diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 881fedb4d..8be6efe33 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -22,14 +22,14 @@ repos:
       - id: detect-private-key
 
   - repo: https://github.com/asottile/pyupgrade
-    rev: v3.4.0
+    rev: v3.8.0
     hooks:
       - id: pyupgrade
         args: [--py38-plus]
         name: Upgrade code
 
   - repo: https://github.com/PyCQA/docformatter
-    rev: v1.7.1
+    rev: v1.7.3
     hooks:
       - id: docformatter
         args: [--in-place, --wrap-summaries=115, --wrap-descriptions=120]
@@ -56,7 +56,7 @@ repos:
         name: Format code
 
   - repo: https://github.com/asottile/yesqa
-    rev: v1.4.0
+    rev: v1.5.0
     hooks:
       - id: yesqa
         additional_dependencies:
@@ -64,8 +64,8 @@ repos:
           - flake8-bandit
           - pep8-naming
 
-  - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: v0.0.272
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.0.276
     hooks:
       - id: ruff
         args: ["--fix"]
diff --git a/setup.py b/setup.py
index 5ba698962..7e88ff306 100755
--- a/setup.py
+++ b/setup.py
@@ -102,13 +102,12 @@ def _load_readme_description(path_dir: str, homepage: str, ver: str) -> str:
     skip_begin = r"<!-- following section will be skipped from PyPI description -->"
     skip_end = r"<!-- end skipping PyPI description -->"
     # todo: wrap content as commented description
-    text = re.sub(rf"{skip_begin}.+?{skip_end}", "<!--  -->", text, flags=re.IGNORECASE + re.DOTALL)
+    return re.sub(rf"{skip_begin}.+?{skip_end}", "<!--  -->", text, flags=re.IGNORECASE + re.DOTALL)
 
     # # https://github.com/Borda/pytorch-lightning/releases/download/1.1.0a6/codecov_badge.png
     # github_release_url = os.path.join(homepage, "releases", "download", ver)
     # # download badge and replace url with local file
     # text = _parse_for_badge(text, github_release_url)
-    return text
 
 
 def _prepare_extras():
diff --git a/src/pl_bolts/losses/self_supervised_learning.py b/src/pl_bolts/losses/self_supervised_learning.py
index ddf3d41ec..04571121f 100644
--- a/src/pl_bolts/losses/self_supervised_learning.py
+++ b/src/pl_bolts/losses/self_supervised_learning.py
@@ -83,8 +83,7 @@ def forward(self, z):
                 if not torch.isnan(loss):
                     losses.append(loss)
 
-        loss = torch.stack(losses).sum()
-        return loss
+        return torch.stack(losses).sum()
 
 
 @under_review()
@@ -255,8 +254,7 @@ def feat_size_w_mask(self, w, feature_map):
         for i in range(w):
             for j in range(w):
                 masks_r5[i, j, 0, i, j] = 1
-        masks_r5 = masks_r5.reshape(-1, 1, w, w)
-        return masks_r5
+        return masks_r5.reshape(-1, 1, w, w)
 
     def _sample_src_ftr(self, r_cnv, masks):
         # get feature dimensions
diff --git a/src/pl_bolts/models/autoencoders/components.py b/src/pl_bolts/models/autoencoders/components.py
index 4868bb5b0..b5c013d15 100644
--- a/src/pl_bolts/models/autoencoders/components.py
+++ b/src/pl_bolts/models/autoencoders/components.py
@@ -74,9 +74,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 @under_review()
@@ -117,8 +115,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-        return out
+        return self.relu(out)
 
 
 @under_review()
@@ -150,9 +147,7 @@ def forward(self, x):
             identity = self.upsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 @under_review()
@@ -192,8 +187,7 @@ def forward(self, x):
             identity = self.upsample(x)
 
         out += identity
-        out = self.relu(out)
-        return out
+        return self.relu(out)
 
 
 @under_review()
@@ -252,8 +246,7 @@ def forward(self, x):
         x = self.layer4(x)
 
         x = self.avgpool(x)
-        x = torch.flatten(x, 1)
-        return x
+        return torch.flatten(x, 1)
 
 
 @under_review()
@@ -325,8 +318,7 @@ def forward(self, x):
         x = self.layer4(x)
         x = self.upscale(x)
 
-        x = self.conv1(x)
-        return x
+        return self.conv1(x)
 
