Onepanel's core concepts can be categorized as follows:
- Data Annotation
Projects allow you to organize your code, workspaces and jobs in a way that you can share and collaborate with your teammates. You easily share project by adding other team members or share them publicly.
See Getting Started with Projects for more information.
Workspaces are full Linux computing environments that come pre-installed with all the tools you need to explore data and build and experiment with your models. They include Python, JupyterLab, TensorFlow, PyTorch and other well known deep learning libraries and tools, as well as full terminal access. You can optionally install custom packages and dependencies into your workspaces.
A single workspace can be paused and resumed at any time or be upgraded to up to 8 GPUs and downgraded back to CPU with ease.
See Getting Started with Workspaces for more information.
Much like workspaces, jobs are also Linux environments that you can in parallel on multiple machines; this allows you to try out different hyper-parameters, code and datasets and then compare the results and metrics for each job and continue iterating on the best results.
While a job is running, you can view running logs, system metrics and model/training metrics via TensorBoard. You also have full terminal access to each running job.
Once a job completes, it automatically saves a snapshot of the logs, code, command, datasets, environment and output. This allows you to share your results with others and be able to fully reproduce/re-run the job at a later time.
See Getting Started with Jobs for more information.
We have integrated CVAT - an open source video and image annotation tool. Now you can quickly and easily leverage third-party data annotators and provide direct access to your datasets for annotation all within Onepanel.
See annotation tools for more information.