license: mit
language:
- en
metrics:
- accuracy
pipeline_tag: image-classification
library_name: tf-keras
tags:
- legal
- code
- not-for-all-audiences
Model Card for Real vs. Fake Image Classifier
This model is designed to classify images as either "real" or "fake" using a Convolutional Neural Network (CNN) built with TensorFlow Keras.
Model Details
Model Description
This CNN model has been developed to differentiate between real and fake images. It utilizes various convolutional layers, pooling layers, and dense layers to effectively learn features from the input images, enabling accurate classification.
- Developed by: [Abhishek Thakur, Mihir Vaid]
- Model type: Image Classification (CNN)
- License: MIT
Uses
Direct Use
This model can be used directly for image classification tasks without the need for further fine-tuning. Users can input images, and the model will output a classification of "real" or "fake."
Out-of-Scope Use
This model is not intended for use in critical applications where misclassification could lead to significant consequences, such as security or legal decisions.
Bias, Risks, and Limitations
While the model aims to provide accurate classifications, it may exhibit biases based on the training data. Users should be aware of potential limitations in performance across different types of images or contexts.
Recommendations
Users should validate the model's predictions with additional methods when deploying it in sensitive applications.
How to Get Started with the Model
To use this model, you can load it using TensorFlow Keras as shown below:
import tensorflow as tf
# Load your trained model
model = tf.keras.models.load_model('path/to/your/model')
# Example prediction
image = tf.keras.preprocessing.image.load_img('path/to/image.jpg', target_size=(image_height, image_width))
image_array = tf.keras.preprocessing.image.img_to_array(image)
image_array = tf.expand_dims(image_array, axis=0)
predictions = model.predict(image_array)
print("Predicted class:", "Real" if predictions > 0.5 else "Fake")