Quick Contact

    Python Tutorial
    Python Panda Tutorial
    Python Selenium
    Python Flask Tutorial
    Python Django
    Numpy
    Tensorflow
    Interview Questions & Answers

    PROGRAM ELEMENTS IN TENSORFLOW

    Program elements in tensorflow include constant, variable, placeholder, sparse tensor.

    – If you want to use tensorflow then you have to use tensors. So by using different program elements you can create different tensors according to requirement.

      1. Constant – constant tensor creates- from a tensor-like object.

    For this you can use syntax – Syntax.tf.constant()

    In this once you create the tensor then you can’t change it.

      1. Variable – you can add new trainable parameters using variables to the graph.

    Syntax is – tf.variable ()

    In this you can make changes in the tensor in future.

      1. Placeholders- with placeholders you can input data to a tensorflow model – from the outside model.

    Syntax is as-: tf.placeholder()

    So when you want to create tensor then no need to put the any value, you just have to mention the data type. So in future whenever you want to add the value in tensor then you can do it using placeholder ().

      1. Sparse Tensor- It allows or enables the processing of tensors as well as efficient storage.

    It includes large number of zero values.

    Syntax- tf.SparseTensor()

    18) Tensorflow 1.0 v/s 2.0

    Tensorflow is no longer what it used to b.

    – Tensorflow 1.O is one of the most widely used deep learning packages, it is versatile. Unfortunately it has major drawback, it is very hard to learn and use, and that’s why many people become dishearted after seeing couple of lines of tensorflow code. Not only its methods are strange but the whole logic of coding is unlike most libraries out there.

    – This led to the development and popularization of higher level packages such as PyTorch and Keras.

    – Keras is interesting as in 2017; it was integrated in the core Tensorflow. In reality though, both Tensorflow and Keras are open source, so such things do happen in programming world.

    – Keras is conceived as an interface for tensorflow rather than different library, making this integration even easier to digest and implement.

    – Even with Keras as part of TF, tensorflow still losing popularity.

    IT was addressed in 2019, when Tensorflow 2.O came on horizon or at least its alpha version.

    – Instead of creating their own high level syntax, the TF developers chose to borrow that of Keras.

    – This decision made sense of Keras was widely adeptly already and people love it.

    – You may hear that “Tensorflow 2 is basically Keras”.

    – TF2 has best of both worlds- most of the versatility of TF1 and the high-level simplicity of Keras.

    – There are also two other major advantages of TF 2 over TF1 – they simplified the API, remove the duplicity and deprecated functions and added some new to the core Tensor Flow.

    – Tensor Flow 2 boasts eager execution or in other words- allowing standard python “rules of physics” to apply to it, rather than complex computational graphs.

     

    Apply now for Advanced Python Training Course

    Copyright 1999- Ducat Creative, All rights reserved.

    Anda bisa mendapatkan server slot online resmi dan terpercaya tentu saja di sini. Sebagai salah satu provider yang menyediakan banyak pilihan permainan.