Machine Learning | Deep Learning | Data Science | Web Dev

When it comes to predictive modeling , A single algorithmic model might be not be enough to make the most optimal predictions.

One of the most effective methodologies in machine learning is **Ensemble Modeling** or **Ensembles**.

Ensemble Modeling is the combination of multiple machine learning models that follow the same or different algorithm to make better overall predictions.

It is usually these types of models that win the machine learning competitions conducted by *netflix* or *kaggle*.

Ensemble modeling methods can be split into various categories.

they are:

**Sequential**

**Probability Distribution :** A probability Distribution shows the list of probabilities associated with each value or a range of values for a discrete or a continuous random variable.

Based on the Type of Random Variable, Probability Distributions can be split into **Continuous Probability Distributions** and **Discrete Probability Distributions**

Convolutional Neural Networks is a type of neural network that is mostly used for Image Datasets and The shared parameters feature of convolutional network reduces the number of parameters for the model and this also makes it efficient in pattern **Edge-Detection. **This blog post contains the implementation of Convolutional Neural Networks with tensorflow 2.0 on the Cats vs Dogs dataset

Cats vs Dogs dataset is one of the most famous dataset used in the field of machine learning, which was developed by the partnership between **pathfinder.com** and **Microsoft.**

*The Dataset is available at :** **https://www.microsoft.com/en-us/download/details.aspx?id=54765*

*Note: The code files will be available at : **https://github.com/ashwinhprasad/Tensorflow-2.0/blob/master/TF2-(4)-ANN/TF2-ANNs.ipynb*

Artificial Neural Networks are the traditional neural networks which means that there are more than or equal to one layer between the input and the output layer. This allows for the model to adapt for the non linearity and form complex functions which makes it useful in real life.

This blog post does not covers only the implementation of feed forward or artificial neural network with tensorflow 2 and not the theory part of the artificial neural network.

*Note : The Program files for tensorflow 2 can be found on - **https://github.com/ashwinhprasad/Tensorflow-2.0*

Logistic Regression is used for Classification tasks and This Blog will take you through the implementation of logistic regression using Tensorflow 2. This blog post won’t be covering about the theories regarding logistic regression and theory is a pre-requisite.

The Dataset that is used in this example is iris dataset from the sklearn library.

we are importing the dataset and storing it in the form of a pandas dataframe

`#importing the libraries import numpy as np import tensorflow as tf import pandas as pd import matplotlib.pyplot…`

*Code files will be available at : **https://github.com/ashwinhprasad/Tensorflow-2.0*

Linear regression is basically using a equation of a line to find out the linear relationship between 2 variables. By finding the linear relationship, I mean finding the parameters ( Slope and Intercept).

**y = m*x + cy : dependent variablex : independent variablem : slopec : intercept**

*Note : All code files will be available at **https://github.com/ashwinhprasad/Tensorflow-2.0*This blog post will cover some basic functions that will be repeatedly used a lot in tensorflow 2.

random.normal generates random values of the given shape, which follow normal distribution

and random.uniform generates random values in such a way that probability of choosing any number from the random bunch is almost uniform

#normal distribution

x1 = tf.random.normal(shape=(5,5),mean=0,stddev=1)

#normal distribution

print(x1)output:tf.Tensor( [[-1.1473149e+00 5.1616412e-01 -2.8656033e-01 -1.4161720e-03 -6.7782238e-02] [ 1.5549400e-01 -1.8609362e+00 7.8299832e-01 -7.3712116e-01 -3.0330741e-01] [ 5.6524660e-02 1.0138390e-01 1.2218195e+00 1.2505690e+00 3.0457941e-01] [ 3.6436683e-01 -8.6699528e-01 1.5152076e+00 7.8330201e-01 -1.4127023e+00] [-1.2999429e+00 1.3505920e+00 1.0376108e+00 -1.5029492e+00 9.7778231e-01]]…

Tensorflow is a DeepLearning library which has a lot of inbuilt classes and functions which allow you to perform these complex deep learning matrix multiplications and gradient calculations easily. The main Goal behind tensorflow is to make developing machine learning models easier and get it to a production environment.

As everyone already know, The updated version of tensorflow allows the user to create models easily whereas it was quite difficult with tensorflow’s first version.

You could directly use tensorflow from google colab (I prefer this) or type

“pip install tensorflow” for windows users and

“pip3 install tensorflow2” for linux users

…

sorry for misspelling network , lol.

All the code files will be available at : https://github.com/ashwinhprasad/PyTorch-For-DeepLearning

Recurrence Neural Network are great for Sequence data and Time Series Data. Long short-term memory is an artificial recurrent neural network architecture used in the field of deep learning. LSTMs and RNNs are used for sequence data and can perform better for timeseries problems.

An LSTM is an advanced version of RNN and LSTM can remember things learnt earlier in the sequence using gates added to a regular RNN. Both LSTM’s and RNN’s working are similar in PyTorch. So, once we coded the Lstm…

Note : All the code files will be available at : https://github.com/ashwinhprasad/SentimentAnalysis

Sentiment analysis in simple words is basically analysing how an user feels about an item or any other thing from the user’s activity such as reviews , tweets, etc.

**Importing the Libraries**

**import** **numpy** **as** **np**

**import** **pandas** **as** **pd**

**import** **matplotlib.pyplot** **as** **plt**

import nltk

**2. Downloading NLTK**

Use the nltk shell to download the english stopwords.

**3. Importing the dataset**

`df = pd.read_csv('IMDB Dataset.csv')`

df.head()

**output :**