Machine Learning – Traditional model building process

Posted on by — Comments
image_pdfimage_print
  1. Identify the ML problem type: Regression, Classification or Time-Series
  2. Specify the data source and format of the labelled training data
  3. Choose your target variable or what you want to predict
  4. Train & Evaluate 100s of Machine Learning algorithms*
  5. Deploy, make predictions & generate insights**

*mltrons Automated Machine Learning curates state-of-the-art libraries like AutoKeras, PyTorch, TensorFlow, H2O, TPOT, Caffe, SageMaker, and AlphaD3M to build the best fit model for the data. Mltrons also provides GPUs to handle larger amount of data.

Gostou? Tire um minutinho e dê sua contribuição para Drall Dev Community no Patreon!

Artigos relacionados:

Default ThumbnailOs 7 passos macros do Machine Learning Default ThumbnailUma idéia sobre Machine Learning Workflow Default ThumbnailMachine Learning com PHP Default ThumbnailUm pequena nota sobre Machine Learning e linguagem de programação Default ThumbnailGoogle Papers – A evolução do TensorFlow visto pelo time interno do Google, os grandes avanços da TPU e a evolução natural do antigo sistema de Machine Learning do Google Default ThumbnailUma pequena nota sobre reconhecimento de imagens do Google e do Facebook Default ThumbnailEdgar F. Codd’s original paper on RDBs – A relational model of data for large shared data banks Default ThumbnailNovo Raspberry Pi 3 Model B+ Default ThumbnailAngularJS – Observando mudanças com $watch no model