ToDo.md

1. Code "training" an AND & OR (1 layer) and XOR (2 layers) model! ;-)

2. https://developers.google.com/machine-learning/crash-course finished?

3. https://codelabs.developers.google.com/tensorflowjs-coral-tflite-node

4. https://www.tensorflow.org/resources/learn-ml

5. https://www.tensorflow.org/tfx/tutorials/serving/rest_simple

6. LLM @ home? https://github.com/Hannibal046/Awesome-LLM

7. Waz https://github.com/edgeimpulse/voice-activated-microbit for https://www.edgeimpulse.com/blog/voice-activated-microbit ?

8. https://colossalai.org & https://github.com/hpcaitech/ColossalAI

9. Use Interpreter, ask https://youtu.be/HxtBIwfy0kM ? Try links in description.

10. Try to hand-draw a digit on 28x28 pixels, or a bigger JPEG or PNG image file, and downsize it with Py, and load that image file and have it recognized!

11. TFX Serving for Web, e.g. https://youtu.be/vhYBaxHs2kI

12. Use model in https://www.tensorflow.org/js and hack a Web App with a 28x28 pixels Canvas to draw on!

13. TF lite instead of full TF in beginner-infer.py AFTER tensorflow/tensorflow#58264 is fixed (or after I can self build lite on a machine with more memory)

14. Run the model with TF lite on the Coral board, using fixed image; then with the webcam!

15. WARNING:tensorflow:No training configuration found in save file, so the model was *not* compiled. Compile it manually.

16. LOW PRIORITIY: Write Dockerfile to run that .py in a container (using https://www.tensorflow.org/install ?)

17. Learn https://fedoramagazine.org/mlcube-and-podman/

18. https://mlcommons.github.io/mlcube/getting-started/mnist/

19. https://colab.research.google.com/notebooks/gpu.ipynb

20. https://colab.research.google.com/notebooks/tpu.ipynb

21. Run on GCP?

    • https://cloud.google.com/tpu/docs/intro-to-tpu
    • https://cloud.google.com/tpu/docs/maintenance-events
    • https://cloud.google.com/tpu/docs/managing-tpus-tpu-vm
    • https://cloud.google.com/tpu/docs/system-architecture-tpu-vm
    • https://cloud.google.com/tpu/docs/training-on-tpu-pods

22. tf.distribute, try bit.ly/keras-TPU from https://youtu.be/W7A-9MYvPwI; also see https://youtu.be/NbID_AqdZXM and https://youtu.be/ZnukSLKEw34

23. https://www.tensorflow.org/datasets

24. https://www.kaggle.com

25. https://modelzoo.co

26. https://www.tensorflow.org/lite/models/modify/model_maker

27. https://github.com/iterative/dvc

28. https://www.youtube.com/c/HuggingFace

29. https://huggingface.co/bert-base-uncased#how-to-use

30. https://github.com/google-research/bert#using-bert-in-colab

31. https://aws.amazon.com/blogs/industries/novartis-ag-uses-amazon-to-build-knowledge-graph-using-bert/

32. https://www.kubeflow.org/docs/started/introduction/

33. How to run Colab or iPython locally on a .ipynb? Try VS Code and locally install missing pip packages? or https://jupyter-docker-stacks.readthedocs.io/en/latest/using/selecting.html or https://www.tensorflow.org/install.