diff --git a/lama_cleaner/__init__.py b/lama_cleaner/__init__.py
index a1ce1a1..82a0711 100644
--- a/lama_cleaner/__init__.py
+++ b/lama_cleaner/__init__.py
@@ -2,11 +2,12 @@ import os
 
 os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
 
-import warnings
+import warnings  # noqa: E402
 
-warnings.simplefilter("ignore", UserWarning)
+warnings.filterwarnings("ignore", category=UserWarning, module="pydantic")
+warnings.filterwarnings("ignore", category=UserWarning, module="lama_cleaner")
 
-from lama_cleaner.parse_args import parse_args
+from lama_cleaner.parse_args import parse_args  # noqa: E402
 
 
 def entry_point():
diff --git a/mobile_sam/__init__.py b/mobile_sam/__init__.py
index 7140702..6e2138b 100644
--- a/mobile_sam/__init__.py
+++ b/mobile_sam/__init__.py
@@ -4,16 +4,14 @@
 # This source code is licensed under the license found in the
 # LICENSE file in the root directory of this source tree.
 
-from .build_sam import (
-    build_sam,
-    build_sam_vit_h,
-    build_sam_vit_l,
-    build_sam_vit_b,
-    build_sam_vit_t,
-    sam_model_registry,
-)
-from .predictor import SamPredictor
-from .automatic_mask_generator import SamAutomaticMaskGenerator
+import warnings
+
+warnings.filterwarnings("ignore", category=UserWarning, module="mobile_sam")
+
+from .automatic_mask_generator import SamAutomaticMaskGenerator  # noqa: E402
+from .build_sam import (build_sam, build_sam_vit_b, build_sam_vit_h, build_sam_vit_l,  # noqa: E402
+                        build_sam_vit_t, sam_model_registry)
+from .predictor import SamPredictor  # noqa: E402
 
 __all__ = [
     "build_sam",
diff --git a/mobile_sam/modeling/tiny_vit_sam.py b/mobile_sam/modeling/tiny_vit_sam.py
index 79205db..f67e3ba 100644
--- a/mobile_sam/modeling/tiny_vit_sam.py
+++ b/mobile_sam/modeling/tiny_vit_sam.py
@@ -8,7 +8,6 @@
 # --------------------------------------------------------
 
 import itertools
-import warnings
 from typing import Tuple
 
 import torch
@@ -19,8 +18,6 @@ from timm.models.layers import DropPath as TimmDropPath
 from timm.models.layers import to_2tuple, trunc_normal_
 from timm.models.registry import register_model
 
-warnings.simplefilter("ignore", category=UserWarning)
-
 
 class Conv2d_BN(torch.nn.Sequential):
     def __init__(self, a, b, ks=1, stride=1, pad=0, dilation=1,