Tensor Flow – Google’s Open Source Machine Learning Software Library
I hope the video must have given you an insight about Tensor Flow. For more information let’s go through some quick referential data.
As per our resources, Google describes project as..
“Tensor Flow is an open source software library for numerical computation with the help of Data Flow Graphs.”
Yes, it is Open-Sourced under the Apache 2.0 Open-Source Licence.
How it came into Picture?
Tensor Flow was originally developed by researchers and engineers working on the Google Brain Team (Artificial Intelligence and Machine Learning Groups). The soul purpose of this team was to conduct machine learning and deep neural networks research. But, they come up with such innovative solution and software that is general enough to be applicable in a wide variety of other domains as well.
How it Works?
The base mechanism behind this whole scenario is based on Data Flow Graphs / Nodes / Edges.
Data flow graphs describe and sort out all different type of mathematical computation with a directed graph of nodes & edges. Nodes typically implement mathematical operations, but it can perform several other operations as well. On the other hand, Edges describe the input/output relationships between nodes.
What Nodes can do?
– Implementation of mathematical operations.
– Representation of endpoints to feed in data.
– Pushing out results.
– Read / Write persistent variables.
What Edges can do?
– Representation of flow.
– Input / Output Relationships between nodes.
These data edges carry dynamically-sized multidimensional data arrays known as Tensors. The flow of tensors through the graph is where Tensor Flow gets its name.
Nodes are assigned to computational devices and execute asynchronously and in parallel once all the tensors on their incoming edges becomes available.
- Language Options- Python / C++
- Flexibility in Operations.
- Optimized Performance.
- Bridges Research and Production.
- Auto Differentiation.
How to Get Started?
You don’t need to worry about that. As google provided structured tutorials for Tensor Flow. You can find these tutorials here
Tensor Flow Tutorials
That’s all from Tensor Flow guys. Keep visiting for more tech updates.
Keep Sharing & Tweaking! 💡