You are on page 1of 4

Name : Sakariya Gaurav Dilipbhai

Roll No : 19BCE233
Subject : Artificial Intelligence (2CSDE85)
Date : 08/09/2022

Practical: 1

Aim: Explore AI Tools.

Tensorflow

 TensorFlow is an end-to-end open-source platform for machine


learning. TensorFlow is a rich system for managing all aspects of a
machine learning system; however, this class focuses on using a
particular TensorFlow API to develop and train machine learning
models. See the TensorFlow documentation for complete details
on the broader TensorFlow system.

 Advantages:
(a) It is an open-source platform that makes it available to all the
users around and ready for the development of any system on
it.
(b) TensorFlow also has its architecture TPU, which performs
computations faster than GPU and CPU. Models built using
TPU can be easily deployed on a cloud at a cheaper rate and
executed at a faster rate.
(c) Tensorflow's Tensorboard makes it simple to debugging nodes.
By doing this, the overhead of accessing the entire code is
decreased.
 Disadvantages:
1. slow speed with respect to its competitors
2. It’s faster TPU only allows the execution of a model but
they can’t trainthem.
3. It Only supports NVIDIA GPU and the Python
programming language forGPU programming.

Pytorch

 PyTorch was created utilising the Torch library for Python programmes.
It was createdby Facebook's AI Research lab and made available as a free
and open-source library inJanuary 2016. It is mostly used in applications
for computer vision, deep learning, andnatural language processing. Due
to PyTorch's flexibility, speed, and ease of use, it is becoming more and
more popular in both the business world and among researchers. PyTorch
is one of the best deep learning tools.

 Advantages:
1. It is flexible, faster, and provides optimizations.
2. Python's IDE and debugging tools make debugging simple.
3. strong APIs rich set to expand the Pytorch Libraries.

 Disadvantages:
1. It is new compared to other AI tools because it was only released
in 2016, so ithas fewer users, and is not well-known.
2. lack of visualisation and monitoring tools, such as a tensor
board which isavailable in tensorflow.

Keras
 Keras is quite helpful for those just getting started with neural networks,
We can quickly construct a neural network model using this high-level
framework, which hides the backend processing. it is based on minimal
structure that provides a clean and easy way to create deep learning models
based on TensorFlow or Theano. Keras is designed to quickly define deep
learning models.

 Advantages:
1. Keras is a user-friendly API and it is very easy to create neural
network models with Keras.
2. Keras provides some deep learning models with their pre-
trained weights. We can use these models directly for making
predictions or feature extraction.

 Disadvantages:
1. Sometimes it is slow on GPU and takes longer time in
computation compared with its backends.
2. Error logs are difficult to debug.

Caffe

 Caffe Convolutional Architecture for Fast Feature Embedding) is the


perfect framework for image classification and segmentation as it
supports various GPU- and CPU-based libraries such as NVIDIA,
cuDNN, Intel MKL, etc. Caffe is aimed at the production of edge
deployment. Caffe interface is somewhat like C++, which means users
need to perform more tasks manually, such as configuration file creation.

 Advantages:
1. Caffe offers very limited abstraction, which means it's
extremely easy to dounconventional, hard-core modifications.
2. It Supports GPU training out-of-the-box.
3. It has Pretty good Matlab and Python interface.
 Disadvantages:
1. Multi-GPU training is partially supported but not all the different
ways of doingthat such as model/data parallelism etc.
2. Dynamic input is error prone in large development programs.

Scikit Learn

 Sklearn (Scikit-learn) is one of the most useful open-source libraries


available for machine learning in Python. The scikit-learn library is a
comprehensive collection of the most efficient statistical modeling and
machine learning tools. These tools include regression, classification,
dimensionality reduction, clustering, and more.

 Advantages:
1. The scikit-learn library is extremely versatile and useful for real-
world purposessuch as predicting consumer behavior, creating
neuroimaging, and more.
2. scikit-learn is supported and updated by numerous authors,
contributors, and alarge international online community.
3. The scikit-learn website offers extensive API documentation for
users who wantto integrate their algorithms into their platform.

 Disadvantages:
1. It is not the best choice for deep learning.

You might also like