예제 #1
0
    def test_train_save_restore_sharded(self):
        """Saves and restores a sharded checkpoint to check for equivalence."""
        if fastmath.local_device_count() < 2:
            return  # multi-accelerator only
        base.N_WEIGHTS_SHARDS = fastmath.local_device_count()
        train_data = data.Serial(lambda _: _very_simple_data(2, 2),
                                 data.CountAndSkip('simple_data'))
        task = training.TrainTask(train_data(), tl.L2Loss(),
                                  optimizers.Adam(.0001))
        eval_task = training.EvalTask(
            _very_simple_data(2, 2),  # deliberately re-using training data
            [tl.L2Loss()],
            metric_names=['SGD.L2Loss'])
        tmp_dir = self.create_tempdir().full_path

        def _make_model_and_session():
            m = tl.Serial(tl.Dense(2))
            ts = training.Loop(m, [task],
                               eval_tasks=[eval_task],
                               eval_at=lambda step_n: step_n % 2 == 0,
                               output_dir=tmp_dir)
            return m, ts

        _, training_session = _make_model_and_session()
        self.assertEqual(0, training_session.step)
        training_session.run(n_steps=1)
        training_session.save_checkpoint('model')
        _, training_session2 = _make_model_and_session()
        training_session2.run(n_steps=1)
        base.N_WEIGHTS_SHARDS = 1
예제 #2
0
 def test_count_and_skip(self):
     dataset = lambda _: ((i, i + 1) for i in range(10))
     examples = data.Serial(dataset, data.CountAndSkip('toy_data'))
     ex_generator = examples()
     ex1 = next(ex_generator)
     self.assertEqual(ex1, (0, 1))
     self.assertEqual(data.inputs.data_counters['toy_data'], 1)
     ex2 = next(ex_generator)
     self.assertEqual(ex2, (1, 2))
     self.assertEqual(data.inputs.data_counters['toy_data'], 2)
     ex3 = next(examples())  # new generator, will skip
     self.assertEqual(ex3, (2, 3))
     self.assertEqual(data.inputs.data_counters['toy_data'], 3)
     data.inputs.data_counters['toy_data'] = 0  # reset
     ex4 = next(examples())  # new generator, was reset
     self.assertEqual(ex4, (0, 1))
     self.assertEqual(data.inputs.data_counters['toy_data'], 1)
예제 #3
0
    def test_train_save_restore_dense(self):
        """Saves and restores a checkpoint to check for equivalence."""
        train_data = data.Serial(lambda _: _very_simple_data(),
                                 data.CountAndSkip('simple_data'))
        task = training.TrainTask(train_data(), tl.L2Loss(),
                                  optimizers.Adam(.0001))
        eval_task = training.EvalTask(
            _very_simple_data(),  # deliberately re-using training data
            [tl.L2Loss()],
            metric_names=['SGD.L2Loss'])
        tmp_dir = self.create_tempdir().full_path

        def _make_model_and_session():
            m = tl.Serial(tl.Dense(1))
            ts = training.Loop(m, [task],
                               eval_tasks=[eval_task],
                               eval_at=lambda step_n: step_n % 2 == 0,
                               output_dir=tmp_dir)
            return m, ts

        model, training_session = _make_model_and_session()
        self.assertEqual(0, training_session.step)
        training_session.run(n_steps=1)
        training_session.save_checkpoint()
        self.assertEqual(data.inputs.data_counters['simple_data'], 2)
        data.inputs.data_counters['simple_data'] = 0  # reset manually
        self.assertEqual(data.inputs.data_counters['simple_data'], 0)  # check
        model2, training_session2 = _make_model_and_session()
        self.assertEqual(data.inputs.data_counters['simple_data'],
                         2)  # restored

        x = np.ones((8, 1))
        y1 = model(x, rng=fastmath.random.get_prng(0))
        y2 = model2(x, rng=fastmath.random.get_prng(0))
        self.assertEqual(str(y1), str(y2))

        training_session2.run(n_steps=1)
        y1 = model(x, rng=fastmath.random.get_prng(0))
        y2 = model2(x, rng=fastmath.random.get_prng(0))
        self.assertNotEqual(str(y1), str(y2))

        slots1 = training_session._trainer_per_task[0].slots
        slots2 = training_session2._trainer_per_task[0].slots
        np.testing.assert_array_equal(slots1, slots2)