Add better examples and confidence threshold

app.py

@@ -73,19 +73,21 @@ def normalize_bbox(box, width, height, padding=0.005):
 EXAMPLES = [
     [
         "invoice.png",
-        "Invoice 1",
+        "East Repair Invoice",
     ],
     [
-        "contract.jpeg",
-        "What is the purchase amount?",
+        "acze.png",
+        "ACZE Invoice",
     ],
     [
-        "statement.png",
-        "What are net sales for 2020?",
+        "north_sea.png",
+        "North Sea Invoice",
     ],
 ]
 
-QUESTION_FILES = {}
+QUESTION_FILES = {
+    "North Sea Invoice": "north_sea.pdf",
+}
 
 FIELDS = {
     "Vendor Name": ["Vendor Name - Logo?", "Vendor Name - Address?"],
@@ -98,6 +100,8 @@ FIELDS = {
     "Invoice Total": ["Invoice Total?"],
     "Amount Due": ["Amount Due?"],
     "Payment Terms": ["Payment Terms?"],
+    "Remit To Name": ["Remit To Name?"],
+    "Remit To Address": ["Remit To Address?"],
 }
 
 
@@ -150,7 +154,7 @@ colors = ["#64A087", "green", "black"]
 
 
 def annotate_page(prediction, pages, document):
-    if "word_ids" in prediction:
+    if prediction is not None and "word_ids" in prediction:
         image = pages[prediction["page"]]
         draw = ImageDraw.Draw(image, "RGBA")
         word_boxes = lift_word_boxes(document, prediction["page"])
@@ -192,9 +196,14 @@ def process_fields(document, fields, model=list(CHECKPOINTS.keys())[0]):
     table = []
 
     for (field_name, questions) in fields.items():
-        answers = [run_pipeline(model, q, document, top_k=1) for q in questions]
+        answers = [
+            a
+            for q in questions
+            for a in ensure_list(run_pipeline(model, q, document, top_k=1))
+            if a.get("score", 1) > 0.5
+        ]
         answers.sort(key=lambda x: -x.get("score", 0) if x else 0)
-        top = answers[0]
+        top = answers[0] if len(answers) > 0 else None
         annotate_page(top, pages, document)
         ret[field_name] = top
         table.append([field_name, top.get("answer") if top is not None else None])
@@ -400,8 +409,8 @@ with gr.Blocks(css=CSS) as demo:
                 None,  # document
                 # {**FIELDS},  # fields
                 gr.update(visible=False, value=None),  # output
-                # gr.update(**empty_table(FIELDS)),  # output_table
-                # gr.update(visible=False),
+                gr.update(**empty_table(fields.value)),  # output_table
+                gr.update(visible=False),
                 None,
                 None,
                 None,
@@ -414,8 +423,8 @@ with gr.Blocks(css=CSS) as demo:
                 document,
                 # fields,
                 output,
-                # output_table,
-                # img_clear_button,
+                output_table,
+                img_clear_button,
                 example_image,
                 upload,
                 url,