https://github.com/shibam120302/gsoc23_ml4sc_lens_finding

https://github.com/shibam120302/gsoc23_ml4sc_lens_finding

Last synced: 7 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/shibam120302/gsoc23_ml4sc_lens_finding
• Owner: shibam120302
• Created: 2023-04-01T15:41:26.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2023-04-01T15:50:05.000Z (over 2 years ago)
• Last Synced: 2025-01-21T17:50:42.867Z (9 months ago)
• Language: Jupyter Notebook
• Size: 13.2 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# ML4SC_GSoc22

### GSOC22 evaluation results

# **Finding** **Rings**

## Datset Used:

#### I used the dataset provided by deepmind on their test document.It contains two types of images one with Rings and one with no rings.I divided the data into two categories Test and Train by 10:90.Here is the image of data.

![](images/batch.png)

## Model:

#### I created my own deep RESNET model with 10+ layers of CNN’s and 2 Layers of Resnets.

![](/images/model.png)

## Evaluation:

#### I trained the data with the batch size of 32 for 20 to 330 epochs and results were awesome.I achieved an accuracy of 89.44% after 20 epochs.

#### Here you can see the training process of model.

![](images/res12_loss.png)

![](images/res12_accuracy.png)

![](images/res12_lr.png)

### Evaluate you own validation set by IPYNB notebook.