diff --git a/examples/uci/tutorial.ipynb b/examples/uci/tutorial.ipynb new file mode 100644 index 0000000..54f657b --- /dev/null +++ b/examples/uci/tutorial.ipynb @@ -0,0 +1,37 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": null, + "id": "initial_id", + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/tests/gpu_tests/cpu_test.py b/tests/gpu_tests/cpu_test.py index 639a99f..2cb4952 100644 --- a/tests/gpu_tests/cpu_test.py +++ b/tests/gpu_tests/cpu_test.py @@ -34,7 +34,9 @@ def setUpClass(cls) -> None: cls.model = cls.model.double() cls.train_dataset = get_mnist_dataset(split="train", data_path="data") + cls.train_dataset = data.Subset(cls.train_dataset, indices=list(range(TRAIN_INDICES))) cls.eval_dataset = get_mnist_dataset(split="valid", data_path="data") + cls.eval_dataset = data.Subset(cls.eval_dataset, indices=list(range(QUERY_INDICES))) cls.task = ClassificationTask() cls.model = prepare_model(cls.model, cls.task) @@ -122,7 +124,6 @@ def test_pairwise_scores(self) -> None: ) new_pairwise_scores = self.analyzer.load_pairwise_scores(scores_name=NEW_SCORE_NAME) - torch.set_printoptions(threshold=30_000) print(f"Previous score: {pairwise_scores[ALL_MODULE_NAME][10]}") print(f"Previous shape: {pairwise_scores[ALL_MODULE_NAME].shape}") print(f"New score: {new_pairwise_scores[ALL_MODULE_NAME][10]}") diff --git a/tests/gpu_tests/prepare_tests.py b/tests/gpu_tests/prepare_tests.py index d46c1f4..f55c211 100644 --- a/tests/gpu_tests/prepare_tests.py +++ b/tests/gpu_tests/prepare_tests.py @@ -99,7 +99,6 @@ def run_analysis() -> None: activation_covariance_dtype=torch.float64, gradient_covariance_dtype=torch.float64, lambda_dtype=torch.float64, - lambda_iterative_aggregate=False, ) analyzer.fit_all_factors( factors_name="single_gpu",