Update app.py

app.py

@@ -14,6 +14,7 @@ import torch
 from PIL import Image
 from transformers import DetrImageProcessor, TableTransformerForObjectDetection
 from paddleocr import PaddleOCR
+import gradio as gr
 
 ocr = PaddleOCR(use_angle_cls=True, lang="en", use_gpu=False, ocr_version='PP-OCRv4')
 
@@ -498,6 +499,30 @@ class TableExtractionPipeline():
         c3.markdown(href, unsafe_allow_html=True)
 
         return df
+    def extract_table_json(img, td_th=0.8, tsr_th=0.7, ocr_th=0.5,
+                       pad_top=90, pad_left=40, pad_bottom=90, pad_right=40):
+        # Convert the uploaded PIL Image to a temp file path
+        img.save("/tmp/input.png")
+        # Call your async pipeline and grab the DataFrame or JSON
+        result = asyncio.run(
+            te.start_process(
+                "/tmp/input.png",
+                TD_THRESHOLD=td_th,
+                TSR_THRESHOLD=tsr_th,
+                OCR_THRESHOLD=ocr_th,
+                padd_top=pad_top,
+                padd_left=pad_left,
+                padd_bottom=pad_bottom,
+                padd_right=pad_right,
+                delta_xmin=10,
+                delta_ymin=3,
+                delta_xmax=10,
+                delta_ymax=3,
+                expand_rowcol_bbox_top=0,
+                expand_rowcol_bbox_bottom=0
+            )
+        )
+        return result  # make sure your start_process returns JSON/dict
 
     async def start_process(self, image_path: str, TD_THRESHOLD, TSR_THRESHOLD,
                             OCR_THRESHOLD, padd_top, padd_left, padd_bottom,
@@ -575,42 +600,17 @@ class TableExtractionPipeline():
         #     st.write('Either incorrectly identified table or no table, to debug remove try/except')
         # break
         # break
-
-
 if __name__ == "__main__":
+            iface = gr.Interface(
+                fn=extract_table_json,
+                inputs=[
+                  gr.Image(type="pil", label="Page Image"),
+                  gr.Slider(0,1,0.8, label="Table-Detection Threshold"),
+                  gr.Slider(0,1,0.7, label="Structure Threshold"),
+                  gr.Slider(0,1,0.5, label="OCR Threshold"),
+                ],
+                outputs=gr.JSON(label="Table JSON"),
+                title="Table→CSV JSON API"
+            )
+            iface.launch(server_name="0.0.0.0", server_port=7860)
 
-    st_up, st_lang = st.columns((1, 1))
-    img_name = st_up.file_uploader("Upload an image with table(s)")
-    lang = st_lang.selectbox('Language', ('en', 'japan'))
-    reload_ocr(lang)
-
-    st1, st2, st3 = st.columns((1, 1, 1))
-    TD_th = st1.slider('Table detection threshold', 0.0, 1.0, 0.8)
-    TSR_th = st2.slider('Table structure recognition threshold', 0.0, 1.0, 0.7)
-    OCR_th = st3.slider("Text Probs Threshold", 0.0, 1.0, 0.5)
-
-    st1, st2, st3, st4 = st.columns((1, 1, 1, 1))
-
-    padd_top = st1.slider('Padding top', 0, 200, 90)
-    padd_left = st2.slider('Padding left', 0, 200, 40)
-    padd_right = st3.slider('Padding right', 0, 200, 40)
-    padd_bottom = st4.slider('Padding bottom', 0, 200, 90)
-
-    te = TableExtractionPipeline()
-    # for img in image_list:
-    if img_name is not None:
-        asyncio.run(
-            te.start_process(img_name,
-                             TD_THRESHOLD=TD_th,
-                             TSR_THRESHOLD=TSR_th,
-                             OCR_THRESHOLD=OCR_th,
-                             padd_top=padd_top,
-                             padd_left=padd_left,
-                             padd_bottom=padd_bottom,
-                             padd_right=padd_right,
-                             delta_xmin=10, # add offset to the left of the table
-                             delta_ymin=3, # add offset to the bottom of the table
-                             delta_xmax=10, # add offset to the right of the table
-                             delta_ymax=3, # add offset to the top of the table
-                             expand_rowcol_bbox_top=0,
-                             expand_rowcol_bbox_bottom=0))