resolve merge conflicts

invokeai/backend/util/__init__.py

@@ -1,7 +1,7 @@
 """
 Initialization file for invokeai.backend.util
 """
-from .devices import (
+from .devices import (  # noqa: F401
     CPU_DEVICE,
     CUDA_DEVICE,
     MPS_DEVICE,
@@ -10,6 +10,11 @@ from .devices import (
     normalize_device,
     torch_dtype,
 )
-from .log import write_log
-from .util import ask_user, download_with_resume, instantiate_from_config, url_attachment_name, Chdir
+from .log import write_log  # noqa: F401
+from .util import (ask_user,
+                   download_with_resume,
+                   instantiate_from_config,
+                   url_attachment_name,
+                   Chdir,
+                   )
 from .attention import auto_detect_slice_size

invokeai/backend/util/hotfixes.py

@@ -25,10 +25,15 @@ from diffusers.models.unet_2d_condition import UNet2DConditionModel
 import diffusers
 from diffusers.models.controlnet import ControlNetConditioningEmbedding, ControlNetOutput, zero_module
 
+from invokeai.backend.util.logging import InvokeAILogger
+
 # TODO: create PR to diffusers
 # Modified ControlNetModel with encoder_attention_mask argument added
 
 
+logger = InvokeAILogger.getLogger(__name__)
+
+
 class ControlNetModel(ModelMixin, ConfigMixin, FromOriginalControlnetMixin):
     """
     A ControlNet model.
@@ -111,7 +116,7 @@ class ControlNetModel(ModelMixin, ConfigMixin, FromOriginalControlnetMixin):
             "DownBlock2D",
         ),
         only_cross_attention: Union[bool, Tuple[bool]] = False,
-        block_out_channels: Tuple[int] = (320, 640, 1280, 1280),
+        block_out_channels: Tuple[int, ...] = (320, 640, 1280, 1280),
         layers_per_block: int = 2,
         downsample_padding: int = 1,
         mid_block_scale_factor: float = 1,

invokeai/backend/util/log.py

@@ -27,8 +27,8 @@ def write_log_message(results, output_cntr):
     log_lines = [f"{path}: {prompt}\n" for path, prompt in results]
     if len(log_lines) > 1:
         subcntr = 1
-        for l in log_lines:
-            print(f"[{output_cntr}.{subcntr}] {l}", end="")
+        for ll in log_lines:
+            print(f"[{output_cntr}.{subcntr}] {ll}", end="")
             subcntr += 1
     else:
         print(f"[{output_cntr}] {log_lines[0]}", end="")

invokeai/backend/util/logging.py

@@ -182,13 +182,13 @@ import urllib.parse
 from abc import abstractmethod
 from pathlib import Path
 
-from invokeai.app.services.config import InvokeAIAppConfig, get_invokeai_config
+from invokeai.app.services.config import InvokeAIAppConfig
 
 try:
     import syslog
 
     SYSLOG_AVAILABLE = True
-except:
+except ImportError:
     SYSLOG_AVAILABLE = False
 
 
@@ -417,7 +417,7 @@ class InvokeAILogger(object):
                     syslog_args["socktype"] = _SOCK_MAP[arg_value[0]]
                 else:
                     syslog_args["address"] = arg_name
-        except:
+        except Exception:
             raise ValueError(f"{args} is not a value argument list for syslog logging")
         return logging.handlers.SysLogHandler(**syslog_args)
 

invokeai/backend/util/mps_fixes.py

@@ -191,7 +191,7 @@ class ChunkedSlicedAttnProcessor:
         assert value.shape[0] == 1
         assert hidden_states.shape[0] == 1
 
-        dtype = query.dtype
+        # dtype = query.dtype
         if attn.upcast_attention:
             query = query.float()
             key = key.float()

invokeai/backend/util/util.py

@@ -84,7 +84,7 @@ def count_params(model, verbose=False):
 
 
 def instantiate_from_config(config, **kwargs):
-    if not "target" in config:
+    if "target" not in config:
         if config == "__is_first_stage__":
             return None
         elif config == "__is_unconditional__":
@@ -234,16 +234,17 @@ def rand_perlin_2d(shape, res, device, fade=lambda t: 6 * t**5 - 15 * t**4 + 10
         .repeat_interleave(d[1], 1)
     )
 
-    dot = lambda grad, shift: (
-        torch.stack(
-            (
-                grid[: shape[0], : shape[1], 0] + shift[0],
-                grid[: shape[0], : shape[1], 1] + shift[1],
-            ),
-            dim=-1,
-        )
-        * grad[: shape[0], : shape[1]]
-    ).sum(dim=-1)
+    def dot(grad, shift):
+        return (
+            torch.stack(
+                (
+                    grid[: shape[0], : shape[1], 0] + shift[0],
+                    grid[: shape[0], : shape[1], 1] + shift[1],
+                ),
+                dim=-1,
+            )
+            * grad[: shape[0], : shape[1]]
+        ).sum(dim=-1)
 
     n00 = dot(tile_grads([0, -1], [0, -1]), [0, 0]).to(device)
     n10 = dot(tile_grads([1, None], [0, -1]), [-1, 0]).to(device)
@@ -287,7 +288,7 @@ def download_with_resume(url: str, dest: Path, access_token: str = None) -> Path
     if dest.is_dir():
         try:
             file_name = re.search('filename="(.+)"', resp.headers.get("Content-Disposition")).group(1)
-        except:
+        except AttributeError:
             file_name = os.path.basename(url)
         dest = dest / file_name
     else:
@@ -342,7 +343,7 @@ def url_attachment_name(url: str) -> dict:
         resp = requests.get(url, stream=True)
         match = re.search('filename="(.+)"', resp.headers.get("Content-Disposition"))
         return match.group(1)
-    except:
+    except Exception:
         return None