6aad1de658a933d3fa376f7fe9abf419da8a8bd2,python/baseline/tf/seq2seq/train.py,Seq2SeqTrainerTf,test,#Seq2SeqTrainerTf#Any#Any#Any#,69

Before Change

            epochs = self.valid_epochs

        total_loss = 0
        steps = len(vs)
        metrics = {}

        start = time.time()
        for batch_dict in vs:

            feed_dict = self.model.make_input(batch_dict)
            vals = self.sess.run([self.test_loss], feed_dict=feed_dict)
            lossv = vals[0]
            total_loss += lossv

        self.log.debug({"phase": phase, "time": time.time() - start})
        avg_loss = total_loss/steps
        metrics["avg_loss"] = avg_loss
        metrics["perplexity"] = np.exp(avg_loss)
        for reporting in reporting_fns:
            reporting(metrics, epochs, phase)
        return metrics


@register_training_func("seq2seq")

After Change

        )
        return metrics

    def test(self, vs, reporting_fns, phase="Valid"):
        epochs = 0
        if phase == "Valid":
            self.valid_epochs += 1
            epochs = self.valid_epochs

        total_loss = 0
        total_toks = 0
        metrics = {}

        start = time.time()
        for batch_dict in vs:

            feed_dict = self.model.make_input(batch_dict)
            lossv = self.model.sess.run(self.test_loss, feed_dict=feed_dict)
            toks = self._num_toks(batch_dict["tgt_lengths"])
            total_loss += lossv * toks
            total_toks += toks

        metrics = self.calc_metrics(total_loss, total_toks)
        self.report(
            epochs, metrics, start,
            phase, "EPOCH", reporting_fns
        )


@register_training_func("seq2seq")
def fit(model, ts, vs, es=None, **kwargs):

Instances