data-science-document-ai 1.45.0__tar.gz → 1.45.1__tar.gz

{data_science_document_ai-1.45.0 → data_science_document_ai-1.45.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: data-science-document-ai
-Version: 1.45.0
+Version: 1.45.1
 Summary: "Document AI repo for data science"
 Author: Naomi Nguyen
 Author-email: naomi.nguyen@forto.com

{data_science_document_ai-1.45.0 → data_science_document_ai-1.45.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "data-science-document-ai"
-version = "1.45.0"
+version = "1.45.1"
 description = "\"Document AI repo for data science\""
 authors = ["Naomi Nguyen <naomi.nguyen@forto.com>", "Kumar Rajendrababu <kumar.rajendrababu@forto.com>", "Igor Tonko <igor.tonko@forto.com>", "Osman Demirel <osman.demirel@forto.com>"]
 packages = [

{data_science_document_ai-1.45.0 → data_science_document_ai-1.45.1}/src/postprocessing/common.py

@@ -134,8 +134,11 @@ def extract_number(data_field_value):
         formatted_value: string
 
     """
+    # Remove container size pattern like 20FT, 40HC, etc from 1 x 40HC
+    value = remove_unwanted_patterns(data_field_value)
+
     formatted_value = ""
-    for c in data_field_value:
+    for c in value:
         if c.isnumeric() or c in [",", ".", "-"]:
             formatted_value += c
 
@@ -320,9 +323,12 @@ def remove_unwanted_patterns(lineitem: str):
     lineitem = lineitem.replace("HIGH CUBE", "")
 
     # Remove container size e.g., 20FT, 40HC, etc.
-    lineitem = re.sub(
-        r"\b(20|22|40|45)(FT|HC|DC|HD|GP|OT|RF|FR|TK|DV)?\b", "", lineitem
-    ).strip()
+    pattern = [
+        f"{s}{t}"
+        for s in ("20|22|40|45".split("|"))
+        for t in ("FT|HC|DC|HD|GP|OT|RF|FR|TK|DV".split("|"))
+    ]
+    lineitem = re.sub(r"|".join(pattern), "", lineitem, flags=re.IGNORECASE).strip()
 
     return lineitem
 

{data_science_document_ai-1.45.0 → data_science_document_ai-1.45.1}/src/prompts/library/customsInvoice/other/prompt.txt

@@ -54,7 +54,7 @@ Your role is to accurately extract specific entities from these invoices to supp
     - unitPrice: Even if the quantity is not mentioned, you can still extract the unit price. Check the naming of the columns in a different languages, it can be "Unit Price", "Prezzo unitario", "Prix Unitaire", "Unitario", etc. Refer to "Prezzo unitario" field in the italian invoice example.
     - totalAmount: The total amount for the item. It can be in different currencies, so ensure to capture the currency as well for the totalAmountCurrency.
     - totalAmountEuro: Few line items contains a total amount in Euro. You can find it by looking for the term "Total EUR" or "Amount in Euro" in the line item but it's always in the EURO / € currency. Sometimes, it can be same as totalAmount if the line item is already in Euro.
-    - quantity: The quantity of the item or service provided in the line item.
+    - quantity: The quantity of the item or service provided in the line item. Pay attention to 2 x 40HC or 2x40HC. It means, quantity is 2 and 40HC is containerSize but not 240.
     - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567).
 
 - hblNumber and mblNumber:

{data_science_document_ai-1.45.0 → data_science_document_ai-1.45.1}/src/prompts/library/partnerInvoice/other/prompt.txt

@@ -52,7 +52,7 @@ Your role is to accurately extract specific entities from these invoices to supp
     - unitPrice: Even if the quantity is not mentioned, you can still extract the unit price. Check the naming of the columns in a different languages, it can be "Unit Price", "Prezzo unitario", "Prix Unitaire", "Unitario", etc. Refer to "Prezzo unitario" field in the italian invoice example.
     - totalAmount: The total amount for the item. It can be in different currencies, so ensure to capture the currency as well for the totalAmountCurrency.
     - totalAmountEuro: Few line items contains a total amount in Euro. You can find it by looking for the term "Total EUR" or "Amount in Euro" in the line item but it's always in the EURO / € currency. Sometimes, it can be same as totalAmount if the line item is already in Euro.
-    - quantity: The quantity of the item or service provided in the line item. Pay attention to 2x40HC. It means, quantity is 2 and containerSize is 40HC but not 240.
+    - quantity: The quantity of the item or service provided in the line item. Pay attention to 2 x 40HC or 2x40HC. It means, quantity is 2 and 40HC is containerSize but not 240.
     - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567, XALU 8593678).
 
 - hblNumber and mblNumber:
@@ -68,6 +68,7 @@ Your role is to accurately extract specific entities from these invoices to supp
     - Example:
         - "COSCO SHIPPING Lines Italy, Poland, or France S.R.L. – Genova Office – As Agent For COSCO SHIPPING Lines Co.,Ltd."
         - vendorName: COSCO SHIPPING Lines Co.,Ltd.
+    - From Hapag-Lloyd invoices, look for "Ballindamm 25" address to extract the vendorAddress.
 
 - agentName: Name of the agent. Agencies are offices authorized to act on behalf of a company. This details usually available including the branch name of the parent company name in the invoice.
 - agentKeyWord: