Upgrade lint. (pytorch#39483)

Summary:
Pull Request resolved: pytorch#39483

I fixed all of the new errors that occurred because of the upgrade.

Signed-off-by: Edward Z. Yang <[email protected]>

Test Plan: Imported from OSS

Differential Revision: D21884575

Pulled By: ezyang

fbshipit-source-id: 45c8e1f1ecb410c8d7c46dd3922ad70e982a0685

Author: ezyang

.flake8

@@ -5,9 +5,12 @@ max-line-length = 120
 # E501 is not flexible enough, we're using B950 instead
 ignore =
     E203,E305,E402,E501,E721,E741,F403,F405,F821,F841,F999,W503,W504,C408,E302,W291,E303,
+    # shebang has extra meaning in fbcode lints, so I think it's not worth trying
+    # to line this up with executable bit
+    EXE001,
     # these ignores are from flake8-bugbear; please fix!
     B007,B008,
     # these ignores are from flake8-comprehensions; please fix!
     C400,C401,C402,C403,C404,C405,C407,C411,C413,C414,C415
 per-file-ignores = __init__.py: F401
-exclude = docs/src,venv,third_party,caffe2,scripts,docs/caffe2,torch/lib/include,torch/lib/tmp_install,build,torch/include,*.pyi,.git
+exclude = docs/src,venv,third_party,caffe2,scripts,docs/caffe2,torch/lib/include,torch/lib/tmp_install,build,torch/include,*.pyi,.git,build,build_test_custom_build,build_code_analyzer

.github/workflows/lint.yml

@@ -66,7 +66,7 @@ jobs:
       - name: Run flake8
         run: |
           set -eux
-          pip install flake8==3.7.9 flake8-mypy flake8-bugbear flake8-comprehensions flake8-executable flake8-pyi mccabe pycodestyle==2.5.0 pyflakes==2.1.1
+          pip install flake8==3.8.2 flake8-mypy flake8-bugbear flake8-comprehensions flake8-executable flake8-pyi==20.5.0 mccabe pycodestyle==2.6.0 pyflakes==2.2.0
           flake8 --version
           flake8 --exit-zero > ${GITHUB_WORKSPACE}/flake8-output.txt
           cat ${GITHUB_WORKSPACE}/flake8-output.txt

benchmarks/operator_benchmark/pt/qpool_test.py

@@ -10,12 +10,12 @@
 # 2D pooling will have input matrix of rank 3 or 4
 qpool2d_long_configs = op_bench.config_list(
     attrs=(
-       #  C    H    W   k       s       p
-       (  1,   3,   3, (3, 3), (1, 1), (0, 0)),  # dummy        # noqa
-       (  3,  64,  64, (3, 3), (2, 2), (1, 1)),  # dummy        # noqa
-       # VGG16 pools with original input shape: (-1, 3, 224, 224)
-       ( 64, 224, 224, (2, 2), (2, 2), (0, 0)),  # MaxPool2d-4  # noqa
-       (256,  56,  56, (2, 2), (2, 2), (0, 0)),  # MaxPool2d-16 # noqa
+        #  C    H    W   k       s       p
+        (  1,   3,   3, (3, 3), (1, 1), (0, 0)),  # dummy        # noqa
+        (  3,  64,  64, (3, 3), (2, 2), (1, 1)),  # dummy        # noqa
+        # VGG16 pools with original input shape: (-1, 3, 224, 224)
+        ( 64, 224, 224, (2, 2), (2, 2), (0, 0)),  # MaxPool2d-4  # noqa
+        (256,  56,  56, (2, 2), (2, 2), (0, 0)),  # MaxPool2d-16 # noqa
     ),
     attr_names=('C', 'H', 'W',   # Input layout
                 'k', 's', 'p'),  # Pooling parameters

test/jit/test_class_type.py

@@ -485,6 +485,7 @@ def forward(self, a):
 
     def test_interface(self):
         global Foo, Bar, OneTwo, OneTwoThree, OneTwoWrong, NotMember, NotMember2
+
         @torch.jit.script
         class Foo(object):
             def __init__(self):
@@ -647,6 +648,7 @@ def __init__(self):
 
     def test_overloaded_fn(self):
         global Foo, MyClass  # see [local resolution in python]
+
         @torch.jit.script
         class Foo(object):
             def __init__(self, x):
@@ -802,6 +804,7 @@ def test():
 
     def test_cast_overloads(self):
         global Foo  # see [local resolution in python]
+
         @torch.jit.script
         class Foo(object):
             def __init__(self, val):

test/print_test_stats.py

@@ -47,11 +47,11 @@ def print_report(self):
         print(f"    avg_time: {self.total_time/test_count:.2f} seconds")
         if test_count > 2:
             print(f"    mean_time: {sorted_tests[test_count>>1].time:.2f} seconds")
-            print(f"    Three longest tests:")
+            print("    Three longest tests:")
             for idx in [-1, -2, -3]:
                 print(f"        {sorted_tests[idx].name} time: {sorted_tests[idx].time:.2f} seconds")
         elif test_count > 0:
-            print(f"    Longest test:")
+            print("    Longest test:")
             print(f"        {sorted_tests[-1].name} time: {sorted_tests[-1].time:.2f} seconds")
         print("")
 

test/quantization/test_quantized_op.py

@@ -1401,7 +1401,8 @@ def test_interpolate(self, X, size, mode, scale_factor, align_corners, nhwc_layo
                         mode=mode, align_corners=align_corners)
             # TODO(#38095): Replace assertEqualIgnoreType. See issue #38095
             self.assertEqualIgnoreType(X_ref, qX_hat.int_repr(), atol=1.0, rtol=0,
-                                       msg="{} results are off".format(name, qX_hat.int_repr(), X_ref))
+                                       msg="{} results are off: qX_hat={} X_ref={}"
+                                           .format(name, qX_hat.int_repr(), X_ref))
             self.assertEqual(scale, qX_hat.q_scale(),
                              msg=error_message.format(name + '.scale', scale, qX_hat.q_scale()))
             self.assertEqual(zero_point, qX_hat.q_zero_point(),
@@ -1455,7 +1456,8 @@ def test_interpolate3d(self, X, size, scale_factor, align_corners, nhwc_layout):
                         mode=mode, align_corners=align_corners)
             # TODO(#38095): Replace assertEqualIgnoreType. See issue #38095
             self.assertEqualIgnoreType(X_ref, qX_hat.int_repr(), atol=1.0, rtol=0,
-                                       msg="{} results are off".format(name, qX_hat.int_repr(), X_ref))
+                                       msg="{} results are off: qX_hat={}, X_ref={}"
+                                           .format(name, qX_hat.int_repr(), X_ref))
             self.assertEqual(scale, qX_hat.q_scale(),
                              msg=error_message.format(name + '.scale', scale, qX_hat.q_scale()))
             self.assertEqual(zero_point, qX_hat.q_zero_point(),

test/test_autograd.py

@@ -6072,6 +6072,7 @@ def test_copy_(self, device):
     def test_simple_reentrant_cross_device(self, device):
         class ReentrantFunc(Function):
             _cpu_mode = True
+
             @staticmethod
             def forward(ctx, x):
                 return x * (x + 2)

test/test_jit.py

@@ -3796,6 +3796,7 @@ def f(i):
     def test_bailout_loop_carried_deps_name_clash(self):
         with enable_profiling_mode_for_profiling_tests():
             NUM_ITERATIONS = 10
+
             @torch.jit.script
             def fct_loop(z, size):
                 # type: (int, int) -> Tuple[Tensor, List[int]]
@@ -3817,6 +3818,7 @@ def fct_loop(z, size):
     def test_bailout_loop_counter_transition(self):
         with enable_profiling_mode_for_profiling_tests():
             NUM_ITERATIONS = 10
+
             @torch.jit.script
             def fct_loop(z, size):
                 # type: (int, int) -> Tuple[Tensor, List[int]]
@@ -7001,9 +7003,10 @@ def test_cast_float(x):
         self.checkScript(test_cast_float, (-1.,))
 
