refactor: update import paths for forensics utilities and enhance logging functionality in inference data

app.backup.py

@@ -13,10 +13,10 @@ import numpy as np
 import io
 import logging
 from utils.utils import softmax, augment_image, convert_pil_to_bytes
-from utils.gradient import gradient_processing
-from utils.minmax import preprocess as minmax_preprocess
-from utils.ela import genELA as ELA
-from utils.wavelet import wavelet_blocking_noise_estimation
+from forensics.gradient import gradient_processing
+from forensics.minmax import minmax_process
+from forensics.ela import ELA
+from forensics.wavelet import wavelet_blocking_noise_estimation
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
@@ -276,7 +276,7 @@ def predict_image_with_html(img, confidence_threshold, augment_methods, rotate_d
     img_np_og = np.array(img)  # Convert PIL Image to NumPy array
 
     gradient_image = gradient_processing(img_np)  # Added gradient processing
-    minmax_image = minmax_preprocess(img_np)  # Added MinMax processing
+    minmax_image = minmax_process(img_np)  # Added MinMax processing
 
     # First pass - standard analysis
     ela1 = ELA(img_np_og, quality=75, scale=50, contrast=20, linear=False, grayscale=True)

app_mcp.py

@@ -13,17 +13,17 @@ import numpy as np
 import io
 import logging
 from utils.utils import softmax, augment_image, convert_pil_to_bytes
-from utils.gradient import gradient_processing
-from utils.minmax import minmax_process
-from utils.ela import ELA
-from utils.wavelet import wavelet_blocking_noise_estimation
-from utils.bitplane import bit_plane_extractor
+from forensics.gradient import gradient_processing
+from forensics.minmax import minmax_process
+from forensics.ela import ELA
+from forensics.wavelet import wavelet_blocking_noise_estimation
+from forensics.bitplane import bit_plane_extractor
 from utils.hf_logger import log_inference_data
 from utils.text_content import QUICK_INTRO, IMPLEMENTATION
 from agents.ensemble_team import EnsembleMonitorAgent, WeightOptimizationAgent, SystemHealthAgent
 from agents.smart_agents import ContextualIntelligenceAgent, ForensicAnomalyDetectionAgent
 
-from forensics.registry import register_model, MODEL_REGISTRY, ModelEntry
+from utils.registry import register_model, MODEL_REGISTRY, ModelEntry
 from agents.ensemble_weights import ModelWeightManager
 from dotenv import load_dotenv
 import json

app_test.py

@@ -9,15 +9,15 @@ import logging
 # Assuming these are available from your utils and agents directories
 # You might need to adjust paths or copy these functions/classes if they are not directly importable.
 from utils.utils import softmax, augment_image
-from utils.gradient import gradient_processing
-from utils.minmax import minmax_process
-from utils.ela import ELA
-from utils.wavelet import wavelet_blocking_noise_estimation
-from utils.bitplane import bit_plane_extractor
+from forensics.gradient import gradient_processing
+from forensics.minmax import minmax_process
+from forensics.ela import ELA
+from forensics.wavelet import wavelet_blocking_noise_estimation
+from forensics.bitplane import bit_plane_extractor
 from utils.hf_logger import log_inference_data
 from agents.ensemble_team import EnsembleMonitorAgent, WeightOptimizationAgent, SystemHealthAgent
 from agents.smart_agents import ContextualIntelligenceAgent, ForensicAnomalyDetectionAgent
-from forensics.registry import register_model, MODEL_REGISTRY, ModelEntry
+from utils.registry import register_model, MODEL_REGISTRY, ModelEntry
 from agents.ensemble_weights import ModelWeightManager
 from transformers import pipeline, AutoImageProcessor, SwinForImageClassification, Swinv2ForImageClassification, AutoFeatureExtractor, AutoModelForImageClassification
 from torchvision import transforms
@@ -32,7 +32,7 @@ logger = logging.getLogger(__name__)
 os.environ['HF_HUB_CACHE'] = './models'
 
 LOCAL_LOG_DIR = "./hf_inference_logs"
-HF_DATASET_NAME="degentic_rd0"
+HF_DATASET_NAME="aiwithoutborders-xyz/degentic_rd0"
 load_dotenv()
 
 # Custom JSON Encoder to handle numpy types
@@ -166,7 +166,18 @@ register_model_with_metadata(
     display_name="ViT", contributor="temp", model_path=MODEL_PATHS["model_7"]
 )
 
