feature: run face enhancement on the GPU

Should run 10x faster

imaginairy/api.py

@@ -401,9 +401,13 @@ def imagine(
                     if not safety_score.is_filtered:
                         if prompt.fix_faces:
                             logger.info("Fixing 😊 's in 🖼  using CodeFormer...")
-                            img = enhance_faces(img, fidelity=prompt.fix_faces_fidelity)
+                            with lc.timing("face enhancement"):
+                                img = enhance_faces(
+                                    img, fidelity=prompt.fix_faces_fidelity
+                                )
                         if prompt.upscale:
                             logger.info("Upscaling 🖼  using real-ESRGAN...")
+                            with lc.timing("upscaling"):
                                 upscaled_img = upscale_image(img)
 
                         # put the newly generated patch back into the original, full size image

imaginairy/enhancers/face_restoration_codeformer.py

@@ -13,6 +13,9 @@ from imaginairy.vendored.codeformer.codeformer_arch import CodeFormer
 
 logger = logging.getLogger(__name__)
 
+face_restore_device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+half_mode = face_restore_device == "cuda"
+
 
 @lru_cache
 def codeformer_model():
@@ -22,12 +25,14 @@ def codeformer_model():
         n_head=8,
         n_layers=9,
         connect_list=["32", "64", "128", "256"],
-    ).to("cpu")
+    ).to(face_restore_device)
     url = "https://github.com/sczhou/CodeFormer/releases/download/v0.1.0/codeformer.pth"
     ckpt_path = get_cached_url_path(url)
     checkpoint = torch.load(ckpt_path)["params_ema"]
     model.load_state_dict(checkpoint)
     model.eval()
+    if half_mode:
+        model = model.half()
     return model
 
 
@@ -39,7 +44,6 @@ def face_restore_helper():
     FaceRestoreHelper loads a model internally so we need to cache it
     or we end up with a memory leak
     """
-    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     face_helper = FaceRestoreHelper(
         upscale_factor=1,
         face_size=512,
@@ -47,7 +51,7 @@ def face_restore_helper():
         det_model="retinaface_resnet50",
         save_ext="png",
         use_parse=True,
-        device=device,
+        device=face_restore_device,
     )
     return face_helper
 
@@ -77,16 +81,17 @@ def enhance_faces(img, fidelity=0):
         # prepare data
         cropped_face_t = img2tensor(cropped_face / 255.0, bgr2rgb=True, float32=True)
         normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True)
-        cropped_face_t = cropped_face_t.unsqueeze(0).to("cpu")
+        cropped_face_t = cropped_face_t.unsqueeze(0).to(face_restore_device)
 
         try:
             with torch.no_grad():
+
                 output = net(cropped_face_t, w=fidelity, adain=True)[0]  # noqa
                 restored_face = tensor2img(output, rgb2bgr=True, min_max=(-1, 1))
             del output
             torch.cuda.empty_cache()
         except Exception as error:  # noqa
-            logger.error(f"\tFailed inference for CodeFormer: {error}")
+            logger.exception(f"\tFailed inference for CodeFormer: {error}")
             restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1))
 
         restored_face = restored_face.astype("uint8")

imaginairy/train.py

@@ -382,6 +382,11 @@ def train_diffusion_model(
     accumulate_grad_batches=32,
     resume=None,
 ):
+    """
+    Train a diffusion model on a single concept.
+
+    accumulate_grad_batches used to simulate a bigger batch size - https://arxiv.org/pdf/1711.00489.pdf
+    """
     batch_size = 1
     seed = 23
     num_workers = 1