Update app.py

app.py

@@ -13,11 +13,23 @@ from transformers import LayoutLMTokenizerFast, LayoutLMForTokenClassification
 # =====================================================
 RESEND_API_KEY = os.getenv("RESEND_API_KEY")
 
-label2id = {"O": 0, "COMPANY": 1, "DATE": 2, "TOTAL": 3}
-id2label = {v: k for k, v in label2id.items()}
+# 🔥 IMPORTANT:
+# Use YOUR VERIFIED DOMAIN EMAIL
+# Example:
+# claims@send.yudham.com
+FROM_EMAIL = "AI Claims <claims@send.yudham.com>"
 
 MODEL_NAME = "ngupta2026/sroie-layoutlm"
 
+label2id = {
+    "O": 0,
+    "COMPANY": 1,
+    "DATE": 2,
+    "TOTAL": 3
+}
+
+id2label = {v: k for k, v in label2id.items()}
+
 # =====================================================
 # LOAD MODEL
 # =====================================================
@@ -30,7 +42,7 @@ model.to(device)
 model.eval()
 
 # =====================================================
-# NORMALIZE BOXES
+# NORMALIZE BOX
 # =====================================================
 def normalize(box, width, height):
     return [
@@ -41,7 +53,7 @@ def normalize(box, width, height):
     ]
 
 # =====================================================
-# CONFIDENCE HELPER
+# CONFIDENCE AVG
 # =====================================================
 def avg_conf(lst):
     if len(lst) == 0:
@@ -49,9 +61,10 @@ def avg_conf(lst):
     return sum(lst) / len(lst)
 
 # =====================================================
-# EXTRACT DATA + CONFIDENCE
+# OCR + EXTRACTION
 # =====================================================
 def extract_receipt(image):
+
     try:
         image = image.convert("RGB")
         image.thumbnail((1200, 1200))
@@ -61,12 +74,15 @@ def extract_receipt(image):
             output_type=pytesseract.Output.DICT
         )
 
-        words, boxes = [], []
+        words = []
+        boxes = []
 
         for i in range(len(data["text"])):
+
             text = data["text"][i].strip()
 
-            if text != "" and len(text) > 2:
+            if text != "" and len(text) > 1:
+
                 x = data["left"][i]
                 y = data["top"][i]
                 w = data["width"][i]
@@ -79,7 +95,7 @@ def extract_receipt(image):
             return {"error": "No text detected"}
 
         width, height = image.size
-        boxes = [normalize(box, width, height) for box in boxes]
+        boxes = [normalize(b, width, height) for b in boxes]
 
         encoding = tokenizer(
             words,
@@ -96,8 +112,8 @@ def extract_receipt(image):
         with torch.no_grad():
             outputs = model(**encoding)
 
-        # 🔥 CONFIDENCE LOGIC
         probs = torch.softmax(outputs.logits, dim=2)
+
         preds = torch.argmax(probs, dim=2)[0][:len(words)]
         confs = torch.max(probs, dim=2)[0][0][:len(words)]
 
@@ -114,39 +130,50 @@ def extract_receipt(image):
         }
 
         for word, pred, conf in zip(words, preds, confs):
+
             label = id2label[pred.item()]
-            confidence = conf.item()
+            c = conf.item()
 
             if label == "COMPANY":
                 result["company"].append(word)
-                conf_store["company"].append(confidence)
+                conf_store["company"].append(c)
 
             if re.search(r"\d{2}[/-]\d{2}[/-]\d{2,4}", word):
                 result["date"].append(word)
-                conf_store["date"].append(confidence)
+                conf_store["date"].append(c)
 
             if re.search(r"\d+(\.\d{2})?", word):
                 try:
                     value = float(word.replace(",", ""))
                     if value > 50:
                         result["total"].append(word)
-                        conf_store["total"].append(confidence)
+                        conf_store["total"].append(c)
                 except:
                     pass
 
-        # FINAL VALUES
-        result["company"] = " ".join(result["company"]) if result["company"] else "Not Found"
-        result["date"] = result["date"][0] if result["date"] else "Not Found"
-        result["total"] = result["total"][-1] if result["total"] else "Not Found"
+        # FINAL CLEAN
+        result["company"] = (
+            " ".join(result["company"])
+            if result["company"] else "Not Found"
+        )
+
+        result["date"] = (
+            result["date"][0]
+            if result["date"] else "Not Found"
+        )
+
+        result["total"] = (
+            result["total"][-1]
+            if result["total"] else "Not Found"
+        )
 
-        # 🔥 FINAL CONFIDENCE
         company_conf = avg_conf(conf_store["company"])
         date_conf = avg_conf(conf_store["date"])
         total_conf = avg_conf(conf_store["total"])
 
-        overall_conf = (company_conf + date_conf + total_conf) / 3
+        overall = (company_conf + date_conf + total_conf) / 3
 
-        result["confidence"] = round(overall_conf, 3)
+        result["confidence"] = round(overall, 3)
 
         return result
 
@@ -156,32 +183,39 @@ def extract_receipt(image):
 # =====================================================
 # DECISION ENGINE
 # =====================================================
-def decision_layer(confidence):
+def decision_layer(conf):
 
-    if confidence >= 0.80:
+    if conf >= 0.80:
         return "AUTO_SEND"
 
-    elif confidence >= 0.60:
+    elif conf >= 0.60:
         return "REVIEW"
 
     else:
         return "REJECT"
 
 # =====================================================
-# SEND EMAIL (RESEND)
+# EMAIL SEND
 # =====================================================
 def send_claim_email(to_email, extracted):
 
     if not RESEND_API_KEY:
-        return "❌ RESEND_API_KEY missing"
+        return "❌ Missing RESEND_API_KEY secret"
 
     subject = "Insurance Claim Request"
 
     html_body = f"""
     <h2>Insurance Claim Request</h2>
-    <p><b>Provider:</b> {extracted['company']}</p>
-    <p><b>Date:</b> {extracted['date']}</p>
-    <p><b>Amount:</b> ₹{extracted['total']}</p>
+
+    <p>Dear Claims Team,</p>
+
+    <p>Please process reimbursement request.</p>
+
+    <p><b>Provider Name:</b> {extracted['company']}</p>
+    <p><b>Bill Date:</b> {extracted['date']}</p>
+    <p><b>Claim Amount:</b> ₹{extracted['total']}</p>
+
+    <p>Regards,<br>AI Claims System</p>
     """
 
     try:
@@ -189,19 +223,20 @@ def send_claim_email(to_email, extracted):
             "https://api.resend.com/emails",
             headers={
                 "Authorization": f"Bearer {RESEND_API_KEY}",
-                "Content-Type": "application/json",
+                "Content-Type": "application/json"
             },
             json={
-                "from": "noreply@yourdomain.com",  # change after domain verify
+                "from": FROM_EMAIL,
                 "to": [to_email],
                 "subject": subject,
-                "html": html_body,
+                "html": html_body
             },
-            timeout=10
+            timeout=20
         )
 
         if response.status_code in [200, 201]:
-            return f"✅ Email sent to {to_email}"
+            return f"✅ Email sent successfully to {to_email}"
+
         else:
             return f"❌ Email failed: {response.text}"
 
@@ -218,20 +253,19 @@ def process_and_send(image, email_id):
     if "error" in extracted:
         return extracted, extracted["error"]
 
-    confidence = extracted.get("confidence", 0)
-    decision = decision_layer(confidence)
+    conf = extracted["confidence"]
+    decision = decision_layer(conf)
+
+    extracted["decision"] = decision
 
-    # 🔥 DECISION CONTROL
     if decision == "AUTO_SEND":
         email_status = send_claim_email(email_id, extracted)
 
     elif decision == "REVIEW":
-        email_status = f"⚠️ Needs human review (confidence={confidence})"
+        email_status = f"⚠️ Human review required (confidence={conf})"
 
     else:
-        email_status = f"❌ Rejected due to low confidence ({confidence})"
-
-    extracted["decision"] = decision
+        email_status = f"❌ Rejected (low confidence={conf})"
 
     return extracted, email_status
 
@@ -240,16 +274,19 @@ def process_and_send(image, email_id):
 # =====================================================
 demo = gr.Interface(
     fn=process_and_send,
+
     inputs=[
         gr.Image(type="pil", label="Upload Receipt"),
-        gr.Textbox(label="Enter Email ID")
+        gr.Textbox(label="Enter Destination Email")
     ],
+
     outputs=[
-        gr.JSON(label="Extracted Data + Confidence + Decision"),
+        gr.JSON(label="AI Extraction"),
         gr.Textbox(label="Email Status")
     ],
-    title="📄 AI Insurance Claim Generator (Enterprise Version)",
-    description="Upload receipt → AI extracts → Confidence check → Safe decision → Email"
+
+    title="📄 AI Insurance Claim Generator",
+    description="Upload receipt → Extract fields → Confidence Check → Auto Email"
 )
 
 demo.launch()