Fix failing tests

Author: ryan.smith

test/classification/test_classifier_convergence.py

@@ -51,7 +51,7 @@ def test_convergence(self):
         model = MultitaskClassifier(tasks=[task1, task2])
 
         # Train
-        trainer = Trainer(lr=0.001, n_epochs=10, progress_bar=False)
+        trainer = Trainer(lr=0.002, n_epochs=50, progress_bar=False)
         trainer.fit(model, dataloaders)
         scores = model.score(dataloaders)
 

test/classification/test_loss.py

@@ -19,15 +19,15 @@ def test_sce_equals_ce(self):
 
         ce_loss = F.cross_entropy(Y_probs, Y_golds, reduction="none")
         ces_loss = cross_entropy_with_probs(Y_probs, Y_golds_probs, reduction="none")
-        np.testing.assert_equal(ce_loss.numpy(), ces_loss.numpy())
+        np.testing.assert_almost_equal(ce_loss.numpy(), ces_loss.numpy())
 
         ce_loss = F.cross_entropy(Y_probs, Y_golds, reduction="sum")
         ces_loss = cross_entropy_with_probs(Y_probs, Y_golds_probs, reduction="sum")
-        np.testing.assert_equal(ce_loss.numpy(), ces_loss.numpy())
+        np.testing.assert_almost_equal(ce_loss.numpy(), ces_loss.numpy())
 
         ce_loss = F.cross_entropy(Y_probs, Y_golds, reduction="mean")
         ces_loss = cross_entropy_with_probs(Y_probs, Y_golds_probs, reduction="mean")
-        np.testing.assert_equal(ce_loss.numpy(), ces_loss.numpy())
+        np.testing.assert_almost_equal(ce_loss.numpy(), ces_loss.numpy())
 
     def test_perfect_predictions(self):
         # Does soft ce loss achieve approx. 0 loss with perfect predictions?