Test.csv contains a list of image identifiers (IDs) that tell you which images to run inference (prediction) on. These images are not used for training but for evaluating the model’s performance

After you've trained your model on your dataset, you use the images specified in Test.csv for testing or inference. This means your model will take these images, make predictions about the objects or classes present in them, and output key information.

The output from the model typically includes: Class: The predicted category or object (e.g., WBC, Trophozites). Confidence: A score that indicates how confident the model is about its prediction. Bounding Box Coordinates: This includes the values for xmin, ymin, xmax, and ymax, which define the area of the image where the object is located.

Once the model has made predictions on the test images, you save the results in a new CSV file. This file becomes your submission file, which will typically have columns like: Image ID: The ID of the image. Class: The predicted class or label. Confidence: The confidence score. xmin, ymin, xmax, ymax: The coordinates of the bounding boxes.

Then submit this file to be scored