Skip to content
No description, website, or topics provided.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
prescription two
prescription_four
prescription_one
prescription_three
prescription_two
support
tests
.gitignore
.travis.yml
README.md
device_check.py
requirements.txt
version_check.py

README.md

TensorFlow BluePill

Build Status

Setup & initial considerations

In this project, the main objective is to use an implementation style for developers applying eager mode. It is also proposed to use the TensorFlow 2 version, currently in alpha version. For the representation of models, the development is oriented to use Keras to solve the experiences.

Prescription 1

general experiences using TensorFlow concepts.

  • tensors, shape, types and slices.
  • constants, variables
  • arrays and tensors operations

Prescription 2

Experience across a regression model representation, case wine-quality dataset model.

  • extract information
  • normalization
  • build a model
  • train a model
  • evaluate a model
  • implement predictions

Prescription 3

Experience of a categorization model representation, case Iris dataset model.

  • extract information
  • build a model
  • train a model
  • evaluate a model
  • implement predictions

Prescription 4

Using the wine quality dataset of prescription 2, we will re design the experience to represent a categorization model.

  • extract information
  • build a model
  • train a model
  • evaluate a model
  • implement predictions

Prescription 5

into this section we will review the experience to

  • save a model
  • restore a model
  • use TensorFlow serving architecture across docker
  • Serve 1 model developed into experiences 2-3-4.

Anexo, Google Cloud Platform

  • VM with GPU support into Google Cloud Platform.
  • Implementation of matrix multiplication
  • Implementing serving production environment into kubernetes engine

General examples

  • device_check: analyzing performance using a generic operation cpu and gpu device

Official site with details

You can’t perform that action at this time.