This method mimics pandas `drop_duplicates` when axis=rows.
For axis=cols this method will drop columns with 0-variance (only
1 unique value). NaNs (`_fill_value`) won't be counted as unique values.
So the following column (`-` would be the `_fill_value`) will be dropped:
foo
-
foo
foo
Intermediate PR in removing path_info from our code base.
It is completely removed from the main codebase and is still
being used in tests, which will be migrated in the followups.
For fs-specific simple path manipulation, we've introduced
`fs.path` helper (similar to `os.path`).
Fixes#6774
It comes with the support for no. of parameters that matches with json.dumps.
See: https://github.com/willmcgugan/rich/pull/1547 and
https://github.com/willmcgugan/rich/pull/1644.
Also added support for printing json to stderr. Before we used to use json.dumps
and rich.json.JSON based on isatty. Now we use rich.print_json on both
tty and non-tty conditions, which may avoid different behaviours in the
future.
Looks like azure-core is dependent on a dependency `chardet` that aiohttp brought it with, which has been replaced with `charset_normalizer`.
But azure-core depends on internal of aiohttp and looks like it tries to set it's own encoding parser, `chardet` which is now not available.
Writing at the end of each step (create_summary) can mess up with some ML Framework progress bars, I moved the message to the monitor task, as the path won't change during training.
Fixesiterative/dvclive#177
We were previously using relpath in *all* cases, mainly to support WSL but relpath doesn't work
on macOS.
Added a check for using relpath *only* for WSL.
Fixes#6092
Currently whether or not link confirmation will run depends on
particular filesystem and existing workspace link type, which introduces
inconsistency in behaviour, where we sometimes catch unsupported links
before touching user data and sometimes we don't.
Depending on a filesyste (e.g. on NTFS), `_remove` might reset
read-only permissions in order to delete a hardlink to protected object,
which will also reset it for the object itself, making it unprotected,
so we need to protect it back.
Currently it depends on the whole `info()` output, which has too much
information that doesn't really matter. To check if file contents didn't
change mtime, size, inode combo is enough.
* executor: move local specific git initialization out of base
* machine: return kwargs for executor init instead of executor object
* executor: add initial ssh executor
* ssh-executor: disable strict host checking for dvc-machine instances
* initial tests for ssh executor
* drop GIT_SSH_COMMAND usage
* push: accept destination odb as a parameter
* separate cache initialization into executor.init_cache
* add test for ssh executor cache init
* remove dead code
* machine: fixture-ize default test scenarios
* test SSHExecutor.from_machine
* tpi: return default SSH port for constructing executors
* iterate on ui, has tree and shows to-be-generated yaml content
* order fs tree structure alphabetically
* always show the heading
* hide dvclive on tree if not on live mode
* Change message at the end
* s/skip/omit
* add link on the header
* remove newline on non-interactive mode
* Add experiment name check.
1. Add experiment name check (https://git-scm.com/docs/git-check-ref-format)
2. Add duplicate exp name check.
3. Add some unit test for it.
* Ban slash / in dvc exp names
* Use dulwich backend for ref name checking
* Some bug fix
* Update dvc/repo/experiments/__init__.py
Co-authored-by: Peter Rowlands (변기호) <[email protected]>
* Update dvc/repo/experiments/__init__.py
Co-authored-by: Peter Rowlands (변기호) <[email protected]>
* Some review changes
* Make some funtion more reusable.
Co-authored-by: Peter Rowlands (변기호) <[email protected]>
If supported, reflink, will provide the fastest transfer possible.
That being said, right now we use `move` in most of the places,
which still takes presidence.
Pre-requisite for #6387
scm.install does 3 different things, and all of those are
very much tied to DVC. So, as we are splitting scm out,
it should now belong to `dvc.install`.
Linux/Windows machines have only 2 cores, whereas macOS machines have 3 cores. Maybe we are stretching available resources by 2x, let's see if `-n=auto` improves test freezes.
* git: support client kwargs for auth in remote methods
* tests: use docker for git/ssh server
* move dulwich backend into subdir
* add base sync<->async support
* dulwich: add initial asyncssh SSH vendor
* dulwich: use asyncssh as the default SSH vendor
* add test for git over ssh
* test read (fetch) over ssh
* ci: re-enable ssh tests
temporary workaround for #6788, #6789
* move async stubs into dvc.scm.asyn
* tests: use sshfs and not dvc.fs.ssh
* allow explicit 0-length read in asyncssh vendor
Ruslan Kuprieiev
David de la Iglesia Castro
dependabot[bot]
Batuhan Taskaya
Guðmundur HeimissonandRuslan Kuprieiev
Saugat Pachhai (सौगात)
karajan1001
Paweł Redzyński
pre-commit-ci[bot]
Ruslan Kuprieiev
Peter Rowlands (변기호)
Dave Berenbaum