README.md

GRE admission prediction

This will help students in predicting their chances for admission in any particular university/college

SYSTEM REQUIREMENTS:

HARDWARE:-

 TPU:-  The TPU is a 28nm, 700MHz ASIC that fits into SATA hard disk slot and is connected to its host via a PCIe Gen3X16 bus that provides an effective bandwidth of 12.5GB/s.

 GPU:- Nvidia GTX 1080 (8 GB VRAM) RECOMMENDED

HOW TO USE:-

1. IMPORT TENSORFLOW IN KAGGLE
2. USED THE KERAS LIBRARY FOR MANIPULATING DATA
3. USED THE GRADUATE ADMISSION DATA(REGRESSION DATA) FROM KAGGLE
4. BY USING THE PANDAS LIBRARY WE CREATED THE DATA FRAME
5. CLEANING THE DATA
6. DIVIDED THE DATA IN TWO PARTS 20% FOR TESTING AND 80% FOR TRAINING
7. SPLIT THE DATA INTO TWO PARTS INPUT AND OUTPUT
8. SCALING THE DATA
9. DESIGNED THE ARTIFICIAL NEURAL NETWORK(ANN) WITH 3 LAYERS INPUT LAYER HIDDEN LAYER OUTPUT LAYER
10. WE USED loss='mean_squared_error',optimizer='Adam'
11. WE TRAINED THE DATA (SPLIT THE DATA BY 20% FOR VALIDATION WHILE TRAINING)
12. THROUGH A GRAPH WE CHECKED FOR THE OVERFITTING
13. WE PREDICTED THE OUTPUT WITH THE TEST DATA AND WE ACHIEVED THE ACCURACY OF ~82%