 
 @under_review()
diff --git a/src/pl_bolts/models/detection/yolo/utils.py b/src/pl_bolts/models/detection/yolo/utils.py
index 2febe1620..0aab13d94 100644
--- a/src/pl_bolts/models/detection/yolo/utils.py
+++ b/src/pl_bolts/models/detection/yolo/utils.py
@@ -148,8 +148,7 @@ def box_size_ratio(wh1: Tensor, wh2: Tensor) -> Tensor:
     """
     wh_ratio = wh1[:, None, :] / wh2[None, :, :]  # [M, N, 2]
     wh_ratio = torch.max(wh_ratio, 1.0 / wh_ratio)
-    wh_ratio = wh_ratio.max(2).values  # [M, N]
-    return wh_ratio
+    return wh_ratio.max(2).values  # [M, N]
 
 
 @torch.jit.script
diff --git a/src/pl_bolts/models/detection/yolo/yolo_module.py b/src/pl_bolts/models/detection/yolo/yolo_module.py
index eb7328085..342f9f5ca 100644
--- a/src/pl_bolts/models/detection/yolo/yolo_module.py
+++ b/src/pl_bolts/models/detection/yolo/yolo_module.py
@@ -314,8 +314,7 @@ class labels.
         """
         images, _ = batch
         detections = self(images)
-        detections = self.process_detections(detections)
-        return detections
+        return self.process_detections(detections)
 
     def infer(self, image: Tensor) -> PRED:
         """Feeds an image to the network and returns the detected bounding boxes, confidence scores, and class
diff --git a/src/pl_bolts/models/gans/basic/components.py b/src/pl_bolts/models/gans/basic/components.py
index 136a0b547..0ac9fe3aa 100644
--- a/src/pl_bolts/models/gans/basic/components.py
+++ b/src/pl_bolts/models/gans/basic/components.py
@@ -20,8 +20,7 @@ def forward(self, z):
         z = F.leaky_relu(self.fc2(z), 0.2)
         z = F.leaky_relu(self.fc3(z), 0.2)
         img = torch.tanh(self.fc4(z))
-        img = img.view(img.size(0), *self.img_shape)
-        return img
+        return img.view(img.size(0), *self.img_shape)
 
 
 class Discriminator(nn.Module):
diff --git a/src/pl_bolts/models/gans/srgan/components.py b/src/pl_bolts/models/gans/srgan/components.py
index 433737125..99ad9d2e6 100644
--- a/src/pl_bolts/models/gans/srgan/components.py
+++ b/src/pl_bolts/models/gans/srgan/components.py
@@ -77,8 +77,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
         x_res = self.input_block(x)
         x = x_res + self.res_blocks(x_res)
         x = self.ps_blocks(x)
-        x = self.output_block(x)
-        return x
+        return self.output_block(x)
 
 
 @under_review()
@@ -131,8 +130,7 @@ def _make_conv_block(
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
         x = self.conv_blocks(x)
-        x = self.mlp(x)
-        return x
+        return self.mlp(x)
 
 
 @under_review()
diff --git a/src/pl_bolts/models/mnist_module.py b/src/pl_bolts/models/mnist_module.py
index f322f6df8..502617740 100644
--- a/src/pl_bolts/models/mnist_module.py
+++ b/src/pl_bolts/models/mnist_module.py
@@ -39,8 +39,7 @@ def __init__(self, hidden_dim: int = 128, learning_rate: float = 1e-3, **kwargs:
     def forward(self, x: Tensor) -> Tensor:
         out = x.view(x.size(0), -1)
         out = torch.relu(self.l1(out))
-        out = torch.relu(self.l2(out))
-        return out
+        return torch.relu(self.l2(out))
 
     def shared_step(self, batch: Any, batch_idx: int, step: str) -> Tensor:
         x, y = batch
diff --git a/src/pl_bolts/models/rl/advantage_actor_critic_model.py b/src/pl_bolts/models/rl/advantage_actor_critic_model.py
index 6e3d157df..43418378b 100644
--- a/src/pl_bolts/models/rl/advantage_actor_critic_model.py
+++ b/src/pl_bolts/models/rl/advantage_actor_critic_model.py
@@ -179,9 +179,7 @@ def compute_returns(
             returns.append(g)
 
