data-science-document-ai 1.60.1__py3-none-any.whl → 1.60.2__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (4) hide show
  1. {data_science_document_ai-1.60.1.dist-info → data_science_document_ai-1.60.2.dist-info}/METADATA +1 -1
  2. {data_science_document_ai-1.60.1.dist-info → data_science_document_ai-1.60.2.dist-info}/RECORD +4 -4
  3. src/constants.py +16 -0
  4. {data_science_document_ai-1.60.1.dist-info → data_science_document_ai-1.60.2.dist-info}/WHEEL +0 -0
{data_science_document_ai-1.60.1.dist-info → data_science_document_ai-1.60.2.dist-info}/METADATA RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: data-science-document-ai
3
- Version: 1.60.1
3
+ Version: 1.60.2
4
4
  Summary: "Document AI repo for data science"
5
5
  Author: Naomi Nguyen
6
6
  Author-email: naomi.nguyen@forto.com
{data_science_document_ai-1.60.1.dist-info → data_science_document_ai-1.60.2.dist-info}/RECORD RENAMED
@@ -1,4 +1,4 @@
1
- src/constants.py,sha256=Mr5HcoMSFQlXY24shJ1EpNKtCBJP_Ai0LZj-0i3BVGk,2845
1
+ src/constants.py,sha256=E1P6CvQ8eoxzI5pDD_4HaNKLnx3SL94Uo4l6Xy5juxM,3545
2
2
  src/constants_sandbox.py,sha256=Iu6HdjCoNSmOX0AwoL9qUQkhq_ZnIN5U9e-Q2UfNuGc,547
3
3
  src/docai.py,sha256=dHuR0ehVjUi1CnoNvdp_yxJtpU_HFXqAZ61ywdz7BEo,5655
4
4
  src/docai_processor_config.yaml,sha256=ZewXqbyiftzmVAaS08BoNp5trY6WXx3HMWDfPwmKfaI,256
@@ -55,6 +55,6 @@ src/prompts/prompt_library.py,sha256=VJWHeXN-s501C2GiidIIvQQuZdU6T1R27hE2dKBiI40
55
55
  src/setup.py,sha256=8-vZWjC8Iwa3xxdk3iR4412VCjtNtgzVqkXcFon7UBE,7309
56
56
  src/tms.py,sha256=UXbIo1QE--hIX6NZi5Qyp2R_CP338syrY9pCTPrfgnE,1741
57
57
  src/utils.py,sha256=Ow5_Jals88o8mbZ1BoHfZpHZoCfig_UQb5aalH-mpWE,17278
58
- data_science_document_ai-1.60.1.dist-info/METADATA,sha256=1FvSj7W8CoSJ--J4VNTM_HFUF1UeF3XXVyYs0NmHTD0,2152
59
- data_science_document_ai-1.60.1.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
60
- data_science_document_ai-1.60.1.dist-info/RECORD,,
58
+ data_science_document_ai-1.60.2.dist-info/METADATA,sha256=OchQC8yfBcQyJrk-zv2uTqC3DWlXYnQMQ7SansDP5eU,2152
59
+ data_science_document_ai-1.60.2.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
60
+ data_science_document_ai-1.60.2.dist-info/RECORD,,
src/constants.py CHANGED
@@ -70,6 +70,22 @@ project_parameters = {
70
70
  "seed": 42,
71
71
  "model_id": "gemini-2.5-flash",
72
72
  },
73
+ # Key to combine the LLM results with the Doc Ai results
74
+ # TODO: remove. No longer using doc ai results. Just to keep track which fields are line item fields
75
+ "key_to_combine": {
76
+ "bookingConfirmation": ["transportLegs"],
77
+ "arrivalNotice": ["containers"],
78
+ "finalMbL": ["containers"],
79
+ "draftMbl": ["containers"],
80
+ "deliveryOrder": ["Equipment", "TransportLeg"],
81
+ "customsAssessment": ["containers"],
82
+ "packingList": ["skuData"],
83
+ "commercialInvoice": ["skus"],
84
+ "shippingInstruction": ["containers"],
85
+ "partnerInvoice": ["lineItem"],
86
+ "customsInvoice": ["lineItem"],
87
+ "bundeskasse": ["lineItem"],
88
+ },
73
89
  }
74
90
 
75
91
  # Hardcoded rules for data points formatting that can't be based on label name alone