Upload 13 files

README.md

@@ -0,0 +1,91 @@
+---
+license: apache-2.0
+datasets:
+- prithivMLmods/Deepfake-vs-Real
+language:
+- en
+pipeline_tag: image-classification
+library_name: transformers
+tags:
+- Deepfake
+base_model:
+- google/vit-base-patch32-224-in21k
+---
+![2.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/UJytx7u0VTx_SAz49640L.png)
+
+# **Deepfake-Detection-Exp-02-22**
+
+Deepfake-Detection-Exp-02-22 is a minimalist, high-quality dataset trained on a ViT-based model for image classification, distinguishing between deepfake and real images. The model is based on Google's **`google/vit-base-patch32-224-in21k`**.
+
+```bitex
+Mapping of IDs to Labels: {0: 'Deepfake', 1: 'Real'} 
+
+Mapping of Labels to IDs: {'Deepfake': 0, 'Real': 1}
+```
+
+```python        
+Classification report:
+        
+                      precision    recall  f1-score   support
+        
+            Deepfake     0.9833    0.9187    0.9499      1600
+                Real     0.9238    0.9844    0.9531      1600
+        
+            accuracy                         0.9516      3200
+           macro avg     0.9535    0.9516    0.9515      3200
+        weighted avg     0.9535    0.9516    0.9515      3200
+```
+
+
+![download (1).png](https://cdn-uploads.huggingface.co/production/uploads/6720824b15b6282a2464fc58/-25Oh3wureg_MI4nvjh7w.png)
+
+# **Inference with Hugging Face Pipeline**
+```python
+from transformers import pipeline
+
+# Load the model
+pipe = pipeline('image-classification', model="prithivMLmods/Deepfake-Detection-Exp-02-22", device=0)
+
+# Predict on an image
+result = pipe("path_to_image.jpg")
+print(result)
+```
+
+# **Inference with PyTorch**
+```python
+from transformers import ViTForImageClassification, ViTImageProcessor
+from PIL import Image
+import torch
+
+# Load the model and processor
+model = ViTForImageClassification.from_pretrained("prithivMLmods/Deepfake-Detection-Exp-02-22")
+processor = ViTImageProcessor.from_pretrained("prithivMLmods/Deepfake-Detection-Exp-02-22")
+
+# Load and preprocess the image
+image = Image.open("path_to_image.jpg").convert("RGB")
+inputs = processor(images=image, return_tensors="pt")
+
+# Perform inference
+with torch.no_grad():
+    outputs = model(**inputs)
+    logits = outputs.logits
+    predicted_class = torch.argmax(logits, dim=1).item()
+
+# Map class index to label
+label = model.config.id2label[predicted_class]
+print(f"Predicted Label: {label}")
+```
+
+# **Limitations**  
+1. **Generalization Issues** – The model may not perform well on deepfake images generated by unseen or novel deepfake techniques.  
+2. **Dataset Bias** – The training data might not cover all variations of real and fake images, leading to biased predictions.  
+3. **Resolution Constraints** – Since the model is based on `vit-base-patch32-224-in21k`, it is optimized for 224x224 image resolution, which may limit its effectiveness on high-resolution images.  
+4. **Adversarial Vulnerabilities** – The model may be susceptible to adversarial attacks designed to fool vision transformers.  
+5. **False Positives & False Negatives** – The model may occasionally misclassify real images as deepfake and vice versa, requiring human validation in critical applications.  
+
+# **Intended Use**  
+1. **Deepfake Detection** – Designed for identifying deepfake images in media, social platforms, and forensic analysis.  
+2. **Research & Development** – Useful for researchers studying deepfake detection and improving ViT-based classification models.  
+3. **Content Moderation** – Can be integrated into platforms to detect and flag manipulated images.  
+4. **Security & Forensics** – Assists in cybersecurity applications where verifying the authenticity of images is crucial.  
+5. **Educational Purposes** – Can be used in training AI practitioners and students in the field of computer vision and deepfake detection.

checkpoint-600/config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "google/vit-base-patch32-224-in21k",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Deepfake",
+    "1": "Real"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Deepfake": 0,
+    "Real": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 32,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.1"
+}

checkpoint-600/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73a60bf9aec06941615d1a2b697736d7fb7ad07c150cf434dfbc09550328e16a
+size 349850288

checkpoint-600/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52cc7200995d9b23db3901761e5ddbea13a0960a155caef18f7cf98315584581
+size 699821498

checkpoint-600/preprocessor_config.json

@@ -0,0 +1,23 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

checkpoint-600/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58d3089cecf8afe9057a8e31154c498a9495581a3deafc501f8af896fb94a8dc
+size 14244

checkpoint-600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52486668edafef828135459be62f8fb60a1c8ac03ea0cc68c8e5fca33899a5e8
+size 1064

checkpoint-600/trainer_state.json

@@ -0,0 +1,80 @@
+{
+  "best_metric": 0.5417714715003967,
+  "best_model_checkpoint": "DP32/checkpoint-600",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.841875,
+      "eval_loss": 0.6350374221801758,
+      "eval_model_preparation_time": 0.0045,
+      "eval_runtime": 64.618,
+      "eval_samples_per_second": 49.522,
+      "eval_steps_per_second": 6.19,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.925,
+      "eval_loss": 0.5866522192955017,
+      "eval_model_preparation_time": 0.0045,
+      "eval_runtime": 62.9502,
+      "eval_samples_per_second": 50.834,
+      "eval_steps_per_second": 6.354,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.94625,
+      "eval_loss": 0.5532988905906677,
+      "eval_model_preparation_time": 0.0045,
+      "eval_runtime": 63.0217,
+      "eval_samples_per_second": 50.776,
+      "eval_steps_per_second": 6.347,
+      "step": 450
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 2.095505475997925,
+      "learning_rate": 1.818181818181818e-07,
+      "loss": 0.5872,
+      "step": 500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9515625,
+      "eval_loss": 0.5417714715003967,
+      "eval_model_preparation_time": 0.0045,
+      "eval_runtime": 62.7241,
+      "eval_samples_per_second": 51.017,
+      "eval_steps_per_second": 6.377,
+      "step": 600
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 600,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.516572596109312e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8b821a7a428ff1821b122e8796fde5b3b9a37a81fff0d4beb6b58653fc2d842
+size 5240

config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "google/vit-base-patch32-224-in21k",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Deepfake",
+    "1": "Real"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Deepfake": 0,
+    "Real": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 32,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73a60bf9aec06941615d1a2b697736d7fb7ad07c150cf434dfbc09550328e16a
+size 349850288

preprocessor_config.json

@@ -0,0 +1,23 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8b821a7a428ff1821b122e8796fde5b3b9a37a81fff0d4beb6b58653fc2d842
+size 5240