Upload folder using huggingface_hub

README.md

@@ -1,68 +1,34 @@
----
-language: en
-license: mit
-tags:
-  - vision
-  - image-classification
-  - drowsiness-detection
-  - driver-monitoring
-  - pytorch
-  - mobilenetv2
-datasets:
-  - custom
----
-
 # Driver Drowsiness Detection Model
 
-This model detects whether a driver is alert or showing signs of drowsiness based on facial features.
-
-## Model Description
-
-This model is fine-tuned on a custom dataset of driver face images to detect drowsiness in real-time. It uses a MobileNetV2 architecture for efficient inference.
-
-### Model Architecture
-- Base model: MobileNetV2
-- Fine-tuned for binary classification (alert vs. drowsy)
-- Input size: 224x224 RGB images
-
-## Training Data
-
-The model was trained on custom-labeled face images extracted from driver-facing camera videos. The dataset includes various lighting conditions and driver appearances.
-
-## Performance
-
-The model can identify signs of drowsiness in drivers with high accuracy. It's designed to be fast enough for real-time inference.
+This model is designed to detect driver drowsiness from facial images using a CNN architecture.
 
-## Limitations
-
-- Requires good lighting conditions to detect faces properly
-- May have reduced accuracy with glasses or face coverings
-- Should be used as part of a comprehensive driver monitoring system and not as the sole safety mechanism
+## Model Details
+- Architecture: CNN
+- Input: Facial images (64x64x3)
+- Output: Binary classification (drowsy/not drowsy)
 
 ## Usage
-
 ```python
-from transformers import AutoImageProcessor, AutoModelForImageClassification
-from PIL import Image
-
-# Load model and processor
-processor = AutoImageProcessor.from_pretrained("ckcl/driver-drowsiness-detector")
-model = AutoModelForImageClassification.from_pretrained("ckcl/driver-drowsiness-detector")
-
-# Load an image
-image = Image.open("path/to/driver/face/image.jpg")
-
-# Process image and get prediction
-inputs = processor(images=image, return_tensors="pt")
-outputs = model(**inputs)
-predicted_class_idx = outputs.logits.argmax(-1).item()
-print(f"Predicted class: {model.config.id2label[str(predicted_class_idx)]}")
+import tensorflow as tf
+import cv2
+import numpy as np
+
+# Load model
+model = tf.keras.models.load_model('drowsiness_model.h5')
+
+# Preprocess image
+img = cv2.imread('face.jpg')
+img = cv2.resize(img, (64, 64))
+img = img / 255.0
+img = np.expand_dims(img, axis=0)
+
+# Make prediction
+prediction = model.predict(img)
+is_drowsy = prediction[0][0] > 0.5
 ```
 
-## Citation
-
-If you use this model in your research or project, please provide a link to this model card.
-
-## Contact
-
-For questions or feedback, please open an issue on the GitHub repository or contact the author on Hugging Face.
+## Training Details
+- Dataset: Custom driver drowsiness dataset
+- Training method: Binary cross-entropy loss with Adam optimizer
+- Validation split: 20%
+- Early stopping with patience=3

config.json

@@ -0,0 +1 @@
+{"model_type": "drowsiness_detector", "input_shape": [64, 64, 3], "architecture": "CNN", "description": "A CNN model for detecting driver drowsiness from facial images"}

model_weights.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be899dd5a95dad19789a8352a7cbd1523a5c2c182795dd4195b4921dde7a3fb6
+size 4613168