Merge pull request #882 from ojaswichopra/playing_cards

Playing cards Image Classification

Playing Cards Image Classification/Dataset/README.md

@@ -0,0 +1 @@
+You can download the dataset from [Kaggle](https://www.kaggle.com/datasets/gpiosenka/cards-image-datasetclassification).

Playing Cards Image Classification/Model/README.md

@@ -0,0 +1,63 @@
+## **Playing Cards Image Classification**
+
+### 🎯 **Goal**
+
+The main goal of this project is to classify different playing cards using deep learning models. This involves building and evaluating models to identify various card types from images with high accuracy.
+
+### 🧵 **Dataset**
+
+The dataset used for this project is sourced from [Kaggle](https://www.kaggle.com/datasets/gpiosenka/cards-image-datasetclassification). It consists of images of playing cards with labels corresponding to different card types. The dataset includes training, validation, and test splits.
+
+### 🧾 **Description**
+
+This project aims to implement and compare several deep learning models to classify playing cards. It involves preprocessing image data, applying data augmentation techniques, training models such as basic CNNs, AlexNet and EfficientNet and evaluating their performance using accuracy metrics.
+
+### 🧮 **What I have done!**
+
+- Preprocessed the image data by resizing and normalizing images.
+- Applied data augmentation techniques to enhance dataset diversity.
+- Implemented three models:
+  - **Basic CNN**: A simple convolutional neural network with custom architecture.
+  - **AlexNet**: A deep convolutional network known for its success in image classification tasks.
+  - **EfficientNet**: A family of convolutional neural networks (CNNs) designed for image classification tasks.
+- Conducted exploratory data analysis (EDA) to understand the dataset's characteristics.
+- Evaluated model performance using accuracy scores.
+
+### 🚀 **Models Implemented**
+
+- **Basic CNN**: A custom convolutional neural network designed for image classification tasks.
+- **AlexNet**: A well-known deep learning model with a AlexNet architecture adapted for playing card classification.
+- **EfficientNetB0**: Base model of EfficientNet known for its efficiency in terms of both computational resources and accuracy. 
+
+### 📚 **Libraries Needed**
+
+- numpy==1.21.2
+- pandas==1.3.3
+- tensorflow==2.7.0
+- matplotlib==3.4.3
+- seaborn==0.11.2
+- scikit-learn==0.24.2
+
+### 📊 **Exploratory Data Analysis Results**
+
+![Dataset Visualization](<../Images/Dataset Visualization.png>)
+![Distribution of Card Types](<../Images/Distribution of Card Types.png>)
+![Distribution of Dataset Splits](<../Images/Distribution of Dataset Splits.png>)
+![Distribution of Labels](<../Images/Distribution of Labels.png>)
+
+### 📈 **Performance of the Models based on Accuracy Scores**
+
+- **Basic CNN**: Accuracy - 74%  
+- **AlexNet**: Accuracy - 71%  
+- **EfficientNet**: Accuracy - 95%  
+
+### 📢 **Conclusion**
+
+Based on the evaluation results, all of the models performed fairly well. Efficient net outperformed CNN and AlexNet with an accuracy of 95%. Further fine-tuning and experimentation with other models or techniques may provide additional insights.
+
+### ✒️ **Your Signature**
+
+Ojaswi Chopra
+---
+
+Connect with me on [LinkedIn](https://www.linkedin.com/ojaswichopra) | [GitHub](https://github.com/ojaswichopra)

Playing Cards Image Classification/Model/requirements.txt

@@ -0,0 +1,7 @@
+numpy==1.21.2
+pandas==1.3.3
+Pillow==8.4.0
+scikit-learn==0.24.2
+matplotlib==3.4.3
+seaborn==0.11.2
+tensorflow==2.7.0