PyTorch and TensorFlow are two of the most popular deep learning frameworks.
- PyTorch is an open-source ML and DL framework developed by Facebook (Meta) using Python and released to the public in 2016.
- Now open source & used by companies such as Tesla, Microsoft, and OpenAI.. etc
- PyTorch is a low-level API for NLP (natural language processing) and computer vision. It is a more powerful version of Numpy.
- PyTorch is used by leading tech companies like
- Tesla to build the computer vision system for their self-driving vehicles
- Meta to power the curation and understanding systems for their timelines.
- Apple to create computationally enhanced photography and even automated weed-killing tractors.
- What can PyTorch be used for?
Problems with long lists of rules — when the traditional approach fails, machine learning/deep learning may help.
Continually changing environments — deep learning can adapt (‘learn’) to new scenarios.
Discovering insights within large collections of data — can you imagine trying to hand‑craft rules for what 101 different kinds of food look like?
Category Algorithms Machine Learning Random forest; Gradient boosted models; Naive Bayes; Nearest neighbour; Support vector machine; …many more Deep Learning Neural networks; Fully connected neural network; Convolutional neural network; Recurrent neural network; Transformer; …many more
Tensor Explanation from root level
9 ---> (single number) ---> scalar (0 D Tensor) (D means Dimensional)
[19, 82] (group of numbers --> Row) ---> Vector (1D Tensor)
[[9 ,88 ] ,
[19 ,8]] (combination of rows & colums) --> Matrix ((2 D Tensor))
3X3 Matrix (Red, Green, and Blue) ---> 3D Tensor (Ex: Image)
4D or 5D tensor (Videos or large data)
- Images are first converted into numerical values called tensors.
- These tensors are given to a neural network, which processes them using mathematical calculations and generates an output tensor.
- This output is then transformed into meaningful information that humans can easily understand.
CUDA (Compute Unified Device Architecture) is a parallel computing platform and programming model developed by NVIDIA that allows software developers to use a GPU for general-purpose computing, not just graphics.
What CUDA does?
Normally, GPUs were designed mainly for graphics. CUDA lets you use the thousands of GPU cores to perform computational tasks much faster than a CPU for certain workloads.
TensorFlow was developed by Google and released to the public in 2015. Production and research are the main uses of TensorFlow.
