Adding Residual DBN's tests.

tests/learnergy/models/test_residual_dbn.py

@@ -0,0 +1,88 @@
+import pytest
+import torch
+import torchvision
+
+from learnergy.models import residual_dbn
+
+
+def test_residual_dbn_zetta1():
+    new_residual_dbn = residual_dbn.ResidualDBN()
+
+    assert new_residual_dbn.zetta1 == 1
+
+
+def test_residual_dbn_zetta1_setter():
+    new_residual_dbn = residual_dbn.ResidualDBN()
+
+    try:
+        new_residual_dbn.zetta1 = 'a'
+    except:
+        new_residual_dbn.zetta1 = 0.1
+
+    assert new_residual_dbn.zetta1 == 0.1
+
+    try:
+        new_residual_dbn.zetta1 = -1
+    except:
+        new_residual_dbn.zetta1 = 0.1
+
+    assert new_residual_dbn.zetta1 == 0.1
+
+
+def test_residual_dbn_zetta2():
+    new_residual_dbn = residual_dbn.ResidualDBN()
+
+    assert new_residual_dbn.zetta2 == 1
+
+
+def test_residual_dbn_zetta2_setter():
+    new_residual_dbn = residual_dbn.ResidualDBN()
+
+    try:
+        new_residual_dbn.zetta2 = 'a'
+    except:
+        new_residual_dbn.zetta2 = 0.1
+
+    assert new_residual_dbn.zetta2 == 0.1
+
+    try:
+        new_residual_dbn.zetta2 = -1
+    except:
+        new_residual_dbn.zetta2 = 0.1
+
+    assert new_residual_dbn.zetta2 == 0.1
+
+
+def test_residual_dbn_calculate_residual():
+    new_residual_dbn = residual_dbn.ResidualDBN(n_visible=784, n_hidden=[128, 128], steps=[1, 1],
+                                                learning_rate=[0.1, 0.1], momentum=[0, 0], decay=[0, 0], temperature=[1, 1], use_gpu=False)
+
+    v = torch.ones(1, 784)
+
+    res = new_residual_dbn.calculate_residual(v)
+
+    assert res.size(1) == 784
+
+
+def test_residual_dbn_fit():
+    train = torchvision.datasets.MNIST(
+        root='./data', train=True, download=True, transform=torchvision.transforms.ToTensor())
+
+    new_residual_dbn = residual_dbn.ResidualDBN(n_visible=784, n_hidden=[128, 128], steps=[1, 1],
+                                                learning_rate=[0.1, 0.1], momentum=[0, 0], decay=[0, 0], temperature=[1, 1], use_gpu=False)
+
+    e, pl = new_residual_dbn.fit(train, batch_size=128, epochs=[1, 1])
+
+    assert len(e) == 2
+    assert len(pl) == 2
+
+
+def test_residual_dbn_forward():
+    new_residual_dbn = residual_dbn.ResidualDBN(n_visible=784, n_hidden=[128, 128], steps=[1, 1],
+                                                learning_rate=[0.1, 0.1], momentum=[0, 0], decay=[0, 0], temperature=[1, 1], use_gpu=False)
+
+    v = torch.ones(1, 784)
+
+    probs = new_residual_dbn.forward(v)
+
+    assert probs.size(1) == 128