Skip to content

In this Repo DataSet, Code and Results have been added. I have created general Neural Networks PArams are taken as input and Different Optimizers and ran on the DataSet to predict Steering angles

Notifications You must be signed in to change notification settings

sid230798/Predict_Steering_Angles_By_Images_Neural_Networks

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

-------------------------------------------------------

Name :- Siddharth Nahar
Entry No :- 2016csb1043
Date :- 30/10/18

------------------------------------------------------

*Copy the Training Set images path as command line argument


------------------------------------------------------
Dir://$ python Main.py PATH_TO_IMAGE_FOLDER

*Code asks for multiple inputs :-

1.Number of Hidden Layers in model
2.Activation function for model as sigmoid,tanh or relu
3.Enter dropouts for input layer and all hidden layers
4.Enter number of nodes in 1st Hidden Layer
5.Enter number of nodes in 2nd Hidden Layer and so on..
6.Batch Size for Gradient Descent
7.Learning Rate 
8.Number of iterations.


Output:-

1. Will give Error on train set and test set after each epoch
2. Plot the graph of validation and train error vs epoch.

-------------------------------------------------------

Sample Run :-

siddharth@Jarvis:~/5thSem/ML/NeuralNetworks$ python Main.py l3/steering/
Started Preprocessing of Data ........
PreProcessing Ended
Number of Hidden Layers in input : 2
Enter the name of Activation Function as sigmoid,tanh, or relu : sigmoid
Enter the Dropout fraction for 1 Layer : 0
Enter the Dropout fraction for 2 Layer : 0
Enter the Dropout fraction for 3 Layer : 0
Enter number of Nodes in 1 Hidden layer : 512
Enter number of Nodes in 2 Hidden layer : 64
Enter the Batch Size for Training : 64
Enter the Learning Rate : 0.05
Enter number of Epochs for training : 100

About

In this Repo DataSet, Code and Results have been added. I have created general Neural Networks PArams are taken as input and Different Optimizers and ran on the DataSet to predict Steering angles

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages