Upload folder using huggingface_hub

README.md

@@ -13,6 +13,12 @@ metrics:
 - accuracy
 - f1
 pipeline_tag: image-classification
+inference: false
+widgets:
+  - text: "Test the DeepGuard Model Live"
+    src: "https://harshasnade-deepfake-detection-system-v1.hf.space"
+  - text: "Deepfake Sample"
+    src: "https://harshasnade-deepfake-detection-system-v1.hf.space"
 ---
 
 # DeepGuard - Deepfake Detection System
@@ -55,8 +61,6 @@ The model is designed to classify single images as either **REAL** or **FAKE**.
 import torch
 import torch.nn as nn
 from torchvision import models
-import albumentations as A
-from albumentations.pytorch import ToTensorV2
 from safetensors.torch import load_file
 import cv2
 
@@ -101,19 +105,6 @@ The model was trained on a diverse dataset comprising:
 - **Deepfake Images:** Generated using StyleGAN2, Diffusion Models, and FaceSwap techniques.
 - **Data Augmentation:** extensive augmentation (compression, noise, blur) was applied to robustify the model against social media re-compression artifacts.
 
-### Training Procedure
-
-- **Optimizer:** AdamW
-- **Loss Function:** BCEWithLogitsLoss
-- **Scheduler:** OneCycleLR
-- **Epochs:** 10+ with Early Stopping
-- **Input Resolution:** 224x224
-
-#### Training Hyperparameters
-
-- **Batch Size:** 32
-- **Precision:** Mixed Precision (FP16)
-
 ## Evaluation
 
 ### Results
@@ -122,19 +113,6 @@ The model achieves high accuracy on standard benchmarks:
 - **Test Accuracy:** ~92-95% (on unseen test split)
 - **Generalization:** Shows strong resilience to JPEG compression compared to standard CNNs.
 
-## Technical Specifications
-
-### Model Architecture
-
-The specific ensemble combines:
-1.  **EfficientNetV2-S:** Excellent at capturing sharp, high-frequency details (e.g., hair textures, eye reflections).
-2.  **Swin Transformer (V2-T):** Captures global semantic inconsistencies (e.g., facial structural alignment).
-
-### Compute Infrastructure
-
-- **Hardware:** Trained on Mac M-Series (MPS) / NVIDIA GPUs.
-- **Framework:** PyTorch 2.6+
-
 ## Citation
 
 ```bibtex