def test_inference_runner_multiple_tensor_features(self):
examples = torch.from_numpy(
np.array([1, 5, 3, 10, -14, 0, 0.5, 0.5],
dtype="float32")).reshape(-1, 2)
examples = [
torch.from_numpy(np.array([1, 5], dtype="float32")),
torch.from_numpy(np.array([3, 10], dtype="float32")),
torch.from_numpy(np.array([-14, 0], dtype="float32")),
torch.from_numpy(np.array([0.5, 0.5], dtype="float32")),
]
expected_predictions = [
PredictionResult(ex, pred) for ex, pred in zip(
examples,
torch.Tensor([f1 * 2.0 + f2 * 3 + 0.5
for f1, f2 in examples]).reshape(-1, 1))
]
model = PytorchLinearRegression(input_dim=2, output_dim=1)
model.load_state_dict(
OrderedDict([('linear.weight', torch.Tensor([[2.0, 3]])),
('linear.bias', torch.Tensor([0.5]))]))
model.eval()
inference_runner = PytorchInferenceRunner(torch.device('cpu'))
predictions = inference_runner.run_inference(examples, model)
for actual, expected in zip(predictions, expected_predictions):
self.assertTrue(_compare_prediction_result(actual, expected))
def test_num_bytes(self):
inference_runner = PytorchInferenceRunner(torch.device('cpu'))
examples = torch.from_numpy(
np.array([1, 5, 3, 10, -14, 0, 0.5, 0.5],
dtype="float32")).reshape(-1, 2)
self.assertEqual((examples[0].element_size()) * 8,
inference_runner.get_num_bytes(examples))
def test_namespace(self):
inference_runner = PytorchInferenceRunner(torch.device('cpu'))
self.assertEqual('RunInferencePytorch',
inference_runner.get_metrics_namespace())