Data Science Laboratory, developed by Dell Technologies, simplifies your work by allowing model implementation, training, and inference tests to be run in containerized JupyterLab or directly from the Linux terminal. Containerized JupyterLab enables data scientists to use notebooks for interactive development in both Python and R, as shown in the following figure:
Figure 9. JupyterLab interactive notebooks
Data Science Laboratory integrates TensorBoard for monitoring deep learning training jobs, as shown in the following figure:
Figure 10. Data Science Laboratory TensorBoard
TensorBoard enables the visualization and tools that are needed for deep learning experimentation. TensorBoard enables features that are used as a visualization toolkit. You can track and visualize metrics for each job, viewing histograms of weights, biases, or other tensors as they change over time.
Data Science Laboratory provides text and markdown editors to help you build documentation on each project. Also, it provides integrated git management for source control, as shown in the following figure:
Figure 11. Integrated git management for source control
Source control integration with a git source control repository allows developers to save and manage multiple versions of a model from a nonproduction instance. Data scientists can track their models throughout the model development cycle and maintain version control.