Update README.md

README.md

@@ -25,9 +25,9 @@ libraries:
 
 ## Trash Classification CNN Model
 
-### What is this?
+### About
 
-This project is a convolutional neural network (CNN) model developed for the purpose of classifying different types of trash items. The dataset used for training this model is the RealWaste electronic dataset, which is available on The UCI Machine Learning Repository. The dataset is provided by the Wollongong City Council and is licensed under CC BY 4.0.
+This project is a convolutional neural network (CNN) model developed for the purpose of classifying different types of trash items.
 
 The CNN model in this project utilizes the TinyVGG architecture, a compact version of the popular VGG neural network architecture. The model is trained to classify trash items into the following subcategories:
 
@@ -49,7 +49,7 @@ For more details about the CNN architecture used in this project, you can refer
 
 Only 30% of the data from the Real Trash Dataset has been used and divided into an 80%-20% split of Train and Test.
 
-The Repository contains 7 files:
+The Huggingface Repository contains 7 files found in the `files and versions` tab:
 
 1. **data_setup.py**: This file contains functions for setting up the data into datasets using ImageFolder and then turning it into batches using DataLoader. It also returns the names of the classes.
 
@@ -82,4 +82,38 @@ Additionally, the repository contains 2 folders:
 
 ## Model Overview
 
-This model is designed for image classification tasks. It requires input images of size 112x112 pixels.
+This model is designed for image classification tasks. It requires input images of size 112x112 pixels. Containing 2 blocks with 2 convulutional layers and then a flattner with a classfier.
+
+The architecture looks like :
+```python
+TrashClassificationCNNModel(
+  (block_1): Sequential(
+    (0): Conv2d(3, 15, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
+    (1): ReLU()
+    (2): Conv2d(15, 15, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
+    (3): ReLU()
+    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
+  )
+  (block_2): Sequential(
+    (0): Conv2d(15, 15, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
+    (1): ReLU()
+    (2): Conv2d(15, 15, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
+    (3): ReLU()
+    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
+  )
+  (classifier): Sequential(
+    (0): Flatten(start_dim=1, end_dim=-1)
+    (1): Linear(in_features=11760, out_features=9, bias=True)
+  )
+)
+```
+
+## Dataset Overview
+
+The dataset used containes images of multiple waste items with multiple classes named RealWaste. It has 4752 samples.
+
+  - Source: [Click here](https://archive.ics.uci.edu/dataset/908/realwaste)
+  - Citation: Single,Sam, Iranmanesh,Saeid, and Raad,Raad. (2023). RealWaste. UCI Machine Learning Repository. https://doi.org/10.24432/C5SS4G.
+
+## Discliamer
+The model mught give inaccurate or worng results BEWARE.