         with self.assertRaisesRegex(RuntimeError, r"Could not cast value of type Tuple\[int, int\] to bool"):  # noqa: W605
+
             @torch.jit.script
             def test_bad_conditional(x):
-                if (1, 2):
+                if (1, 2):  # noqa F634
                     return
                 else:
                     return 0
@@ -7669,6 +7672,7 @@ def bar(c, b):
 
     def test_error_stacktrace_interface(self):
         global IFace
+
         @torch.jit.script
         def baz(c, b):
             return c + b
@@ -8318,6 +8322,7 @@ def foo(a):
                 return 4
         self.assertEqual(foo(4), 7)
         self.assertEqual(foo(None), 4)
+
         @torch.jit.script
         def foo2(a, b):
             # type: (Optional[int], Optional[int]) -> int
@@ -16082,7 +16087,7 @@ def fn():
         def identity(x1):  # noqa: F811
             # type: (str) -> str
             pass
-        #
+
         @torch.jit._overload  # noqa: F811
         def identity(x1):  # noqa: F811
             # type: (float) -> float

test/test_jit_py3.py

@@ -16,7 +16,7 @@ def test_joined_str(self):
         def func(x):
             hello, test = "Hello", "test"
             print(f"{hello + ' ' + test}, I'm a {test}") # noqa E999
-            print(f"format blank")
+            print(f"format blank") # noqa F541
             hi = 'hi'
             print(f"stuff before {hi}")
             print(f"{hi} stuff after")
@@ -39,6 +39,7 @@ def func(x):
     @unittest.skipIf(sys.version_info[:2] < (3, 7), "`dataclasses` module not present on < 3.7")
     def test_dataclass_error(self):
         from dataclasses import dataclass
+
         @dataclass
         class NormalizationInfo(object):
             mean: float = 0.0
@@ -256,6 +257,7 @@ def __call__(self, *args) -> str:
                 return str(type(args[0]))
 
         the_class = MyPythonClass()
+
         @torch.jit.script
         def fn(x):
             return the_class(x)
@@ -424,6 +426,7 @@ def foo():
 
     def test_export_opnames_interface(self):
         global OneTwoModule
+
         @torch.jit.interface
         class OneTwoModule(nn.Module):
             def one(self, x, y):

test/test_tensorboard.py

@@ -363,28 +363,32 @@ def test_image_with_boxes(self):
                                             self))
 
     def test_image_with_one_channel(self):
-        self.assertTrue(compare_image_proto(summary.image('dummy',
-                                                    tensor_N(shape=(1, 8, 8)),
-                                                    dataformats='CHW'),
-                                                    self))  # noqa E127
+        self.assertTrue(compare_image_proto(
+            summary.image('dummy',
+                          tensor_N(shape=(1, 8, 8)),
+                          dataformats='CHW'),
+                          self))  # noqa E127
 
     def test_image_with_one_channel_batched(self):
-        self.assertTrue(compare_image_proto(summary.image('dummy',
-                                                    tensor_N(shape=(2, 1, 8, 8)),
-                                                    dataformats='NCHW'),
-                                                    self))  # noqa E127
+        self.assertTrue(compare_image_proto(
+            summary.image('dummy',
+                          tensor_N(shape=(2, 1, 8, 8)),
+                          dataformats='NCHW'),
+                          self))  # noqa E127
 
     def test_image_with_3_channel_batched(self):
-        self.assertTrue(compare_image_proto(summary.image('dummy',
-                                                    tensor_N(shape=(2, 3, 8, 8)),
-                                                    dataformats='NCHW'),
-                                                    self))  # noqa E127
+        self.assertTrue(compare_image_proto(
+            summary.image('dummy',
+                          tensor_N(shape=(2, 3, 8, 8)),
+                          dataformats='NCHW'),
+                          self))  # noqa E127
 
     def test_image_without_channel(self):
-        self.assertTrue(compare_image_proto(summary.image('dummy',
-                                                    tensor_N(shape=(8, 8)),
-                                                    dataformats='HW'),
-                                                    self))  # noqa E127
+        self.assertTrue(compare_image_proto(
+            summary.image('dummy',
+                          tensor_N(shape=(8, 8)),
+                          dataformats='HW'),
+                          self))  # noqa E127
 
     def test_video(self):
         try: