def test_integration_median_absolute_percentage_error_with_output_transform(): np.random.seed(1) size = 105 np_y_pred = np.random.rand(size, 1) np_y = np.random.rand(size, 1) np.random.shuffle(np_y) np_median_absolute_percentage_error = 100.0 * np.median( np.abs(np_y - np_y_pred) / np.abs(np_y)) batch_size = 15 def update_fn(engine, batch): idx = (engine.state.iteration - 1) * batch_size y_true_batch = np_y[idx:idx + batch_size] y_pred_batch = np_y_pred[idx:idx + batch_size] return idx, torch.from_numpy(y_pred_batch), torch.from_numpy( y_true_batch) engine = Engine(update_fn) m = MedianAbsolutePercentageError(output_transform=lambda x: (x[1], x[2])) m.attach(engine, "median_absolute_percentage_error") data = list(range(size // batch_size)) median_absolute_percentage_error = engine.run( data, max_epochs=1).metrics["median_absolute_percentage_error"] assert np_median_absolute_percentage_error == pytest.approx( median_absolute_percentage_error)
def _test(metric_device): metric_device = torch.device(metric_device) m = MedianAbsolutePercentageError(device=metric_device) torch.manual_seed(10 + rank) size = 105 y_pred = torch.randint(1, 10, size=(size, 1), dtype=torch.double, device=device) y = torch.randint(1, 10, size=(size, 1), dtype=torch.double, device=device) m.update((y_pred, y)) # gather y_pred, y y_pred = idist.all_gather(y_pred) y = idist.all_gather(y) np_y_pred = y_pred.cpu().numpy().ravel() np_y = y.cpu().numpy().ravel() res = m.compute() e = np.abs(np_y - np_y_pred) / np.abs(np_y) # The results between numpy.median() and torch.median() are Inconsistant # when the length of the array/tensor is even. So this is a hack to avoid that. # issue: https://github.com/pytorch/pytorch/issues/1837 if np_y_pred.shape[0] % 2 == 0: e_prepend = np.insert(e, 0, e[0], axis=0) np_res_prepend = 100.0 * np.median(e_prepend) assert pytest.approx(res) == np_res_prepend else: np_res = 100.0 * np.median(e) assert pytest.approx(res) == np_res
def test_zero_sample(): m = MedianAbsolutePercentageError() with pytest.raises( NotComputableError, match= r"EpochMetric must have at least one example before it can be computed" ): m.compute()
def test_wrong_input_shapes(): m = MedianAbsolutePercentageError() with pytest.raises(ValueError): m.update((torch.rand(4, 1, 2), torch.rand(4, 1))) with pytest.raises(ValueError): m.update((torch.rand(4, 1), torch.rand(4, 1, 2))) with pytest.raises(ValueError): m.update((torch.rand(4, 1, 2), torch.rand(4, ))) with pytest.raises(ValueError): m.update((torch.rand(4, ), torch.rand(4, 1, 2)))
def test_wrong_input_shapes(): m = MedianAbsolutePercentageError() with pytest.raises(ValueError, match=r"Predictions should be of shape"): m.update((torch.rand(4, 1, 2), torch.rand(4, 1))) with pytest.raises(ValueError, match=r"Targets should be of shape"): m.update((torch.rand(4, 1), torch.rand(4, 1, 2))) with pytest.raises(ValueError, match=r"Predictions should be of shape"): m.update((torch.rand(4, 1, 2), torch.rand(4))) with pytest.raises(ValueError, match=r"Targets should be of shape"): m.update((torch.rand(4), torch.rand(4, 1, 2)))
def _test(n_epochs, metric_device): metric_device = torch.device(metric_device) n_iters = 80 size = 105 y_true = torch.rand(size=(size, )).to(device) y_preds = torch.rand(size=(size, )).to(device) def update(engine, i): return ( y_preds[i * size:(i + 1) * size], y_true[i * size:(i + 1) * size], ) engine = Engine(update) m = MedianAbsolutePercentageError(device=metric_device) m.attach(engine, "mape") data = list(range(n_iters)) engine.run(data=data, max_epochs=n_epochs) assert "mape" in engine.state.metrics res = engine.state.metrics["mape"] np_y_true = y_true.cpu().numpy().ravel() np_y_preds = y_preds.cpu().numpy().ravel() e = np.abs(np_y_true - np_y_preds) / np.abs(np_y_true) np_res = 100.0 * np.median(e) e_prepend = np.insert(e, 0, e[0], axis=0) np_res_prepend = 100.0 * np.median(e_prepend) # The results between numpy.median() and torch.median() are Inconsistant # when the length of the array/tensor is even. So this is a hack to avoid that. # issue: https://github.com/pytorch/pytorch/issues/1837 if np_y_preds.shape[0] % 2 == 0: assert pytest.approx(res) == np_res_prepend else: assert pytest.approx(res) == np_res
def test_median_absolute_percentage_error(): # See https://github.com/torch/torch7/pull/182 # For even number of elements, PyTorch returns middle element # NumPy returns average of middle elements # Size of dataset will be odd for these tests size = 51 np_y_pred = np.random.rand(size,) np_y = np.random.rand(size,) np_median_absolute_percentage_error = 100.0 * np.median(np.abs(np_y - np_y_pred) / np.abs(np_y)) m = MedianAbsolutePercentageError() y_pred = torch.from_numpy(np_y_pred) y = torch.from_numpy(np_y) m.reset() m.update((y_pred, y)) assert np_median_absolute_percentage_error == pytest.approx(m.compute())
def test_median_absolute_percentage_error_2(): np.random.seed(1) size = 105 np_y_pred = np.random.rand(size, 1) np_y = np.random.rand(size, 1) np.random.shuffle(np_y) np_median_absolute_percentage_error = 100.0 * np.median(np.abs(np_y - np_y_pred) / np.abs(np_y)) m = MedianAbsolutePercentageError() y_pred = torch.from_numpy(np_y_pred) y = torch.from_numpy(np_y) m.reset() batch_size = 16 n_iters = size // batch_size + 1 for i in range(n_iters): idx = i * batch_size m.update((y_pred[idx : idx + batch_size], y[idx : idx + batch_size])) assert np_median_absolute_percentage_error == pytest.approx(m.compute())