metadata
license: apache-2.0
datasets:
- microsoft/cats_vs_dogs
language:
- en
metrics:
- accuracy
- f1
- recall
- precision
pipeline_tag: image-classification
language: en tags:
- pytorch
- image-classification
- cats-vs-dogs
- computer-vision datasets:
- microsoft/cats_vs_dogs model-index:
- name: Dogs vs Cats Classifier results:
- task:
type: image-classification
name: Image Classification
metrics:
- type: accuracy value: 93.25 name: Validation Accuracy
- type: roc_auc value: 0.9942 name: ROC AUC
- type: precision value: 0.9769 name: Precision
- type: recall value: 0.9615 name: Recall
- type: f1 value: 0.9691 name: F1-Score
license: mit
Dogs vs Cats Classifier
This model classifies images as either cats or dogs using a Convolutional Neural Network (CNN) architecture.
Model description
Architecture:
- 4 convolutional blocks (Conv2D → ReLU → BatchNorm → MaxPool)
- Feature channels: 3→64→128→256→512
- Global average pooling
- Fully connected layers: 512→256→1
- Binary classification output
Training
- Dataset: microsoft/cats_vs_dogs
- Training/Validation split: 80/20
- Input size: 224x224 RGB images
- Trained for 10 epochs
- Best validation accuracy: 93.25%
Intended uses
- Image classification between cats and dogs
- Transfer learning base for similar pet/animal classification tasks
Limitations
- Only trained on cats and dogs
- May not perform well on:
- Low quality/blurry images
- Unusual angles/poses
- Multiple animals in one image
Input
RGB images resized to 224x224 pixels, normalized using ImageNet statistics:
- mean=[0.485, 0.456, 0.406]
- std=[0.229, 0.224, 0.225]
khouya abd elhamid rak bacha bla mona9acha