will be stored there anyway?
too large to be stored in the .cache path? It will be stored there anyway?
oh that is exactly why the latest release supports chunks, so you can get a partial copy 🙂
nonetheless, the assumption is that you will have to end up with the data locally, otherwise the network becomes a huge bottleneck
make sense ?
Indeed it does! But what still puzzles me so badly is why I get below path when running dataset.get_local_copy() on one of the machines of my cluster:/home/user/.clearml/cache/storage_manager/datasets/.lock.000.ds_61ff8d4335dd4b74bd78c3576fa44131.clearml
Why is it pointing to a .lock file?
In fact, the datasets directory does not even exist
FileNotFoundError: [Errno 2] No such file or directory: '/home/user/.clearml/cache/storage_manager/datasets/.lock.000.ds_38e9acc8d56441999e806815abddee82.clearml'
Let me check this issue, it seems like the locking mechanism should have figured that there is no lock...
Found it
GiganticTurtle0 you are 🧨 ! thank you for stumbling across this one as well.
Fix will be pushed later today 🙂
