# isort: skip_file # __use_pretrained_model_start__ import ray import tensorflow as tf from ray.train.batch_predictor import BatchPredictor from ray.train.tensorflow import ( to_air_checkpoint, TensorflowPredictor, ) # to simulate having a pretrained model. def build_model() -> tf.keras.Model: model = tf.keras.Sequential([ tf.keras.layers.InputLayer(input_shape=(1, )), tf.keras.layers.Dense(1), ]) return model model = build_model() checkpoint = to_air_checkpoint(model) batch_predictor = BatchPredictor(checkpoint, TensorflowPredictor, model_definition=build_model) predict_dataset = ray.data.range(3) predictions = batch_predictor.predict(predict_dataset) # __use_pretrained_model_end__
# __batch_prediction_start__ import pandas as pd from ray.train.batch_predictor import BatchPredictor batch_predictor = BatchPredictor(checkpoint, TensorflowPredictor, model_definition=build_model) # Create a dummy dataset. ds = ray.data.from_pandas( pd.DataFrame({ "feature_1": [1, 2, 3], "label": [1, 2, 3] })) # Use `feature_columns` to specify the input columns to your model. predictions = batch_predictor.predict(ds, feature_columns=["feature_1"]) print(predictions.show()) # {'predictions': array([-1.2789773], dtype=float32)} # {'predictions': array([-2.5579545], dtype=float32)} # {'predictions': array([-3.8369317], dtype=float32)} # __batch_prediction_end__ # __compute_accuracy_start__ def calculate_accuracy(df): return pd.DataFrame({"correct": int(df["predictions"][0]) == df["label"]}) predictions = batch_predictor.predict(ds, feature_columns=["feature_1"],