diff --git a/main.py b/main.py
index ff6c248e..3854e208 100644
--- a/main.py
+++ b/main.py
@@ -694,7 +694,7 @@ def init_wandb(save_dir, opt, config, group_name, name_str):
             # TODO change once leaving "swiffer" config directory
             try:
                 group_name = nowname.split(now)[-1].split("-")[1]
-            except:
+            except Exception:
                 group_name = nowname
             default_logger_cfg["params"]["group"] = group_name
             init_wandb(
@@ -842,7 +842,7 @@ def init_wandb(save_dir, opt, config, group_name, name_str):
                 print(
                     f"{k}, {data.datasets[k].__class__.__name__}, {len(data.datasets[k])}"
                 )
-        except:
+        except Exception:
             print("datasets not yet initialized.")
 
         # configure learning rate
diff --git a/scripts/demo/detect.py b/scripts/demo/detect.py
index 96e9f212..6e219204 100644
--- a/scripts/demo/detect.py
+++ b/scripts/demo/detect.py
@@ -45,7 +45,7 @@ def decode(self, cv2Image, method="dwtDct", **configs):
                 bits = embed.decode(cv2Image)
                 return self.reconstruct(bits)
 
-    except:
+    except Exception:
         raise e
 
 
diff --git a/scripts/demo/streamlit_helpers.py b/scripts/demo/streamlit_helpers.py
index 01ffe612..631c967b 100644
--- a/scripts/demo/streamlit_helpers.py
+++ b/scripts/demo/streamlit_helpers.py
@@ -489,7 +489,7 @@ def do_sample(
                     if isinstance(batch[key], torch.Tensor):
                         print(key, batch[key].shape)
                     elif isinstance(batch[key], list):
-                        print(key, [len(l) for l in batch[key]])
+                        print(key, [len(lst) for lst in batch[key]])
                     else:
                         print(key, batch[key])
                 c, uc = model.conditioner.get_unconditional_conditioning(
diff --git a/sgm/modules/attention.py b/sgm/modules/attention.py
index 15d66a36..f943bb52 100644
--- a/sgm/modules/attention.py
+++ b/sgm/modules/attention.py
@@ -9,10 +9,10 @@
 from torch import nn
 
 from ..util import default, exists
+from .diffusionmodules.util import checkpoint
 
 logger = logging.getLogger(__name__)
 
-
 if version.parse(torch.__version__) >= version.parse("2.0.0"):
     SDP_IS_AVAILABLE = True
     from torch.backends.cuda import SDPBackend, sdp_kernel
@@ -51,12 +51,10 @@
     import xformers.ops
 
     XFORMERS_IS_AVAILABLE = True
-except:
+except Exception:
     XFORMERS_IS_AVAILABLE = False
     logger.debug("no module 'xformers'. Processing without...")
 
-from .diffusionmodules.util import checkpoint
-
 
 def uniq(arr):  # TODO: this seems unused
     return {el: True for el in arr}.keys()
diff --git a/sgm/modules/diffusionmodules/model.py b/sgm/modules/diffusionmodules/model.py
index 33c25212..72786176 100644
--- a/sgm/modules/diffusionmodules/model.py
+++ b/sgm/modules/diffusionmodules/model.py
@@ -10,6 +10,8 @@
 from einops import rearrange
 from packaging import version
 
+from ...modules.attention import LinearAttention, MemoryEfficientCrossAttention
+
 logger = logging.getLogger(__name__)
 
 try:
@@ -17,12 +19,10 @@
     import xformers.ops
 
     XFORMERS_IS_AVAILABLE = True
-except:
+except Exception:
     XFORMERS_IS_AVAILABLE = False
     logger.debug("no module 'xformers'. Processing without...")
 
-from ...modules.attention import LinearAttention, MemoryEfficientCrossAttention
-
 
 def get_timestep_embedding(timesteps, embedding_dim):
     """
@@ -299,7 +299,9 @@ def make_attn(in_channels, attn_type="vanilla", attn_kwargs=None):
         )
         attn_type = "vanilla"
         attn_kwargs = None
-    logger.debug(f"making attention of type '{attn_type}' with {in_channels} in_channels")
+    logger.debug(
+        f"making attention of type '{attn_type}' with {in_channels} in_channels"
+    )
     if attn_type == "vanilla":
         assert attn_kwargs is None
         return AttnBlock(in_channels)
diff --git a/sgm/modules/diffusionmodules/openaimodel.py b/sgm/modules/diffusionmodules/openaimodel.py
index 1c87442a..0c0458df 100644
--- a/sgm/modules/diffusionmodules/openaimodel.py
+++ b/sgm/modules/diffusionmodules/openaimodel.py
@@ -649,7 +649,8 @@ def __init__(
 
         self.use_fairscale_checkpoint = False
         checkpoint_wrapper_fn = (
-            partial(checkpoint_wrapper, offload_to_cpu=offload_to_cpu)
+            # TODO: this can't work since `checkpoint_wrapper` is not defined
+            partial(checkpoint_wrapper, offload_to_cpu=offload_to_cpu)  # noqa: F821
             if self.use_fairscale_checkpoint
             else lambda x: x
         )
diff --git a/sgm/util.py b/sgm/util.py
index 3e16865a..414458c3 100644
--- a/sgm/util.py
+++ b/sgm/util.py
@@ -31,7 +31,7 @@ def get_string_from_tuple(s):
                 return t[0]
             else:
                 pass
-    except:
+    except Exception:
         pass
     return s
 
@@ -164,7 +164,9 @@ def mean_flat(tensor):
 def count_params(model, verbose=False):
     total_params = sum(p.numel() for p in model.parameters())
     if verbose:
-        logger.info(f"{model.__class__.__name__} has {total_params * 1.e-6:.2f} M params.")
+        logger.info(
+            f"{model.__class__.__name__} has {total_params * 1.e-6:.2f} M params."
+        )
     return total_params