def __init__()

in src/trainer.py [0:0]

• 39 lines of code
• 12 McCabe index (conditional complexity)

    def __init__(self, model, params, ftmodel=None, teacher_model=None):
        """
        Initialize trainer.
        """
        # pretrained model / model / params
        self.model = model
        self.ftmodel = ftmodel
        self.teacher_model = teacher_model
        self.params = params
        assert params.fp16 is False

        # set parameters
        self.set_parameters()

        # set optimizers
        self.set_optimizers()

        # stopping criterion used for early stopping
        if params.stopping_criterion != '':
            split = params.stopping_criterion.split(',')
            assert len(split) == 2 and split[1].isdigit()
            self.decrease_counts_max = int(split[1])
            self.decrease_counts = 0
            if split[0][0] == '_':
                self.stopping_criterion = (split[0][1:], False)
            else:
                self.stopping_criterion = (split[0], True)
            self.best_stopping_criterion = -1e12 if self.stopping_criterion[1] else 1e12
        else:
            self.stopping_criterion = None
            self.best_stopping_criterion = None

        # validation metrics
        self.metrics = []
        metrics = [m for m in params.validation_metrics.split(',') if m != '']
        for m in metrics:
            m = (m[1:], False) if m[0] == '_' else (m, True)
            self.metrics.append(m)
        self.best_metrics = {metric: (-1e12 if biggest else 1e12) for (metric, biggest) in self.metrics}

        # training statistics
        self.epoch = 0
        self.indices = []
        self.n_iter = 0
        self.embeddings = None
        self.stats = OrderedDict(
            [('processed_i', 0)] +
            [('MSE', [])] +
            [('XE', [])] +
            [('triplet', [])] +
            [('time', [])]
        )
        self.last_time = time.time()