-
Notifications
You must be signed in to change notification settings - Fork 74k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
standalone pip package for tf.io.gfile.GFile #45942
Comments
@kpe Do you mean something like this https://github.com/dask/gcsfs ? |
@vnvo2409 - thank you, I've missed (I still like the way how |
I think it is possible to compile a pip package for gfile without tensorflow but it will require a lot of efforts. |
After conversion to modular filesystems fully lands, we might revisit this, as it becomes possible. |
+1 gfile is a beautiful tool which is not only for GCS but also other file systems like HDFS. It will be great we can have it as a standalone package. |
+1 for a standalone package. HDFS access through GFile is nice to have independently of TF |
+1 |
Is there a chance to somehow split the
io.gfile
into a standalone pip package? This would be great!The current
tf.io.gfile.GFile
API is very handy for transparently accessing local or GCS bucket files.It is so great, I wish I could use it even in projects where I don't have tensorflow. As far as I'm aware the only/main alternative is to use the
google-cloud-storage
pip package (google.cloud.storage.Client
), but it does not provide a python ideomatic API and cannot transparently handle local files.I'm not sure where and what would be the best way to handle this, therefore I posted a related issue at googleapis/python-storage#354.
The text was updated successfully, but these errors were encountered: