sdg-hub 0.3.1__py3-none-any.whl → 0.4.1__py3-none-any.whl

sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/generate_questions_responses_ja.yaml

@@ -0,0 +1,55 @@
+- role: system
+  content: You are a very knowledgeable AI Assistant that will faithfully assist the user with their task.
+
+- role: user
+  content: |
+    Develop a series of as many educational question and answer pairs as possible from a chapter in a {{domain}} textbook.
+
+    The questions should:
+    * Be self-contained, not requiring references to tables, figures, or specific sections in the text for understanding.
+    * Focus on teaching and reinforcing the key knowledge and concepts presented in the chapter.
+    * Avoid sections with minimal educational content like index pages or prefaces. In such cases, respond with [UNANSWERABLE].
+    * Be directly relevant to the textbook's domain. For instance, in a science textbook, questions should revolve around scientific terms, definitions, and practical applications, while in a legal textbook, they should cover legal principles, case law, and precedents.
+    * Be formulated to allow for independent answers, avoiding direct references to specific theorems or text sections. For example, rather than asking 'Under what conditions is the fixed point of a function unique according to Theorem 3.1.5?', ask 'How does the Fixed Point Iteration method contribute to understanding function uniqueness?'
+    * Span a range of difficulty levels to accommodate a diverse student audience, from basic understanding to advanced comprehension.
+    * Include a variety of question types such as multiple-choice for basic recall, short answer for deeper understanding, and essay or problem-solving questions to test application and analysis skills.
+    * Align closely with the learning objectives of the textbook or the specific chapter, ensuring that the questions test the fundamental concepts and skills that the chapter aims to impart.
+    * Be in Japanese.
+
+    Strictly follow this format for each question answer pair your generate while responding:
+
+    [QUESTION]
+    <Insert question here>
+    [ANSWER]
+    <Insert answer here>
+    [END]
+
+    Each question and answer pair should stand alone as a mini-lesson, encapsulating a key concept or idea from the chapter in a way that is accessible and informative without requiring the reader to refer back to the textbook.
+
+    Here are some examples of questions:
+
+    [Document]
+    {{icl_document}}
+
+    [QUESTION]
+    {{icl_query_1}}
+    [ANSWER]
+    {{icl_response_1}}
+    [END]
+
+    [QUESTION]
+    {{icl_query_2}}
+    [ANSWER]
+    {{icl_response_2}}
+    [END]
+
+    [QUESTION]
+    {{icl_query_3}}
+    [ANSWER]
+    {{icl_response_3}}
+    [END]
+
+    Now, here is the document:
+    [DOCUMENT]
+    {{document_outline}}
+    {{document}}

sdg_hub/flows/text_analysis/structured_insights/flow.yaml

@@ -24,7 +24,6 @@ metadata:
     - "entity-extraction"
     - "keyword-extraction"
   license: "Apache-2.0"
-  min_sdg_hub_version: "0.2.0"
   dataset_requirements:
     required_columns:
       - "text"
@@ -50,10 +49,16 @@ blocks:
       max_tokens: 1024
       temperature: 0.3
       async_mode: true
+  - block_type: "LLMParserBlock"
+    block_config:
+      block_name: "extract_summary"
+      input_cols: "raw_summary"
+      extract_content: true
+      expand_lists: true
   - block_type: "TextParserBlock"
     block_config:
       block_name: "parse_summary"
-      input_cols: "raw_summary"
+      input_cols: "extract_summary_content"
       output_cols: "summary"
       start_tags:
         - "[SUMMARY]"
@@ -76,10 +81,16 @@ blocks:
       max_tokens: 512
       temperature: 0.3
       async_mode: true
+  - block_type: "LLMParserBlock"
+    block_config:
+      block_name: "extract_keywords"
+      input_cols: "raw_keywords"
+      extract_content: true
+      expand_lists: true
   - block_type: "TextParserBlock"
     block_config:
       block_name: "parse_keywords"
-      input_cols: "raw_keywords"
+      input_cols: "extract_keywords_content"
       output_cols: "keywords"
       start_tags:
         - "[KEYWORDS]"
@@ -102,10 +113,16 @@ blocks:
       max_tokens: 1024
       temperature: 0.3
       async_mode: true
+  - block_type: "LLMParserBlock"
+    block_config:
+      block_name: "extract_entities"
+      input_cols: "raw_entities"
+      extract_content: true
+      expand_lists: true
   - block_type: "TextParserBlock"
     block_config:
       block_name: "parse_entities"
-      input_cols: "raw_entities"
+      input_cols: "extract_entities_content"
       output_cols: "entities"
       start_tags:
         - "[ENTITIES]"
@@ -128,10 +145,16 @@ blocks:
       max_tokens: 256
       temperature: 0.1
       async_mode: true
+  - block_type: "LLMParserBlock"
+    block_config:
+      block_name: "extract_sentiment"
+      input_cols: "raw_sentiment"
+      extract_content: true
+      expand_lists: true
   - block_type: "TextParserBlock"
     block_config:
       block_name: "parse_sentiment"
-      input_cols: "raw_sentiment"
+      input_cols: "extract_sentiment_content"
       output_cols: "sentiment"
       start_tags:
         - "[SENTIMENT]"

{sdg_hub-0.3.1.dist-info → sdg_hub-0.4.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sdg_hub
-Version: 0.3.1
+Version: 0.4.1
 Summary: Synthetic Data Generation
 Author-email: Red Hat AI Innovation <abhandwa@redhat.com>
 License: Apache-2.0
@@ -65,6 +65,7 @@ Requires-Dist: pytest-html; extra == "dev"
 Requires-Dist: tox<5,>=4.4.2; extra == "dev"
 Requires-Dist: ruff; extra == "dev"
 Requires-Dist: pytest-env; extra == "dev"
+Requires-Dist: nbconvert>=7.0.0; extra == "dev"
 Dynamic: license-file
 
 # `sdg_hub`: Synthetic Data Generation Toolkit

{sdg_hub-0.3.1.dist-info → sdg_hub-0.4.1.dist-info}/RECORD

@@ -1,9 +1,9 @@
-sdg_hub/__init__.py,sha256=Tw-6R5a8_W1kJcTAsW3R9ltBDP1dy5-fe7Tvt3cSyCQ,550
-sdg_hub/_version.py,sha256=gGLpQUQx-ty9SEy9PYw9OgJWWzJLBnCpfJOfzL7SjlI,704
+sdg_hub/__init__.py,sha256=TlkZT40-70urdcWLqv3kupaJj8s-SVgd2QyvlSFwb4A,510
+sdg_hub/_version.py,sha256=k7cu0JKra64gmMNU_UfA5sw2eNc_GRvf3QmesiYAy8g,704
 sdg_hub/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sdg_hub/core/__init__.py,sha256=NwqB4fwhC29W50VW7QXZssLxx122YvgO9LHDLdgAnrI,496
-sdg_hub/core/blocks/__init__.py,sha256=9sCkCvDQzJGSedaePVlEIpbNwrkBz_K500VW_6FLhuE,1601
-sdg_hub/core/blocks/base.py,sha256=TrzUAkG7Tiquk0Z3SOFsb5mRnHd1IbHH6gFPVH1P7T8,10424
+sdg_hub/core/__init__.py,sha256=e3BoejbqjYhasf9t__L4qE52lkD9EBjx4o--2kqKdro,460
+sdg_hub/core/blocks/__init__.py,sha256=5FsbkcO-dmBv6MqO96TPn9FKKPTQZQCv20j4wR7UvQw,1502
+sdg_hub/core/blocks/base.py,sha256=-SOdBpJwtRTMsrmCEuLjUBQMRCo_PLYlHEBRrz8sF9g,13031
 sdg_hub/core/blocks/registry.py,sha256=FuEN_pnq-nSH1LguY3_oCubT6Kz3SuJjk3TcUpLT-lw,10695
 sdg_hub/core/blocks/deprecated_blocks/__init__.py,sha256=RDu3MWFStDQko-TKkx8tGoB1UTatP_RSldZK43zHDvY,889
 sdg_hub/core/blocks/deprecated_blocks/combine_columns.py,sha256=HCvpaYsAwgx1Dm0vIshcWsKoVsRT0KrmKp9j4oqtByc,2757
@@ -15,20 +15,15 @@ sdg_hub/core/blocks/deprecated_blocks/rename_columns.py,sha256=thp-mHtkRmUw_nYKp
 sdg_hub/core/blocks/deprecated_blocks/sample_populator.py,sha256=UdueMApxOmPWaxxMrw7b1v74fKJBfqqRATEBqgmVtNw,1737
 sdg_hub/core/blocks/deprecated_blocks/selector.py,sha256=nWecsVsW8DvBcqAF_LOqXmW-5MQ28uN3d1y6wkSy38c,2960
 sdg_hub/core/blocks/deprecated_blocks/set_to_majority_value.py,sha256=44TQu-rK5isia-otMVB1zHd8D-wWmu3C8CI1NLtfY5s,2729
-sdg_hub/core/blocks/evaluation/__init__.py,sha256=kFXee-vsVVdU2XtLio9qHgPx_a0zoB_rQr509EKBGJc,357
-sdg_hub/core/blocks/evaluation/evaluate_faithfulness_block.py,sha256=vFi3YIxVPNnzgdenIeAl7yUb4OOUY_uUOXS-pWLsDmw,12223
-sdg_hub/core/blocks/evaluation/evaluate_relevancy_block.py,sha256=NXT1lixR-JnOXNlBCbMjULcpu4kh2SthhwCWEobiBt0,12115
-sdg_hub/core/blocks/evaluation/verify_question_block.py,sha256=LKoIHdxUuTVO24n_M9cAliEj56uEe2kQAecKTRz65zI,12465
 sdg_hub/core/blocks/filtering/__init__.py,sha256=isxSVSvDqkMjG8dQSl3Q2M4g5c1t9fTjBSA21icf-yA,275
-sdg_hub/core/blocks/filtering/column_value_filter.py,sha256=H8Gif0q9Wc_d1TnVow8Zpsg7blJOFGN1EZmV6OPpkcg,5971
-sdg_hub/core/blocks/llm/__init__.py,sha256=N6-Prgd4X85oWbMQzhYMrq7OX-NTJm57cghowK-val0,844
-sdg_hub/core/blocks/llm/client_manager.py,sha256=6RNqYvFIh4SF6jopI6tTY5MA01y8Qo-tAhsE0GeHZZ0,16109
-sdg_hub/core/blocks/llm/config.py,sha256=gc4xp5D20MSlKMFEos0QAaKUwgbZpBtMGXmn6LsIk78,11289
+sdg_hub/core/blocks/filtering/column_value_filter.py,sha256=2Z9j_CiiTn5mHZ9gfXU-itLXDmeXSh0UI0x1x7j-LQ0,6001
+sdg_hub/core/blocks/llm/__init__.py,sha256=AyS0dd3pkPPXH5a9aj4mT5HsKjX2vjXfkmQc6rkFV4A,795
 sdg_hub/core/blocks/llm/error_handler.py,sha256=7T-019ZFB9qgZoX1ybIiXyaLjPzrF96qcKmUu6vmO6g,12178
-sdg_hub/core/blocks/llm/llm_chat_block.py,sha256=9ytjxjADM0FydkLapZPSQPfzjjrFIdFONs3EJEoKnaw,23007
-sdg_hub/core/blocks/llm/llm_chat_with_parsing_retry_block.py,sha256=H7LqYpEiFO1g2cxncAl4vkLhQxAYgGpV6cUyQTSG03k,27628
+sdg_hub/core/blocks/llm/llm_chat_block.py,sha256=MHhI2x9i6LrfDXgvAy2_6YxgyoD7j6BpCgNGsM69xDg,22194
+sdg_hub/core/blocks/llm/llm_chat_with_parsing_retry_block.py,sha256=DW4b09IqXmcshvXawFheDyaLp3rz7vpO5VBrKdUQYW8,31703
+sdg_hub/core/blocks/llm/llm_parser_block.py,sha256=aoHqsDDhaIgCDfPpv7acc0DVN-zUgzFflRVB4win0aM,12012
 sdg_hub/core/blocks/llm/prompt_builder_block.py,sha256=fkJd718X1oYlMY1cjo_8WCO16Gl8Tm0bUPWR78E_uws,13935
-sdg_hub/core/blocks/llm/text_parser_block.py,sha256=8oRlXEkw8ULA8XVa7WtQZUojodl_ihs1omZpvbwoJQE,17165
+sdg_hub/core/blocks/llm/text_parser_block.py,sha256=975HK6NfXiU9Any4UDMpBNidRpyhHmc76BXUN69SVyc,12566
 sdg_hub/core/blocks/transform/__init__.py,sha256=lF9InjOzA6p_mjiwV-a2Kwstq9kqRiQ-dEwbsmR9yQs,825
 sdg_hub/core/blocks/transform/duplicate_columns.py,sha256=SaP7rIF4ZFEFFa50aU2xGNIuddXaEZrKxdWfHjzFpVI,2833
 sdg_hub/core/blocks/transform/index_based_mapper.py,sha256=XC_a7Skbd3mu7f4ra8fGWPxMwqUMSjJkQ7Ag7vflwJA,8235
@@ -37,12 +32,12 @@ sdg_hub/core/blocks/transform/melt_columns.py,sha256=vaYa5Taq6GhNZYWFL4uPK3-SfN2
 sdg_hub/core/blocks/transform/rename_columns.py,sha256=qeB5L2utqDQnutUetH1VKZSqDiJSH_yUp5EFCV-XCVI,1998
 sdg_hub/core/blocks/transform/text_concat.py,sha256=_-B__Hob1WwgwkILPIZvTnsDzuwtoX1hKviyzHlnnes,3149
 sdg_hub/core/blocks/transform/uniform_col_val_setter.py,sha256=XnjiT29z3PzIPy8M-mmE2w-Miab6Ed5ahy32SaxTCTE,3263
-sdg_hub/core/flow/__init__.py,sha256=N2NZGngvd7qpT5FI_knKukUFM0IkD9K5jdTi-gDeUI4,475
-sdg_hub/core/flow/base.py,sha256=8Xacytg9M82Mbv8r2GLbQgNltH-hCtFS1Fa1WpfFlSw,56488
+sdg_hub/core/flow/__init__.py,sha256=0_m_htuZfPxk8xQ9IKfp0Pz-JRE4O7lYMUFrKyLNoLA,409
+sdg_hub/core/flow/base.py,sha256=IRnNEZ3laDmR4sW_MTseL4syhLuUylyHY_0tS5QaS-A,54084
 sdg_hub/core/flow/checkpointer.py,sha256=stm5ZtjjEiLk9ZkAAnoQQn5Y8Yl_d7qCsQLZTrCXR48,11867
-sdg_hub/core/flow/metadata.py,sha256=h9jpvAzWsF5n4ztZMzwa9ZNgnzKTHmFWdn7YbyJLHCw,12977
+sdg_hub/core/flow/metadata.py,sha256=cFrpJjWOaK87aCuRFyC3Pdf83oYU93mrmZEMdUnhsN8,10540
 sdg_hub/core/flow/migration.py,sha256=6and-RBqV0t2gRipr1GiOOVnyBJdtyyjw1kO08Z--d4,7558
-sdg_hub/core/flow/registry.py,sha256=DzCqEEgwhvwnCBAGLogoMVdwXh4pCHrxOWqoxam7O8I,12162
+sdg_hub/core/flow/registry.py,sha256=N6KfX-L7QRkooznIFxDuhRZYuDA5g3N5zC-KRm2jVhk,12109
 sdg_hub/core/flow/validation.py,sha256=pUJvgaUjLpKNwvW6djcqVOF-HShOjegEmGOnUnoX4BA,9722
 sdg_hub/core/utils/__init__.py,sha256=C2FzLn3dHprwGJDEgI4fyFS3aoCJR-9PhHsunxropJ8,351
 sdg_hub/core/utils/datautils.py,sha256=__HkUe1DxcJVHKrFX68z_hDXwxJygBlJDfjJLnj7rHc,4230
@@ -59,12 +54,14 @@ sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/gener
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/generate_question_list.yaml,sha256=qHOgUNrQz2vjUjJiEHNGWxDDXwjJlP1kofTxeGgLyPI,1461
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/detailed_summary/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/detailed_summary/detailed_summary.yaml,sha256=Ik6gAml0O-jPq8jpXBAkURzYkQuFOnDZb4LDwjmfAiE,381
-sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/detailed_summary/flow.yaml,sha256=va9ESTlEaZozy8pXTJ8OICjRg08KSP4l305YUKFuGAE,4417
+sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/detailed_summary/flow.yaml,sha256=fUdzY9dtU69o99Uq8FIPycgVWdLD-1kbY97Bh-Vo2A0,5538
+sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/doc_direct_qa/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/doc_direct_qa/flow.yaml,sha256=smPWVUZRCt58EagWDmJVmTBQj8qMcjpzh-Q3GSuFrz0,4413
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/extractive_summary/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/extractive_summary/extractive_summary.yaml,sha256=SeapWoOx3fhN5SvWYuHss_9prLE8xSkOic7JkbDHSR0,4081
-sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/extractive_summary/flow.yaml,sha256=Iv4AlbE9PFtTn6teekgiNtrTiYio_nYWS8gyD6eFLUA,4580
+sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/extractive_summary/flow.yaml,sha256=iNNIfofFE7awK7iivtIFWxjfjy8QviMugOPPnOTySKA,5706
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/key_facts/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/key_facts/flow.yaml,sha256=Rv0c4s5vim2I5jKzQgjcUfVMdla6czzmZUU67hlTAbg,2898
+sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/key_facts/flow.yaml,sha256=CIUZNYhvszT-jpz1Hvh6nS2y5W34P529ZOMp8thEQ9k,3219
 sdg_hub/flows/qa_generation/document_grounded_qa/enhanced_multi_summary_qa/key_facts/key_facts_summary.yaml,sha256=YKMX_CuvcThG_bdNCAIXdVBkMvB72I89RGq2ltSSgc8,3298
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/README.md,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -74,17 +71,24 @@ sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/ev
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/evaluate_question.yaml,sha256=zwzklXup6khRkR88avgrJTcjaMcV1wnbeYaML5oPuNs,1767
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/evaluate_relevancy.yaml,sha256=cA8igo7jMrRXaWW6k0of6KOp7YnxLtPj0fP4DbrmZNQ,3647
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/extractive_summary.yaml,sha256=fcMV7LaCFZo4D29nwhGJXqFFuZMYVLo9XYjv8zcU6zs,364
-sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/flow.yaml,sha256=oyrLRjEnmioMa_G_sd9yQK_nBt4arwWV5fvKgzYE2ds,6090
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/flow.yaml,sha256=HR8sf7RUZKr8UqKztBj-nlvyrve1UMUu8x8qgYM6O14,9055
 sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/instructlab/generate_questions_responses.yaml,sha256=yX8aLY8dJSDML9ZJhnj9RzPbN8tH2xfcM4Gc6xZuwqQ,2596
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/README.md,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/atomic_facts_ja.yaml,sha256=OjPZaSCOSLxEWgW3pmNwF7mmLhGhFGTmKL_3rKdqeW4,2488
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/detailed_summary_ja.yaml,sha256=nEy_RcotHGiiENrmUANpKkbIFsrARAeSwECrBeHi2so,391
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/extractive_summary_ja.yaml,sha256=V90W0IeJQZTFThA8v0UOs3DtZbtU3BI9jkpChw1BULo,402
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/flow.yaml,sha256=iY1N6CY97fEkqI5oqaamSfqmiXpHPhWH_aOppsMxVjY,9176
+sdg_hub/flows/qa_generation/document_grounded_qa/multi_summary_qa/multilingual/japanese/generate_questions_responses_ja.yaml,sha256=96SQqXG7fmb-50SdX85sgVtrFcQ-oNKe_0BoQdZmY5g,2638
 sdg_hub/flows/text_analysis/__init__.py,sha256=WStks4eM_KHNTVsHglcj8vFghmI0PH9P1hUrijBLbwc,125
 sdg_hub/flows/text_analysis/structured_insights/__init__.py,sha256=_DT4NR05JD9CZoSWROPr2lC6se0VjSqQPZJJlEV79mk,274
 sdg_hub/flows/text_analysis/structured_insights/analyze_sentiment.yaml,sha256=1YGPypFJYS8qfYFj2J6ERTgodKJvMF4YHNGt_vOF5qc,1000
 sdg_hub/flows/text_analysis/structured_insights/extract_entities.yaml,sha256=Q_SDy14Zu-qS2sbKfUBmGlYj3k7CUg6HzzXlFCXRKuU,1169
 sdg_hub/flows/text_analysis/structured_insights/extract_keywords.yaml,sha256=_nPPMdHnxag_lYbhYUjGJGo-CvRwWvwdGX7cQhdZ1S0,847
-sdg_hub/flows/text_analysis/structured_insights/flow.yaml,sha256=2HuGTyNwYe6a8Ev-QdKZXwe29NL4wOkq4ecEV9a7NDg,4221
+sdg_hub/flows/text_analysis/structured_insights/flow.yaml,sha256=BBV18SdvuVTAESjwkJ7V1jbb-cSTBvNl3SCycd0oEQ4,4934
 sdg_hub/flows/text_analysis/structured_insights/summarize.yaml,sha256=WXwQak1pF8e1OwnOoI1EHu8QB6iUNW89rfkTdi1Oq54,687
-sdg_hub-0.3.1.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-sdg_hub-0.3.1.dist-info/METADATA,sha256=-dPDzTaPfnMb_n6p7Jcvkqv3Y-Ihi76psItQL7DQBX8,9735
-sdg_hub-0.3.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-sdg_hub-0.3.1.dist-info/top_level.txt,sha256=TqI7d-HE1n6zkXFkU0nF3A1Ct0P0pBaqI675uFokhx4,8
-sdg_hub-0.3.1.dist-info/RECORD,,
+sdg_hub-0.4.1.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+sdg_hub-0.4.1.dist-info/METADATA,sha256=pLRs5oOsVI9515UEZxcUEZFZhCoZ0kli0KLpBPPPB7w,9783
+sdg_hub-0.4.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+sdg_hub-0.4.1.dist-info/top_level.txt,sha256=TqI7d-HE1n6zkXFkU0nF3A1Ct0P0pBaqI675uFokhx4,8
+sdg_hub-0.4.1.dist-info/RECORD,,

sdg_hub/core/blocks/evaluation/__init__.py

@@ -1,9 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""Evaluation blocks for SDG Hub."""
-
-# Local
-from .evaluate_faithfulness_block import EvaluateFaithfulnessBlock
-from .evaluate_relevancy_block import EvaluateRelevancyBlock
-from .verify_question_block import VerifyQuestionBlock
-
-__all__ = ["EvaluateFaithfulnessBlock", "EvaluateRelevancyBlock", "VerifyQuestionBlock"]

sdg_hub/core/blocks/evaluation/evaluate_faithfulness_block.py

@@ -1,323 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-"""Thin wrapper for faithfulness evaluation using 4 composed blocks.
-
-This module provides a simple, lightweight wrapper that composes:
-- PromptBuilderBlock: builds evaluation prompts
-- LLMChatBlock: generates LLM responses
-- TextParserBlock: parses structured output
-- ColumnValueFilterBlock: filters based on judgment
-
-The wrapper exposes minimal LLM interface for flow detection while
-delegating all functionality to the internal blocks.
-"""
-
-# Standard
-from typing import Any, Optional
-
-# Third Party
-from datasets import Dataset
-from pydantic import ConfigDict, Field, field_validator
-
-# Local
-from ...utils.error_handling import BlockValidationError
-from ...utils.logger_config import setup_logger
-from ..base import BaseBlock
-from ..filtering.column_value_filter import ColumnValueFilterBlock
-from ..llm.llm_chat_block import LLMChatBlock
-from ..llm.prompt_builder_block import PromptBuilderBlock
-from ..llm.text_parser_block import TextParserBlock
-from ..registry import BlockRegistry
-
-logger = setup_logger(__name__)
-
-
-@BlockRegistry.register(
-    "EvaluateFaithfulnessBlock",
-    "evaluation",
-    "Thin wrapper composing 4 blocks for faithfulness evaluation",
-)
-class EvaluateFaithfulnessBlock(BaseBlock):
-    """Thin wrapper for faithfulness evaluation using composed blocks.
-
-    Composes PromptBuilderBlock + LLMChatBlock + TextParserBlock + ColumnValueFilterBlock
-    into a single evaluation pipeline with smart parameter routing.
-
-    Parameters
-    ----------
-    block_name : str
-        Name of the block.
-    input_cols : List[str]
-        Input columns: ["document", "response"]
-    output_cols : List[str]
-        Output columns: ["faithfulness_explanation", "faithfulness_judgment"]
-    model : Optional[str]
-        LLM model identifier.
-    api_base : Optional[str]
-        API base URL.
-    api_key : Optional[str]
-        API key.
-    prompt_config_path : str
-        Path to YAML prompt template file (required).
-    **kwargs : Any
-        All other parameters are automatically routed to appropriate internal blocks
-        based on each block's accepted parameters. This includes all LLM parameters
-        (temperature, max_tokens, extra_body, extra_headers, etc.), text parser
-        parameters, and filter parameters.
-    """
-
-    model_config = ConfigDict(
-        extra="allow"
-    )  # Allow extra fields for dynamic forwarding
-
-    # --- Core configuration ---
-    prompt_config_path: str = Field(
-        ...,
-        description="Path to YAML file containing the faithfulness evaluation prompt template",
-    )
-
-    # --- LLM interface (for flow detection) ---
-    model: Optional[str] = Field(None, description="LLM model identifier")
-    api_base: Optional[str] = Field(None, description="API base URL")
-    api_key: Optional[str] = Field(None, description="API key")
-
-    # --- Filter configuration ---
-    filter_value: str = Field(
-        "YES", description="Value to filter on for faithfulness judgment"
-    )
-    operation: str = Field("eq", description="Filter operation")
-    convert_dtype: Optional[str] = Field(
-        None, description="Data type conversion for filter column"
-    )
-
-    # --- Parser configuration ---
-    start_tags: list[str] = Field(
-        ["[Start of Explanation]", "[Start of Answer]"],
-        description="Start tags for parsing explanation and judgment",
-    )
-    end_tags: list[str] = Field(
-        ["[End of Explanation]", "[End of Answer]"],
-        description="End tags for parsing explanation and judgment",
-    )
-    parsing_pattern: Optional[str] = Field(
-        None,
-        description="Regex pattern for custom parsing. If provided, takes precedence over tag-based parsing",
-    )
-
-    # --- Internal blocks (composition) ---
-    prompt_builder: PromptBuilderBlock = Field(None, exclude=True)  # type: ignore
-    llm_chat: LLMChatBlock = Field(None, exclude=True)  # type: ignore
-    text_parser: TextParserBlock = Field(None, exclude=True)  # type: ignore
-    filter_block: ColumnValueFilterBlock = Field(None, exclude=True)  # type: ignore
-
-    @field_validator("input_cols")
-    @classmethod
-    def validate_input_cols(cls, v):
-        """Validate input columns."""
-        if v != ["document", "response"]:
-            raise ValueError(
-                f"EvaluateFaithfulnessBlock expects input_cols ['document', 'response'], got {v}"
-            )
-        return v
-
-    @field_validator("output_cols")
-    @classmethod
-    def validate_output_cols(cls, v):
-        """Validate output columns."""
-        expected = ["faithfulness_explanation", "faithfulness_judgment"]
-        if v != expected:
-            raise ValueError(
-                f"EvaluateFaithfulnessBlock expects output_cols {expected}, got {v}"
-            )
-        return v
-
-    def __init__(self, **kwargs):
-        """Initialize with smart parameter routing."""
-        super().__init__(**kwargs)
-        self._create_internal_blocks(**kwargs)
-
-        # Log initialization if model is configured
-        if self.model:
-            logger.info(
-                f"Initialized EvaluateFaithfulnessBlock '{self.block_name}' with model '{self.model}'"
-            )
-
-    def _extract_params(self, kwargs: dict, block_class) -> dict:
-        """Extract parameters for specific block class based on its model_fields."""
-        # Exclude parameters that are handled by this wrapper's structure
-        wrapper_params = {
-            "block_name",
-            "input_cols",
-            "output_cols",
-        }
-
-        # Extract parameters that the target block accepts
-        params = {
-            k: v
-            for k, v in kwargs.items()
-            if k in block_class.model_fields and k not in wrapper_params
-        }
-
-        # Also include declared fields from this composite block that the target block accepts
-        for field_name in self.__class__.model_fields:
-            if (
-                field_name in block_class.model_fields
-                and field_name not in wrapper_params
-            ):
-                field_value = getattr(self, field_name)
-                if field_value is not None:  # Only forward non-None values
-                    params[field_name] = field_value
-
-        return params
-
-    def _create_internal_blocks(self, **kwargs):
-        """Create internal blocks with parameter routing."""
