fix: cleanup logging - remove unnecessary version checks

docs/changelog.md

@@ -4,11 +4,6 @@
 
 - feature: integrates spandrel for upscaling 
 
-**14.1.1**
-
-- tests: add installation tests for windows, mac, and conda
-- fix: dependency issues
-
 **14.2.0**
 - 🎉 feature: add image prompt support via `--image-prompt` and `--image-prompt-strength`
 

imaginairy/cli/upscale.py

@@ -62,7 +62,6 @@ def upscale_cmd(
     """
     Upscale an image 4x using AI.
     """
-    from tqdm import tqdm
 
     from imaginairy.enhancers.face_restoration_codeformer import enhance_faces
     from imaginairy.enhancers.upscale import upscale_image, upscale_model_lookup
@@ -92,15 +91,15 @@ def upscale_cmd(
     elif format_template == "DEFAULT":
         format_template = DEFAULT_FORMAT_TEMPLATE
 
-    for p in tqdm(image_filepaths):
-        savepath = os.path.join(outdir, os.path.basename(p))
+    for n, p in enumerate(image_filepaths):
         if p.startswith("http"):
             img = LazyLoadingImage(url=p)
         else:
             img = LazyLoadingImage(filepath=p)
+        orig_height = img.height
         for model in upscale_model:
             logger.info(
-                f"Upscaling {p} from {img.width}x{img.height} to {img.width * 4}x{img.height * 4} and saving it to {savepath}"
+                f"Upscaling ({n + 1}/{len(image_filepaths)}) {p} ({img.width}x{img.height})..."
             )
 
             img = upscale_image(img, model)
@@ -131,4 +130,7 @@ def upscale_cmd(
             new_file_name = format_filename(format_template, new_file_name_data)
             new_file_path = os.path.join(outdir, new_file_name)
             img.save(new_file_path)
-            print(f"Saved to {new_file_path}")
+            scale = int(img.height / orig_height)
+            logger.info(
+                f"Upscaled {scale}x to {img.width}x{img.height} and saved to {new_file_path}"
+            )

imaginairy/enhancers/upscale.py

@@ -69,10 +69,7 @@ def upscale_image(
         model_path = upscaler_model
 
     model = ModelLoader().load_from_file(model_path)
-    logger.info(
-        f"Upscaling from {img.width}x{img.height} to {img.width * model.scale}x{img.height * model.scale}"
-    )
-    print(f"Upscaling image with model {model.architecture}@{upscaler_model}")
+    logger.debug(f"Upscaling image with model {model.architecture}@{upscaler_model}")
 
     assert isinstance(model, ImageModelDescriptor)
 

imaginairy/utils/tile_up.py

@@ -40,7 +40,7 @@ def tile_process(
     output = img.new_zeros(output_shape)
     tiles_x = math.ceil(width / tile_size)
     tiles_y = math.ceil(height / tile_size)
-    logger.info(f"Tiling with {tiles_x}x{tiles_y} ({tiles_x*tiles_y}) tiles")
+    logger.debug(f"Tiling with {tiles_x}x{tiles_y} ({tiles_x*tiles_y}) tiles")
 
     for y in range(tiles_y):
         for x in range(tiles_x):
@@ -79,13 +79,13 @@ def tile_process(
             )
 
             # Place the processed tile in the output image
-            output[
-                :, :, output_start_y:output_end_y, output_start_x:output_end_x
-            ] = output_tile[
-                :,
-                :,
-                tile_output_start_y:tile_output_end_y,
-                tile_output_start_x:tile_output_end_x,
-            ]
+            output[:, :, output_start_y:output_end_y, output_start_x:output_end_x] = (
+                output_tile[
+                    :,
+                    :,
+                    tile_output_start_y:tile_output_end_y,
+                    tile_output_start_x:tile_output_end_x,
+                ]
+            )
 
     return output

imaginairy/vendored/clip/clip.py

@@ -8,7 +8,6 @@
 
 import torch
 from PIL import Image
-from pkg_resources import packaging
 from torchvision.transforms import CenterCrop, Compose, Normalize, Resize, ToTensor
 from tqdm import tqdm
 
@@ -23,9 +22,6 @@
     BICUBIC = Image.BICUBIC
 
 
-if packaging.version.parse(torch.__version__) < packaging.version.parse("1.7.1"):
-    warnings.warn("PyTorch version 1.7.1 or higher is recommended")
-
 
 __all__ = ["available_models", "load", "tokenize"]
 _tokenizer = _Tokenizer()
@@ -272,10 +268,7 @@ def tokenize(
     sot_token = _tokenizer.encoder["<|startoftext|>"]
     eot_token = _tokenizer.encoder["<|endoftext|>"]
     all_tokens = [[sot_token] + _tokenizer.encode(text) + [eot_token] for text in texts]
-    if packaging.version.parse(torch.__version__) < packaging.version.parse("1.8.0"):
-        result = torch.zeros(len(all_tokens), context_length, dtype=torch.long)
-    else:
-        result = torch.zeros(len(all_tokens), context_length, dtype=torch.int)
+    result = torch.zeros(len(all_tokens), context_length, dtype=torch.int)
 
     for i, tokens in enumerate(all_tokens):
         if len(tokens) > context_length: