TensorFlow
TensorFlow is an endtoend open source machine learning platform, developed by Google Brain and first made public in 2015. It is very famous for its deep learning functionalities. Its basic data structure is multidimensional array, called tensor. The computations are expressed as dataflow graphs, where each node is a mathematical operation and the connecting arrows are tensors.
Its ecosystem is extremely rich, offering many advanced functionalities to developers and researchers, such as:

plethora of libraries, ranging from probabilistic reasoning to music creation

tools improving development, debugging, compiling and benchmarking

officially supported models and datasets
It can run on CPU, GPU and TPU (tensor processing unit). APIs for 13 different programming languages are available, Python being the most complete and the only one covered by the API stability promises.
Related articles
TensorFlow Extended (TFX): the components and their functionalities
Categories: Big Data, Data Engineering, Data Science, Learning  Tags: Beam, CI/CD, Data Engineering, Deep Learning, Pipeline, Data Science, Deployment, Machine Learning, MLOps, Open source, Python, TensorFlow
Putting Machine Learning (ML) and Deep Learning (DL) models in production certainly is a difficult task. It has been recognized as more failureprone and time consuming than the modeling itself, yet…
Mar 3, 2021
MLflow tutorial: an open source Machine Learning (ML) platform
Categories: Data Engineering, Data Science, Learning  Tags: Deep Learning, AWS, Databricks, Deployment, Machine Learning, Azure, MLflow, MLOps, Python, Scikitlearn
Introduction and principles of MLflow With increasingly cheaper computing power and storage and at the same time increasing data collection in all walks of life, many companies integrated Data Science…
Mar 23, 2020
Avoid Bottlenecks in distributed Deep Learning pipelines with Horovod
Categories: Data Science  Tags: Deep Learning, GPU, Horovod, Keras, TensorFlow
The Deep Learning training process can be greatly speed up using a cluster of GPUs. When dealing with huge amounts of data, distributed computing quickly becomes a challenge. A common obstacle which…
By Grégor JOUET
Nov 15, 2019
Machine Learning model deployment
Categories: Big Data, Data Engineering, Data Science, DevOps & SRE  Tags: DevOps, Operation, AI, Cloud, Machine Learning, MLOps, Onpremises, Schema
“Enterprise Machine Learning requires looking at the big picture … from a data engineering and a data platform perspective,” lectured Justin Norman during the talk on the deployment of Machine…
Sep 30, 2019
TensorFlow installation on Docker
Categories: Containers Orchestration, Data Science, Learning  Tags: CPU, Deep Learning, Docker, Jupyter, Linux, AI, TensorFlow
TensorFlow is an Open Source software from Google for numerical computation using a graph representation: Vertex (nodes) represent mathematical operations Edges represent Ndimensional data array…
Aug 5, 2019
Nvidia and AI on the edge
Categories: Data Science  Tags: Caffe, Deep Learning, Edge computing, GPU, NVIDIA, AI, Keras, PyTorch, TensorFlow
In the last four years, corporations have been investing a lot in AI and particularly in Deep Learning and Edge Computing. While the theory has taken huge steps forward and new algorithms are invented…
By Yliess HATI
Oct 10, 2018
Deep learning on YARN: running Tensorflow and friends on Hadoop cluster
Categories: Data Science  Tags: Spark, YARN, Deep Learning, GPU, Hadoop, Spark MLlib, PyTorch, TensorFlow, XGBoost, MXNet
With the arrival of Hadoop 3, YARN offer more flexibility in resource management. It is now possible to perform Deep Learning analysis on GPUs with specific development environments, leveraging…
Jul 24, 2018
YARN and GPU Distribution for Machine Learning
Categories: Data Science, DataWorks Summit 2018  Tags: YARN, GPU, Machine Learning, Neural Network, Storage
This article goes over the fundamental principles of Machine Learning and what tools are currently used to run machine learning algorithms. We will then see how a resource manager such as YARN can be…
By Grégor JOUET
May 30, 2018
TensorFlow on Spark 2.3: The Best of Both Worlds
Categories: Data Science, DataWorks Summit 2018  Tags: Mesos, Spark, YARN, C++, CPU, GPU, JavaScript, Tuning, Keras, Kubernetes, Machine Learning, Python, TensorFlow
The integration of TensorFlow With Spark has a lot of potential and creates new opportunities. This article is based on a conference seen at the DataWorks Summit 2018 in Berlin. It was about the new…
By Yliess HATI
May 29, 2018