root logger

app.py

@@ -1,13 +1,5 @@
 import traceback
 import logging
-logger = logging.getLogger()
-logger.setLevel(logging.WARNING)
-if not logger.handlers:
-    handler = logging.StreamHandler()
-    handler.setFormatter(logging.Formatter("%(asctime)s [%(levelname)s] %(name)s: %(message)s"))
-    logger.addHandler(handler)
-logger.warning("here")
-
 from typing import Optional
 import spaces
 import gradio as gr
@@ -30,6 +22,14 @@ import torch
 from transformers import AutoModelForCausalLM
 from transformers import AutoProcessor 
 
+logger = logging.getLogger()
+logger.setLevel(logging.WARNING)
+if not logger.handlers:
+    handler = logging.StreamHandler()
+    handler.setFormatter(logging.Formatter("%(asctime)s [%(levelname)s] %(name)s: %(message)s"))
+    logger.addHandler(handler)
+logger.warning("here")
+
 # Define repository and local directory
 repo_id = "microsoft/OmniParser-v2.0"  # HF repo
 local_dir = "weights"  # Target local directory
@@ -105,6 +105,7 @@ def get_som_response(instruction, image_som):
         inputs = magma_processor(images=[image_som], texts=prompt, return_tensors="pt")
         inputs['pixel_values'] = inputs['pixel_values'].unsqueeze(0).to(torch.bfloat16) # Add .to(torch.bfloat16) here for explicit casting
         inputs['image_sizes'] = inputs['image_sizes'].unsqueeze(0)
+        logger.warning(inputs['pixel_values'].dtype)
         # inputs = inputs.to("cuda")
         inputs = inputs.to("cuda", dtype=torch.bfloat16)
 

util/utils.py

@@ -18,7 +18,8 @@ import numpy as np
 from matplotlib import pyplot as plt
 import easyocr
 from paddleocr import PaddleOCR
-reader = easyocr.Reader(['en'])
+# reader = easyocr.Reader(['en'])
+reader = None
 paddle_ocr = PaddleOCR(
     lang='en',  # other lang also available
     use_angle_cls=False,
@@ -520,6 +521,8 @@ def check_ocr_box(image_source: Union[str, Image.Image], display_img = True, out
     else:  # EasyOCR
         if easyocr_args is None:
             easyocr_args = {}
+        if reader is None:
+            reader = easyocr.Reader(['en'])
         result = reader.readtext(image_np, **easyocr_args)
         coord = [item[0] for item in result]
         text = [item[1] for item in result]