• [ ] watch -n 0.5 nvidia-smi, glances, ncdu, (conda install ncdu glances). Screen record.

• [ ] Shoutout a few of my favorite models?

  • [ ] Detectron2, GPT, Dall-E, Stable-Baselines 3 (reinforcement agent/environment)
  • [ ] High-level libraries
    • [ ] https://github.com/pytorch/ignite, Here is Poutyne,

• [ ] Graph learning (TorchGeometric, Deep Graph Library (DGL), NetworkX (build graph data structures here and import into DL libs))

  • [ ] SPARQL. Can be hard to go from unstructured data to SPARQL.

• [ ] Ensemble. Common pre-processing steps.

• [ ] Data pipelines.

  • [ ] Rsync It uses an algorithm to minimize the amount of data copied by only moving the portions of files that have changed.

• [ ] Compressed CSVs in pandas

MLFlow exists, no password: MLflow (ncsa.cloud)

• [ ] Building Machine Learning Pipelines: Common Pitfalls - neptune.ai
• [ ] NLOps: clearML https://github.com/allegroai/clearml Auto-Magical Suite of tools to streamline your ML workflowExperiment Manager, MLOps and Data-Management