         # reverse list and stop the gradients
-        returns = torch.tensor(returns[::-1])
-
-        return returns
+        return torch.tensor(returns[::-1])
 
     def loss(
         self,
diff --git a/src/pl_bolts/models/rl/vanilla_policy_gradient_model.py b/src/pl_bolts/models/rl/vanilla_policy_gradient_model.py
index 344ebddd1..de463d351 100644
--- a/src/pl_bolts/models/rl/vanilla_policy_gradient_model.py
+++ b/src/pl_bolts/models/rl/vanilla_policy_gradient_model.py
@@ -172,9 +172,7 @@ def compute_returns(self, rewards):
             returns.insert(0, reward)
 
         returns = torch.tensor(returns)
-        returns = (returns - returns.mean()) / (returns.std() + self.eps)
-
-        return returns
+        return (returns - returns.mean()) / (returns.std() + self.eps)
 
     def loss(self, states, actions, scaled_rewards) -> Tensor:
         """Calculates the loss for VPG.
diff --git a/src/pl_bolts/models/self_supervised/cpc/networks.py b/src/pl_bolts/models/self_supervised/cpc/networks.py
index 7acdb7906..dd67d8a4c 100644
--- a/src/pl_bolts/models/self_supervised/cpc/networks.py
+++ b/src/pl_bolts/models/self_supervised/cpc/networks.py
@@ -114,8 +114,7 @@ def _make_layer(self, sample_batch, block, planes, blocks, stride=1, dilate=Fals
 
     def flatten(self, x):
         x = x.view(self.batch_size, -1)
-        x = F.avg_pool1d(x.unsqueeze(1), 4).squeeze(1)
-        return x
+        return F.avg_pool1d(x.unsqueeze(1), 4).squeeze(1)
 
     def forward(self, x):
         x = self.conv1(x)
@@ -126,9 +125,7 @@ def forward(self, x):
         x = self.layer1(x)
         x = self.layer2(x)
         x = self.layer3(x)
-        x = self.layer4(x)
-
-        return x
+        return self.layer4(x)
 
 
 @under_review()
@@ -202,9 +199,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 @under_review()
diff --git a/src/pl_bolts/models/self_supervised/resnets.py b/src/pl_bolts/models/self_supervised/resnets.py
index 04b90a150..60dcfc8f5 100644
--- a/src/pl_bolts/models/self_supervised/resnets.py
+++ b/src/pl_bolts/models/self_supervised/resnets.py
@@ -88,9 +88,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 @under_review()
@@ -133,9 +131,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 @under_review()
diff --git a/src/pl_bolts/models/self_supervised/simclr/simclr_module.py b/src/pl_bolts/models/self_supervised/simclr/simclr_module.py
index b2f61efd0..d17ddf7a1 100644
--- a/src/pl_bolts/models/self_supervised/simclr/simclr_module.py
+++ b/src/pl_bolts/models/self_supervised/simclr/simclr_module.py
@@ -27,9 +27,7 @@ def forward(ctx, tensor):
         gathered_tensor = [torch.zeros_like(tensor) for _ in range(torch.distributed.get_world_size())]
 
         torch.distributed.all_gather(gathered_tensor, tensor)
-        gathered_tensor = torch.cat(gathered_tensor, 0)
-
-        return gathered_tensor
+        return torch.cat(gathered_tensor, 0)
 
     @staticmethod
     def backward(ctx, grad_output):
diff --git a/src/pl_bolts/models/self_supervised/swav/swav_resnet.py b/src/pl_bolts/models/self_supervised/swav/swav_resnet.py
index 9c702756c..65b09dbe9 100644
--- a/src/pl_bolts/models/self_supervised/swav/swav_resnet.py
+++ b/src/pl_bolts/models/self_supervised/swav/swav_resnet.py
@@ -67,9 +67,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 class Bottleneck(nn.Module):
@@ -120,9 +118,7 @@ def forward(self, x):
             identity = self.downsample(x)
 
         out += identity
-        out = self.relu(out)
-
-        return out
+        return self.relu(out)
 
 
 class ResNet(nn.Module):
@@ -294,9 +290,7 @@ def forward_backbone(self, x):
             return x
 
         x = self.avgpool(x)
-        x = torch.flatten(x, 1)
-
-        return x
+        return torch.flatten(x, 1)
 
     def forward_head(self, x):
         if self.projection_head is not None:
diff --git a/src/pl_bolts/models/vision/image_gpt/gpt2.py b/src/pl_bolts/models/vision/image_gpt/gpt2.py
index 08048f189..4d6b4ec2e 100644
--- a/src/pl_bolts/models/vision/image_gpt/gpt2.py
+++ b/src/pl_bolts/models/vision/image_gpt/gpt2.py
@@ -26,8 +26,7 @@ def forward(self, x):
         a, _ = self.attn(x, x, x, attn_mask=attn_mask, need_weights=False)
         x = x + a
         m = self.mlp(self.ln_2(x))
-        x = x + m
-        return x
+        return x + m
 
