pipeline.py

from transformers import Pipeline
import numpy as np
import tensorflow as tf
from PIL import Image

class BrainTumorDetectionPipeline(Pipeline):
    def __init__(self, model_path: str, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self.model = tf.keras.models.load_model(model_path)

    def preprocess(self, images: list):
        processed_images = []
        for image in images:
            image = Image.open(image).resize((128, 128))  # Resize to model input size
            image = np.array(image) / 255.0  # Normalize the image
            processed_images.append(np.expand_dims(image, axis=0))  # Add batch dimension
        return np.vstack(processed_images)  # Stack for batch processing

    def forward(self, images: list):
        preprocessed_images = self.preprocess(images)
        predictions = self.model.predict(preprocessed_images)
        return predictions