Hello everyone,

We have been working on a very interesting AI hobby project for quite sometime now. It's about the work around of a data selection technique such that a predictor as low as the mixtures of gaussian model can be used for probability predictions on the image dataset. It has shown some success on MNIST dataset. We would like your criticism and corrections as there would be the implementation of the major AI techniques and other useful ones.

We would like the current progress of the project in source code to be made available to other hackers for any contributions at derrickdonkorsci@gmail.com
Thank you,

Yours sincerely,
Derrick Donkor
Felix Acquah

TorchVision v0.9 comes with a series of new mobile-friendly models that can be used for Classification, Object Detection and Semantic Segmentation. This article looks into the code of the models and shares implementation, training, configuration and more.
https://pytorch.org/blog/torchvision-mobilenet-v3-implementation/

PyTorch 1.9 is finally out!!

Main features, and improvements from the release for the domain libraries including:

TorchVision: iOS support, new SSD + SSDLite models and more

TorchAudio: non-Python wav2vec 2.0 model support, improved resampling and more

Torch Text: Vocab module for NLP workflows

PyTorch Mobile interpreter and
New models for detection is now out in beta. Few demo apps are also there.

See full details here.

The quickest way to becoming a Deep Learning engineer

- Fine-tune a pre-trained language model
- Learn some word embeddings
- Create a simple custom feed-forward network
- Customize a loss function

You don't need to buy a GPU for machine learning work!

There are other alternatives. Here are some:

1. Google Colab
2. Kaggle
3. Deepnote
4. AWS SageMaker
5. GCP Notebooks
6. Azure Notebooks
7. Cocalc
8. Binder
9. Saturncloud
10. Datablore
11. IBM Notebooks

Spend your time focusing on your problem💪💪. Let others worry about the hardware!!

A research team from Facebook AI and UC Berkeley finds a solution for vision transformers’ optimization instability problem by simply using a standard, lightweight convolutional stem for ViT models. The approach dramatically increases optimizer stability and improves peak performance without sacrificing computation efficiency. | https://bit.ly/3yqqNds

The official introduction example guides on the official site https://www.tensorflow.org/tutorials

more example code https://www.tensorflow.org/text/tutorials/

another one: https:/like/www.tensorflow.org/federated/tutorials/tutorials_overview?hl=sk

keras examples
https://keras.io/examples

The biggest mistake I see with deep learning practitioners new to the PyTorch library is forgetting and/or mixing up the following steps:
1) Zeroing out gradients from the previous steps (opt.zero_grad())
2)Performing backpropagation (loss.backward())
3)Updating model parameters (opt.step())

Failure to perform these steps in this exact order is a surefire way to shoot yourself in the foot when using PyTorch, and worse, PyTorch doesn’t report an error if you mix up these steps, so you may not even know you shot yourself!

The PyTorch library is super powerful, but you’ll need to get used to the fact that training a neural network with PyTorch is like taking off your bicycle’s training wheels — there’s no safety net to catch you if you mix up important steps (unlike with Keras/TensorFlow which allow you to encapsulate entire training procedures into a single model.fit call).

That’s not to say that Keras/TensorFlow are “better” than PyTorch — it’s just a difference between the two deep learning libraries of which you need to be aware. Source: pyimageserch

🎓 Introduction to Deep Learning (by MIT) 🎓

This is one of the top high-quality courses to learn the foundational knowledge of deep learning.

All lectures have been uploaded. 100% Free!
https://youtube.com/playlist?list=PLtBw6njQRU-rwp5__7C0oIVt26ZgjG9NI

fastchan, a new conda mini-distribution
https://www.fast.ai/2021/07/15/fastconda/

Chainer A Powerful, Flexible, and Intuitive Framework for Neural Networks.

✅ Bridge the gap between Algorithms and Implementations OF Deep Learning.

✅ Chainer supports CUDA computation. It only requires a few lines of code to leverage a GPU. It also runs on multiple GPUs with little effort.

✅ Chainer supports various network architectures including feed-forward nets, convnets, recurrent nets and recursive nets. It also supports per-batch architectures.

✅ Chainer is an open source deep learning framework written purely in Python on top of NumPy and CuPy Python libraries. The development is led by Japanese venture company Preferred Networks in partnership with IBM, Intel, Microsoft, and Nvidia.
https://chainer.org/

OpenCV 4.5.3 and 3.4.15 have been
https://opencv.org/opencv-4-5-3/

Introducing Triton:
Open-Source GPU Programming for building a Neural Networks

OpenAI has released an open-source Python like language for writing highly efficient GPU code.
https://openai.com/blog/triton/