test

app.py

@@ -12,12 +12,11 @@ import gradio as gr
 import torch
 
 
-from mmocr.apis import MMOCRInferencer
-ocr = MMOCRInferencer(det='TextSnake', rec='ABINet_Vision')
+# from mmocr.apis import MMOCRInferencer
 
-url = (
-    "https://upload.wikimedia.org/wikipedia/commons/thumb/5/5b/Draft_Marks_on_the_Bow_of_Kruzenshtern_Port_of_Tallinn_16_July_2011.jpg/1600px-Draft_Marks_on_the_Bow_of_Kruzenshtern_Port_of_Tallinn_16_July_2011.jpg"
-)
+# ocr = MMOCRInferencer(det="TextSnake", rec="ABINet_Vision")
+
+url = "https://upload.wikimedia.org/wikipedia/commons/thumb/5/5b/Draft_Marks_on_the_Bow_of_Kruzenshtern_Port_of_Tallinn_16_July_2011.jpg/1600px-Draft_Marks_on_the_Bow_of_Kruzenshtern_Port_of_Tallinn_16_July_2011.jpg"
 path_input = "./example1.jpg"
 urllib.request.urlretrieve(url, filename=path_input)
 
@@ -27,23 +26,25 @@ urllib.request.urlretrieve(url, filename=path_input)
 
 
 path_img_output_folder = "./demo-out"
-if not os.path.exists(path_img_output_folder):
-    os.makedirs(path_img_output_folder)
+# if not os.path.exists(path_img_output_folder):
+#     os.makedirs(path_img_output_folder)
 
 path_img_input_folder = "./demo-input"
-if not os.path.exists(path_img_input_folder):
-    os.makedirs(path_img_input_folder)
+# if not os.path.exists(path_img_input_folder):
+#     os.makedirs(path_img_input_folder)
+
 
 def do_process(img):
-    img_name = 'tmp.jpg'
+    img_name = "tmp.jpg"
     path_input = os.path.join(path_img_input_folder, img_name)
-    path_output = os.path.join(path_img_output_folder, 'vis',img_name)
+    path_output = os.path.join(path_img_output_folder, "vis", img_name)
     img.save(path_input)
     img.save(path_output)
     # result = ocr(path_input, out_dir=path_img_output_folder, save_vis=True)
     img_res = Image(filename=path_output)
     return img_res
 
+
 input_im = gr.inputs.Image(
     shape=None, image_mode="RGB", invert_colors=False, source="upload", type="pil"
 )
@@ -53,7 +54,9 @@ output_img = gr.outputs.Image(label="Output of Integrated Gradients", type="pil"
 # output_label = gr.outputs.Label(label="Classification results", num_top_classes=3)
 
 title = "Reading draught marks"
-description = "Playground: Reading draught marks using pre-trained models. Tools: MMOCR, Gradio."
+description = (
+    "Playground: Reading draught marks using pre-trained models. Tools: MMOCR, Gradio."
+)
 examples = [["./example1.jpg"], ["./example2.jpg"]]
 article = "<p style='text-align: center'><a href='https://github.com/mawady' target='_blank'>By Dr. Mohamed Elawady</a></p>"
 iface = gr.Interface(