imrkash's picture
Upload sequential-algo/predict.py with huggingface_hub
28c212d verified
Raw History Blame Contribute Delete
1.76 kB
import os
import numpy as np
import tensorflow as tf
class_names= ['cocci', 'healthy', 'ncd', 'salmo']
def predict():
prediction = []
image_files = load_test_data()
# Iterate over the image files
for i, image_file in enumerate(image_files):
# Load and preprocess the image
image_path = os.path.join(get_upload_dir(), image_file)
img = tf.keras.preprocessing.image.load_img(image_path, target_size=(256, 256))
img_array = tf.keras.preprocessing.image.img_to_array(img)
img_array = tf.expand_dims(img_array, 0) # Create a batch axis
# Make predictions
saved_model=load_model(get_model_path())
predictions = saved_model.predict(img_array)
predicted_label = class_names[np.argmax(predictions[0])]
accuracy = round(np.max(predictions[0]) * 100, 2)
prediction.append({"label": predicted_label, "accuracy": accuracy})
return prediction
def load_test_data():
upload_dir = os.path.abspath(get_upload_dir())
image_files = [f for f in os.listdir(upload_dir) if f.endswith(('.jpg', '.png', '.jpeg'))]
return image_files or []
def load_model(model_path):
try:
return tf.keras.models.load_model(model_path)
except Exception as e:
raise ValueError(f"Failed to load model from '{model_path}': {str(e)}")
def get_pwd():
# Define the path relative to the script's directory
return os.path.dirname(os.path.abspath(__file__))
def get_upload_dir():
upload_dir = os.path.join(get_pwd(), '..', 'uploads')
# Get the absolute path to the uploads directory
return os.path.abspath(upload_dir)
def get_model_path():
current_dir = get_pwd()
return os.path.join(current_dir, 'model', 'sequential-model.keras')