An AI/ML team's remote storage can quickly become a mess. Dave Berenbaum shares how DVC's new cloud versioning feature works and how it can help you organize your storage and ensure reproducibility.
📸 DVC supplements cloud versioning, recording the version IDs of all files in your project. Snapshots of your cloud-versioned data are maintained, tracked in Git; ensuring reproducibility.
If you are already a DVC user, you get all the benefits you are used to, plus the advantage of seeing an expected file structure in your remote.
No more having to explain the hash folders to colleagues new to DVC! 😉
Read the blog post to get yourself set up with DVC Cloud Versioning and start making your data versioning life a whole lot easier! 😅
[ Ссылка ]
To learn more about Iterative's open-source and SaaS tools please visit:
🧑🏽💻 Our online course: [ Ссылка ]
✍🏼 Our docs: [ Ссылка ]
[ Ссылка ]
[ Ссылка ]
[ Ссылка ]
Join our Discord server: [ Ссылка ]
DVC Cloud Versioning Demo
Теги
Data ScienceMachine learningAIml engineervs codedata version controlci/cdweights and biasesmlflowcomet mlneptune.aipachydermmachine learning experimentationhyperparametersdeep learningconfusion matrixROC AUCaccuracymetaflowdoltDagshubGit LFSLabel StudioAirflowversion controlepochfeature extractionVS CodeGitHubGitLabMLOpsGitOpsclassificationloss curveiterationsparametersneural networkgradient descentNLPcomputer vision