The story begins with a assignment given to me that needed me to deploy a Monocular Single Human Pose Estimation model on AWS Lambda. Me… — The story begins with a assignment given to me that needed me to deploy a Monocular Single Human Pose Estimation model on AWS Lambda. Me being a student, i prefer to be in the free tier of Lambda, where we get about 3GB of RAM and 500MB storage, the storage…

This is not a post for benchmarking TPUs vs GPUs, so let’s not get into too much of the technical details, its just my experience with… — This is not a post for strictly benchmarking TPUs vs GPUs on a highly technical level, so let’s not get into too much of the technical details, its just my experience with TPUs and how the deep learning world is affected by it. Model: ResNet-Unet like architecture, has ~30M parameters

I’ll keep this simple and sweet — I’ll keep this simple and sweet Open this Notebook in Colab and Run: https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/Utils/Colab_25GBRAM_GPU.ipynb That is it, you have a 25GB High RAM 🤩 session in Colab, Go Crazy !

Recently i had to make a dataset of 400K images + 400K image masks, and then had to train them on a Deep Neural Network using the Google… — Recently i had to make a dataset of 400K images + 400K image masks, and then had to train them on a Deep Neural Network using the Google Colab free Tesla P100 GPUs, this article is about the journey i had to go through, and learnt quite…