+
 def infer(image: Image.Image, model_id: str, confidence_threshold: float = 0.75) -> dict:
+    """Predict using a specific model.
+
+    Args:
+        image (Image.Image): The input image to classify.
+        model_id (str): The ID of the model to use for classification.
+        confidence_threshold (float, optional): The confidence threshold for classification. Defaults to 0.75.
+
+    Returns:
+        dict: A dictionary containing the model details, classification scores, and label.
+    """
     entry = MODEL_REGISTRY[model_id]
     img = entry.preprocess(image)
     try:
@@ -194,6 +205,22 @@ def infer(image: Image.Image, model_id: str, confidence_threshold: float = 0.75)
         }
 
 def predict_with_ensemble(img, confidence_threshold, augment_methods, rotate_degrees, noise_level, sharpen_strength):
+    """Full ensemble prediction pipeline.
+
+    Args:
+        img (Image.Image): The input image to classify.
+        confidence_threshold (float): The confidence threshold for classification.
+        augment_methods (list): The augmentation methods to apply to the image.
+        rotate_degrees (int): The degrees to rotate the image.
+        noise_level (int): The noise level to add to the image.
+        sharpen_strength (int): The strength of the sharpening to apply to the image.
+
+    Raises:
+        ValueError: If the input image could not be converted to a PIL Image.
+
+    Returns:
+        tuple: A tuple containing the processed image, forensic images, model predictions, raw model results, and consensus.
+    """    
     if not isinstance(img, Image.Image):
         try:
             img = Image.fromarray(img)

forensics/__init__.py

@@ -0,0 +1,15 @@
+from .bitplane import bit_plane_extractor
+from .ela import ELA
+from .exif import exif_full_dump
+from .gradient import gradient_processing
+from .minmax import minmax_process
+from .wavelet import wavelet_blocking_noise_estimation
+
+__all__ = [
+    'bit_plane_extractor',
+    'ELA',
+    'exif_full_dump',
+    'gradient_processing',
+    'minmax_process',
+    'wavelet_blocking_noise_estimation'
+] 

{utils → forensics}/ela.py

@@ -61,4 +61,4 @@ def ELA(img, quality=75, scale=50, contrast=20, linear=False, grayscale=False):
     if grayscale:
         ela = desaturate(ela)
     
-    return Image.fromarray(ela)
+    return Image.fromarray(ela) 

{utils → forensics}/gradient.py

@@ -47,4 +47,4 @@ def gradient_processing(image, intensity=90, blue_mode="Abs", invert=False, equa
         gradient = equalize_img(gradient)
     elif intensity > 0:
         gradient = cv.LUT(gradient, create_lut(intensity, intensity))
-    return Image.fromarray(gradient)
+    return Image.fromarray(gradient) 

{utils → forensics}/minmax.py

@@ -91,4 +91,4 @@ def minmax_process(image, channel=4, radius=2):
         elif channel == 3:
             minmax[low] = [255, 255, 255]
             minmax[high] = [255, 255, 255]
-    return Image.fromarray(minmax)
+    return Image.fromarray(minmax) 

requirements.txt

@@ -1,20 +1,36 @@
-gradio[mcp]
-gradio_leaderboard
-gradio_client
-transformers
-huggingface_hub[hf_xet]
+--index-url https://download.pytorch.org/whl/nightly/cpu
+
+# Core ML/AI libraries
+transformers>=4.48.2
+torch 
 torchvision
-torch
-spaces
-# pillow
+torchaudio
+# Image processing
 opencv-python
-modelscope_studio
-pydantic==2.10.6
-tf-keras
-PyWavelets
-pyexiftool
-psutil
-datasets
 Pillow
+
+# Wavelet processing
+PyWavelets==1.8.0
+
+# System utilities
+psutil
 python-dotenv
-smolagents[all]
+
+# Gradio and UI
+gradio[mcp]>=5.33.1
+# gradio_leaderboard==0.0.13
+gradio_client==1.10.3
+spaces
+
+# HuggingFace ecosystem
+huggingface_hub[hf_xet]>=0.32.0
+datasets>=3.6.0
+
+# Data validation and utilities
+pydantic==2.11.5
+
+# AI agents
+smolagents[all]
+
+# Optional: EXIF metadata (if needed)
+pyexiftool

utils/hf_logger.py

@@ -5,7 +5,7 @@ import io
 import datetime
 from PIL import Image
 import logging
-from huggingface_hub import HfApi, CommitOperationAdd # Keep HfApi for repo creation, but remove CommitOperationAdd for direct upload
+from huggingface_hub import HfApi, CommitScheduler # Keep HfApi for repo creation, but remove CommitOperationAdd for direct upload
 import numpy as np
 
 logger = logging.getLogger(__name__)
@@ -92,6 +92,22 @@ def log_inference_data(
         with open(log_file_path, 'w', encoding='utf-8') as f:
             json.dump(new_entry, f, cls=NumpyEncoder, indent=2)
         
+        # Schedule commit to Hugging Face dataset repository
+        scheduler = CommitScheduler(
+            repo_id=HF_DATASET_NAME,
+            repo_type="dataset",
+            folder_path=LOCAL_LOG_DIR,
+            path_in_repo="logs",
+            token=os.getenv("HF_TOKEN"),
+            every=10  # Commit every 10 files
+        )
+        
+        # Add the file to the scheduler
+        scheduler.push_to_hub(
+            path_or_fileobj=log_file_path,
+            path_in_repo=f"logs/log_{timestamp_str}.json"
+        )
+        
         logger.info(f"Inference data logged successfully to local file: {log_file_path}")
 
     except Exception as e: