From ddef0cd5644df31e27eda988d350bffe70641132 Mon Sep 17 00:00:00 2001
From: Maximilian Schmidt <schmidt.maximilian@posteo.de>
Date: Mon, 25 May 2020 20:51:42 +0900
Subject: [PATCH] Fix issues with non-installed extra requirements

Stringify annotations using optional dependencies
Mark test relying on optional dependencies as skip
Demote f-string to normal string
Fix flake8 issue in setup.py
---
 cgp/cartesian_graph.py             | 6 +++---
 cgp/local_search/gradient_based.py | 6 +++---
 setup.py                           | 4 ++--
 test/test_individual.py            | 6 +++---
 4 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/cgp/cartesian_graph.py b/cgp/cartesian_graph.py
index e60d90d4..34b779f6 100644
--- a/cgp/cartesian_graph.py
+++ b/cgp/cartesian_graph.py
@@ -309,7 +309,7 @@ def _f(x):
 
     def to_torch(
         self, parameter_names_to_values: Optional[Dict[str, float]] = None
-    ) -> torch.nn.Module:
+    ) -> "torch.nn.Module":
         """Compile the function(s) represented by the graph to a Torch class.
 
         Generates a definition of the Torch class in Python code and
@@ -335,7 +335,7 @@ def to_torch(
                     node.format_parameter_str()
                     all_parameter_str.append(node.parameter_str)
         forward_str = ", ".join(node.output_str for node in self.output_nodes)
-        class_str = f"""\
+        class_str = """\
 class _C(torch.nn.Module):
 
     def __init__(self):
@@ -370,7 +370,7 @@ def to_sympy(
         self,
         simplify: Optional[bool] = True,
         parameter_names_to_values: Optional[Dict[str, float]] = None,
-    ) -> List[sympy_expr.Expr]:
+    ) -> List["sympy_expr.Expr"]:
         """Compile the function(s) represented by the graph to a SymPy expression.
 
         Generates one SymPy expression for each output node.
diff --git a/cgp/local_search/gradient_based.py b/cgp/local_search/gradient_based.py
index 10e4f4d1..4b3fdbf7 100644
--- a/cgp/local_search/gradient_based.py
+++ b/cgp/local_search/gradient_based.py
@@ -2,7 +2,7 @@
 
 try:
     import torch  # noqa: F401
-    from torch.optim.optimizer import Optimizer
+    from torch.optim.optimizer import Optimizer  # noqa: F401
 
     torch_available = True
 except ModuleNotFoundError:
@@ -16,10 +16,10 @@
 
 def gradient_based(
     individual: Individual,
-    objective: Callable[[torch.nn.Module], torch.Tensor],
+    objective: Callable[["torch.nn.Module"], "torch.Tensor"],
     lr: float,
     gradient_steps: int,
-    optimizer: Optional[Optimizer] = None,
+    optimizer: Optional["Optimizer"] = None,
     clip_value: Optional[float] = None,
 ) -> None:
     """Perform a local search for numeric leaf values for an individual
diff --git a/setup.py b/setup.py
index 220eaefa..35c96488 100644
--- a/setup.py
+++ b/setup.py
@@ -13,8 +13,8 @@ def read_extra_requirements():
     extra_requirements = {}
     extra_requirements["all"] = []
     with open("./extra-requirements.txt") as f:
-        for l in f:
-            req = l.replace("\n", " ")
+        for dep in f:
+            req = dep.replace("\n", " ")
             extra_requirements[req] = [req]
             extra_requirements["all"].append(req)
 
diff --git a/test/test_individual.py b/test/test_individual.py
index d1cb7596..ffdd1f78 100644
--- a/test/test_individual.py
+++ b/test/test_individual.py
@@ -1,7 +1,6 @@
 import math
 import pickle
 import pytest
-import torch
 
 import cgp
 from cgp.individual import Individual
@@ -57,7 +56,7 @@ def test_individual_with_parameter_python():
 
 
 def test_individual_with_parameter_torch():
-
+    torch = pytest.importorskip("torch")
     primitives = (cgp.Add, cgp.Parameter)
     genome = cgp.Genome(1, 1, 2, 1, 2, primitives)
     # f(x) = x + c
@@ -97,7 +96,7 @@ def test_individual_with_parameter_torch():
 
 
 def test_individual_with_parameter_sympy():
-
+    sympy = pytest.importorskip("sympy")  # noqa
     primitives = (cgp.Add, cgp.Parameter)
     genome = cgp.Genome(1, 1, 2, 1, 2, primitives)
     # f(x) = x + c
@@ -135,6 +134,7 @@ def test_individual_with_parameter_sympy():
 
 
 def test_to_and_from_torch_plus_backprop():
+    torch = pytest.importorskip("torch")
     primitives = (cgp.Mul, cgp.Parameter)
     genome = cgp.Genome(1, 1, 2, 2, 1, primitives)
     # f(x) = c * x