fix unit test wrt deprecated functions and trying to remove nonexiste…

…nt files

README.md

@@ -86,6 +86,7 @@ graph = MNISTGraph(
 )
 graph.train(data_sets, max_steps=5000, precision=0.95)
 
+# verify the training worked by testing if a handwritten number is recognized
 image_data = MNISTDataSets.read_one_image_from_url(
     'http://github.com/lene/nn-wtf/blob/master/nn_wtf/data/7_from_test_set.raw?raw=true'
 )

nn_wtf/tests/save_and_restore_test.py

@@ -8,7 +8,7 @@
 
 from tempfile import gettempdir
 from os import remove
-from os.path import join
+from os.path import join, isfile
 
 __author__ = 'Lene Preuss <lene.preuss@gmail.com>'
 
@@ -30,7 +30,8 @@ def setUp(self):
 
     def tearDown(self):
         for filename in self.generated_filenames:
-            remove(join(gettempdir(), filename))
+            if isfile(join(gettempdir(), filename)):
+                remove(join(gettempdir(), filename))
 
     def test_save_untrained_network_runs(self):
         graph = init_graph(SavableNetwork())
@@ -46,22 +47,23 @@ def test_prediction_with_trained_graph(self):
         savefile = self._save_trained_graph()
 
         new_graph = init_graph(SavableNetwork(), session=tf.Session())
-        for v in tf.all_variables():
+        for v in tf.global_variables():
             self.assertFalse(self.is_ndarray_equal(v.eval(new_graph.session), self.variables[v.op.name]))
         new_graph.restore()#savefile)
-        for v in tf.all_variables():
+        for v in tf.global_variables():
             self.assertTrue(self.is_ndarray_equal(v.eval(new_graph.session), self.variables[v.op.name]))
 
     def _save_trained_graph(self):
         tf.reset_default_graph()
         with train_neural_network(create_train_data_set(), SavableNetwork()) as graph:
-            for v in tf.all_variables():
+            for v in tf.global_variables():
                 self.variables[v.op.name] = v.eval(graph.session)
             return graph.save(global_step=graph.trainer.num_steps())
 
     def _add_savefiles_to_list(self, savefile):
         self.generated_filenames.extend([savefile, '{}.meta'.format(savefile), 'checkpoint'])
 
-    def is_ndarray_equal(self, array_1, array_2):
+    @staticmethod
+    def is_ndarray_equal(array_1, array_2):
         import numpy
         return numpy.array_equal(array_1, array_2)