Fix typos

Found via `codespell -S ./imaginairy/vendored`

LICENSE

@@ -1,4 +1,4 @@
-Copyright 2022 Bryce Drennan (and numerous other contributers as documented)
+Copyright 2022 Bryce Drennan (and numerous other contributors as documented)
 (for modifications on top of CompVis code)
 
 The MIT License

README.md

@@ -245,7 +245,7 @@ docker run -it --gpus all -v $HOME/.cache/huggingface:/root/.cache/huggingface -
 **5.0.0**
 - feature: 🎉 inpainting support using new inpainting model from RunwayML. It works really well! By default, the 
 inpainting model will automatically be used for any image-masking task 
-- feature: 🎉 new default sampler makes image generation mor than twice as fast
+- feature: 🎉 new default sampler makes image generation more than twice as fast
 - feature: added `DPM++ 2S a` and `DPM++ 2M` samplers.
 - feature: improve progress image logging
 - fix: fix bug with `--show-work`. fixes #84
@@ -315,7 +315,7 @@ would be uncorrelated to the rest of the surrounding image.  It created terrible
  - fix: another bfloat16 fix
 
 **1.6.1**
- - fix: make sure image tensors come to the CPU as float32 so there aren't compatability issues with non-bfloat16 cpus
+ - fix: make sure image tensors come to the CPU as float32 so there aren't compatibility issues with non-bfloat16 cpus
 
 **1.6.0**
  - fix: *maybe* address #13 with `expected scalar type BFloat16 but found Float`
@@ -361,7 +361,7 @@ would be uncorrelated to the rest of the surrounding image.  It created terrible
    - https://github.com/CompVis/stable-diffusion/pull/177
    - https://github.com/huggingface/diffusers/pull/532/files
    - https://github.com/HazyResearch/flash-attention
-   - xformers improvments https://www.photoroom.com/tech/stable-diffusion-100-percent-faster-with-memory-efficient-attention/
+   - xformers improvements https://www.photoroom.com/tech/stable-diffusion-100-percent-faster-with-memory-efficient-attention/
  - Development Environment
    - ✅ add tests
    - ✅ set up ci (test/lint/format)
@@ -458,7 +458,7 @@ would be uncorrelated to the rest of the surrounding image.  It created terrible
    - find similar images https://knn5.laion.ai/?back=https%3A%2F%2Fknn5.laion.ai%2F&index=laion5B&useMclip=false
    - https://github.com/vicgalle/stable-diffusion-aesthetic-gradients
 
-## Noteable Stable Diffusion Implementations
+## Notable Stable Diffusion Implementations
  - https://github.com/ahrm/UnstableFusion
  - https://github.com/AUTOMATIC1111/stable-diffusion-webui
  - https://github.com/blueturtleai/gimp-stable-diffusion

imaginairy/configs/stable-diffusion-v1-inpaint.yaml

@@ -17,7 +17,7 @@ model:
     scale_factor: 0.18215
     finetune_keys: null
 
-    scheduler_config: # 10000 warmup steps
+    scheduler_config: # 10000 warm-up steps
       target: ldm.lr_scheduler.LambdaLinearScheduler
       params:
         warm_up_steps: [ 2500 ] # NOTE for resuming. use 10000 if starting from scratch

imaginairy/configs/stable-diffusion-v1.yaml

@@ -16,7 +16,7 @@ model:
     monitor: val/loss_simple_ema
     scale_factor: 0.18215
 
-    scheduler_config: # 10000 warmup steps
+    scheduler_config: # 10000 warm-up steps
       target: imaginairy.lr_scheduler.LambdaLinearScheduler
       params:
         warm_up_steps: [ 10000 ]

imaginairy/modules/diffusion/ddpm.py

@@ -493,7 +493,7 @@ class DDPM(pl.LightningModule):
             target = self.get_v(x_start, noise, t)
         else:
             raise NotImplementedError(
-                f"Paramterization {self.parameterization} not yet supported"
+                f"Parameterization {self.parameterization} not yet supported"
             )
 
         loss = self.get_loss(model_out, target, mean=False).mean(dim=[1, 2, 3])
@@ -1051,7 +1051,7 @@ class LatentDiffusion(DDPM):
     def apply_model(self, x_noisy, t, cond, return_ids=False):
 
         if isinstance(cond, dict):
-            # hybrid case, cond is exptected to be a dict
+            # hybrid case, cond is expected to be a dict
             pass
         else:
             if not isinstance(cond, list):
@@ -1109,7 +1109,7 @@ class LatentDiffusion(DDPM):
                 num_downs = self.first_stage_model.encoder.num_resolutions - 1
                 rescale_latent = 2 ** (num_downs)
 
-                # get top left postions of patches as conforming for the bbbox tokenizer, therefore we
+                # get top left positions of patches as conforming for the bbbox tokenizer, therefore we
                 # need to rescale the tl patch coordinates to be in between (0,1)
                 tl_patch_coordinates = [
                     (
@@ -1175,7 +1175,7 @@ class LatentDiffusion(DDPM):
             ]
             assert not isinstance(
                 output_list[0], tuple
-            )  # todo cant deal with multiple model outputs check this never happens
+            )  # todo can't deal with multiple model outputs check this never happens
 
             o = torch.stack(output_list, axis=-1)
             o = o * weighting

imaginairy/modules/diffusion/openaimodel.py

@@ -359,7 +359,7 @@ def count_flops_attn(model, _x, y):
 
 class QKVAttentionLegacy(nn.Module):
     """
-    A module which performs QKV attention. Matches legacy QKVAttention + input/ouput heads shaping
+    A module which performs QKV attention. Matches legacy QKVAttention + input/output heads shaping
     """
 
     def __init__(self, n_heads):

imaginairy/samplers/plms.py

@@ -198,7 +198,7 @@ class PLMSSampler:
             # 2nd order Pseudo Linear Multistep (Adams-Bashforth)
             e_t_prime = (3 * noise_pred - old_eps[-1]) / 2
         elif len(old_eps) == 2:
-            # 3nd order Pseudo Linear Multistep (Adams-Bashforth)
+            # 3rd order Pseudo Linear Multistep (Adams-Bashforth)
             e_t_prime = (23 * noise_pred - 16 * old_eps[-1] + 5 * old_eps[-2]) / 12
         elif len(old_eps) >= 3:
             # 4nd order Pseudo Linear Multistep (Adams-Bashforth)

tests/test_guidance.py

@@ -78,7 +78,7 @@
 # to_tensor_tfm = transforms.ToTensor()
 #
 #
-# # mismatch of tons of image encoding / decoding / loading functions i cant be asked to clean up right now
+# # mismatch of tons of image encoding / decoding / loading functions i can't be asked to clean up right now
 #
 # def pil_to_latent(input_im):
 #     # Single image -> single latent in a batch (so size 1, 4, 64, 64)
@@ -154,7 +154,7 @@
 #             text = text[idx + 1:]
 #             # find value for weight
 #             if " " in text:
-#                 idx = text.index(" ")  # first occurence
+#                 idx = text.index(" ")  # first occurrence
 #             else:  # no space, read to end
 #                 idx = len(text)
 #             if idx != 0:
@@ -303,7 +303,7 @@
 #                 linx = np.linspace(0, 5, h // 8, endpoint=False)
 #                 liny = np.linspace(0, 5, w // 8, endpoint=False)
 #                 x, y = np.meshgrid(liny, linx)
-#                 p = [np.expand_dims(perlin(x, y, seed=i), 0) for i in range(4)]  # reproducable seed
+#                 p = [np.expand_dims(perlin(x, y, seed=i), 0) for i in range(4)]  # reproducible seed
 #                 p = np.concatenate(p, 0)
 #                 p = torch.tensor(p).unsqueeze(0).cuda()
 #                 latents = latents + (p * args.perlin_multi).to(get_device()).half()