Create unit tests for extraction and utils

README.md

@@ -72,6 +72,8 @@ The other classes of attacks can be launched in a similar fashion. See the `exam
 2. Git clone this repository.
 3. Run `poetry install`.
 
+If you'd like to use a Jupyter Notebook environment, run `poetry shell` followed by `jupyter notebook`.
+
 Feel free to join our #privacyraven channel in [Empire Hacking](https://empireslacking.herokuapp.com/) if you need help using or extending PrivacyRaven.
 The official pip release will arrive soon.
 

examples/README.md

@@ -1,4 +1,4 @@
 Overview:
 
-+ `example_mnist_cpu.py`
-+ `example_mnist_gpu.py`
++ `example_mnist_cpu.py`: Applies a model extraction attack with the copycat synthesizer to an example MNIST model using a CPU
++ `example_mnist_gpu.py`: Applies a model extraction attack with the copycat synthesizer to an example MNIST model using a GPU

examples/extract_mnist_gpu.py

@@ -7,6 +7,7 @@
 """
 
 import privacyraven as pr
+
 from privacyraven.utils.data import get_emnist_data
 from privacyraven.extraction.core import ModelExtractionAttack
 from privacyraven.utils.query import get_target
@@ -19,13 +20,13 @@
 
 def query_mnist(input_data):
     # PrivacyRaven provides built-in query functions
-    return get_target(model, input_data)
+    return get_target(model, input_data, (1, 28, 28, 1))
 
 
 # Obtain seed (or public) data to be used in extraction
 emnist_train, emnist_test = get_emnist_data()
 
-# Run a Model Extraction Attack
+# Run a model extraction attack
 attack = ModelExtractionAttack(
     query_mnist,
     100,
@@ -38,3 +39,5 @@ def query_mnist(input_data):
     emnist_train,
     emnist_test,
 )
+
+# Use emnist_train.data and emnist_test.data for unlabeled data