 
 @under_review()
diff --git a/src/pl_bolts/models/vision/image_gpt/igpt_module.py b/src/pl_bolts/models/vision/image_gpt/igpt_module.py
index b491e0f20..89c48c536 100644
--- a/src/pl_bolts/models/vision/image_gpt/igpt_module.py
+++ b/src/pl_bolts/models/vision/image_gpt/igpt_module.py
@@ -13,8 +13,7 @@
 def _shape_input(x):
     """Shape batch of images for input into GPT2 model."""
     x = x.view(x.shape[0], -1)  # flatten images into sequences
-    x = x.transpose(0, 1).contiguous()  # to shape [seq len, batch]
-    return x
+    return x.transpose(0, 1).contiguous()  # to shape [seq len, batch]
 
 
 @under_review()
diff --git a/src/pl_bolts/models/vision/pixel_cnn.py b/src/pl_bolts/models/vision/pixel_cnn.py
index ef66e3a24..89237195f 100644
--- a/src/pl_bolts/models/vision/pixel_cnn.py
+++ b/src/pl_bolts/models/vision/pixel_cnn.py
@@ -1,8 +1,7 @@
 """PixelCNN.
 
 Implemented by: William Falcon Reference
-: https: //arxiv.org/pdf/1905.09272.pdf (page 15
-Accessed: May 14, 2020.
+: https: //arxiv.org/pdf/1905.09272.pdf (page 15 Accessed: May 14, 2020.
 """
 from torch import nn
 from torch.nn import functional as F  # noqa: N812
diff --git a/src/pl_bolts/transforms/self_supervised/moco_transforms.py b/src/pl_bolts/transforms/self_supervised/moco_transforms.py
index ec68cf0c4..f6117c39d 100644
--- a/src/pl_bolts/transforms/self_supervised/moco_transforms.py
+++ b/src/pl_bolts/transforms/self_supervised/moco_transforms.py
@@ -255,5 +255,4 @@ def __init__(self, sigma: List[float] = [0.1, 2.0]) -> None:
 
     def __call__(self, x):
         sigma = random.uniform(self.sigma[0], self.sigma[1])  # noqa: S311
-        x = x.filter(ImageFilter.GaussianBlur(radius=sigma))
-        return x
+        return x.filter(ImageFilter.GaussianBlur(radius=sigma))
diff --git a/src/pl_bolts/transforms/self_supervised/ssl_transforms.py b/src/pl_bolts/transforms/self_supervised/ssl_transforms.py
index 99fec9e1c..8db1069c3 100644
--- a/src/pl_bolts/transforms/self_supervised/ssl_transforms.py
+++ b/src/pl_bolts/transforms/self_supervised/ssl_transforms.py
@@ -49,10 +49,9 @@ def __call__(self, old_image):
         new_image.paste(flipped_both, (xpad - xsize + 1, ypad + ysize - 1))
         new_image.paste(flipped_both, (xpad + xsize - 1, ypad + ysize - 1))
 
-        new_image = new_image.crop(
+        return new_image.crop(
             (xpad - xtranslation, ypad - ytranslation, xpad + xsize - xtranslation, ypad + ysize - ytranslation)
         )
-        return new_image
 
 
 @under_review()
@@ -75,6 +74,4 @@ def __call__(self, x):
         # reshape to have (b x patches, c, h, w)
         x = x.view(-1, c, self.patch_size, self.patch_size)
 
-        x = x.squeeze(0)
-
-        return x
+        return x.squeeze(0)
diff --git a/tests/callbacks/test_data_monitor.py b/tests/callbacks/test_data_monitor.py
index ce51bd7ce..3d3ddc0c4 100644
--- a/tests/callbacks/test_data_monitor.py
+++ b/tests/callbacks/test_data_monitor.py
@@ -131,8 +131,7 @@ def forward(self, x):
         self.layer2_input = x
         x = self.layer2(x)
         self.layer2_output = x
-        x = torch.relu(x - 2)
-        return x
+        return torch.relu(x - 2)
 
 
 @mock.patch("pl_bolts.callbacks.data_monitor.ModuleDataMonitor.log_histogram")