Scribd
Upload
62 views

Mid Term

The 9 steps for developing a neural network model are: 1) Initialization, 2) Data generation, 3) Train-test splitting, 4) Neural net construction, 5) Propagation, 6) Back-propagation, 7) Iterative optimization, and 8) Testing. The key steps involve generating unique data, splitting it for training and testing, constructing and connecting the neural network layers, propagating data through the network, and iteratively optimizing and testing the model.

Uploaded by

aks
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
62 views

Mid Term

The 9 steps for developing a neural network model are: 1) Initialization, 2) Data generation, 3) Train-test splitting, 4) Neural net construction, 5) Propagation, 6) Back-propagation, 7) Iterative optimization, and 8) Testing. The key steps involve generating unique data, splitting it for training and testing, constructing and connecting the neural network layers, propagating data through the network, and iteratively optimizing and testing the model.

Uploaded by

aks
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Midterm 1

Answer 2

The 9 steps are

Initialization- The first step is to take each step during the network designing process seriously

Data generation-The next step is to make sure that the data developed is unique and generated

without any external sources. The huge collective data will be used to make assumptions using

the generation method.

Train-test Splitting- the datasets will be distributed into two sections so that it could be further

processed for training and testing process. The sections will be 30% testing and 70% training.

The evaluation of the data performance will be the next step and training and testing will be

done.

Neural net Construction- the programming step of class python will be utilized to develop a

layer. The list will be prepared that describes each layer. The applicable formulas will be used to

take care of hyper parameters.

Propagation- the connection needs to be created and that will be done using the biases and data

weights.

Back- Propagation- once the connection made the data sensitivity needs to be maintained and

that will be done through coding.

Iterative Optimization – regular updates will be planned to reduce the data sensitivity

Testing – once all the processes are done the design of the model will be tested to identify

vulnerabilities.

Answer 3
Midterm 2

Due to the similar processes, it is very difficult to find a difference between the two that is

clustering and classification. There are very small differences between the two and the same will

be discussed here.

Classification- the input instance in classification is assigned with labels those are predetermined

on the basis of their properties.

Clustering- The clustering process do not have any labels

The differences are:-

1. Classification is a process for supervised learning, and clustering is of unsupervised

learning

2. Corresponding class labels are used in the process of classification whereas, clustering do

not use such labels.

3. Use of labels requires training and testing process for classification process whereas,

clustering does not require and testing or training.

4. Classification is more complex as compared to clustering.

5. Examples of classification are Logistic regression and support vector machine. The

examples for clustering are Fuzzy c-means, K-means, Gaussian clustering alogorithm.

References

Kartoun, U. (2019). A glimpse of the difference between predictive modeling and classification

modeling. Journal Of Clinical Epidemiology, 109, 142. doi:

10.1016/j.jclinepi.2019.01.001

You might also like