| import cv2 | |
| import numpy as np | |
| from tensorflow.keras.models import load_model | |
| # Load the model | |
| model_output = '/content/drive/MyDrive/saved_models/gender_model/gender_compiled_model' | |
| model = load_model(model_output) | |
| # Load and preprocess an image (assuming 'image_path' is the path to your image) | |
| def preprocess_image(image_path): | |
| # Load the image using OpenCV | |
| img = cv2.imread(image_path) | |
| # Resize the image to 160x160, which is the expected input size for InceptionResNetV1 | |
| img = cv2.resize(img, (224, 224)) | |
| # Convert the image to RGB | |
| img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) | |
| # Normalize the image (the model expects pixel values between -1 and 1) | |
| img = img.astype('float32') / 127.5 - 1 | |
| # Add a batch dimension | |
| img = np.expand_dims(img, axis=0) | |
| return img | |
| # Path to the image you want to test | |
| image_path = '/content/pic.jpeg' | |
| # Preprocess the image | |
| input_image = preprocess_image(image_path) | |
| # Perform inference to get the face embedding | |
| pred = model.predict(input_image) | |
| # Labels for the genders | |
| labels = ["Woman", "Man"] | |
| predicted_label_index = np.argmax(pred) | |
| # Get the corresponding label | |
| predicted_gender = labels[predicted_label_index] | |
| print(f"Predicted Gender: {predicted_gender}") | |