-        # Route parameters to appropriate blocks
-        prompt_params = self._extract_params(kwargs, PromptBuilderBlock)
-        llm_params = self._extract_params(kwargs, LLMChatBlock)
-        parser_params = self._extract_params(kwargs, TextParserBlock)
-        filter_params = self._extract_params(kwargs, ColumnValueFilterBlock)
-
-        self.prompt_builder = PromptBuilderBlock(
-            block_name=f"{self.block_name}_prompt_builder",
-            input_cols=["document", "response"],
-            output_cols=["eval_faithfulness_prompt"],
-            **prompt_params,
-        )
-
-        # Create LLM chat block with dynamic LLM parameter forwarding
-        llm_config = {
-            "block_name": f"{self.block_name}_llm_chat",
-            "input_cols": ["eval_faithfulness_prompt"],
-            "output_cols": ["raw_eval_faithfulness"],
-            **llm_params,
-        }
-
-        # Only add LLM parameters if they are provided
-        if self.model is not None:
-            llm_config["model"] = self.model
-        if self.api_base is not None:
-            llm_config["api_base"] = self.api_base
-        if self.api_key is not None:
-            llm_config["api_key"] = self.api_key
-
-        self.llm_chat = LLMChatBlock(**llm_config)
-
-        # Create text parser
-        self.text_parser = TextParserBlock(
-            block_name=f"{self.block_name}_text_parser",
-            input_cols=["raw_eval_faithfulness"],
-            output_cols=["faithfulness_explanation", "faithfulness_judgment"],
-            **parser_params,
-        )
-
-        self.filter_block = ColumnValueFilterBlock(
-            block_name=f"{self.block_name}_filter",
-            input_cols=["faithfulness_judgment"],
-            output_cols=[],  # Filter doesn't create new columns
-            **filter_params,
-        )
-
-    def generate(self, samples: Dataset, **kwargs: Any) -> Dataset:
-        """Execute the 4-block faithfulness evaluation pipeline.
-
-        Parameters
-        ----------
-        samples : Dataset
-            Input dataset with 'document' and 'response' columns.
-        **kwargs : Any
-            Additional arguments passed to internal blocks.
-
-        Returns
-        -------
-        Dataset
-            Filtered dataset with faithfulness evaluation results.
-        """
-        # Validate model is configured
-        if not self.model:
-            raise BlockValidationError(
-                f"Model not configured for block '{self.block_name}'. "
-                f"Call flow.set_model_config() before generating."
-            )
-
-        logger.info(
-            f"Starting faithfulness evaluation for {len(samples)} samples",
-            extra={"block_name": self.block_name, "model": self.model},
-        )
-
-        try:
-            # Execute 4-block pipeline with validation delegation
-            result = self.prompt_builder(samples, **kwargs)
-            result = self.llm_chat(result, **kwargs)
-            result = self.text_parser(result, **kwargs)
-            result = self.filter_block(result, **kwargs)
-
-            logger.info(
-                f"Faithfulness evaluation completed: {len(samples)} → {len(result)} samples",
-                extra={"block_name": self.block_name},
-            )
-
-            return result
-
-        except Exception as e:
-            logger.error(
-                f"Error during faithfulness evaluation: {e}",
-                extra={"block_name": self.block_name, "error": str(e)},
-            )
-            raise
-
-    def __getattr__(self, name: str) -> Any:
-        """Forward attribute access to appropriate internal block."""
-        # Check each internal block to see which one has this parameter
-        for block_attr, block_class in [
-            ("prompt_builder", PromptBuilderBlock),
-            ("llm_chat", LLMChatBlock),
-            ("text_parser", TextParserBlock),
-            ("filter_block", ColumnValueFilterBlock),
-        ]:
-            if hasattr(self, block_attr) and name in block_class.model_fields:
-                internal_block = getattr(self, block_attr)
-                if internal_block is not None:
-                    return getattr(internal_block, name)
-        raise AttributeError(
-            f"'{self.__class__.__name__}' object has no attribute '{name}'"
-        )
-
-    def __setattr__(self, name: str, value: Any) -> None:
-        """Handle dynamic parameter updates from flow.set_model_config()."""
-        super().__setattr__(name, value)
-
-        # Forward to appropriate internal blocks
-        for block_attr, block_class in [
-            ("prompt_builder", PromptBuilderBlock),
-            ("llm_chat", LLMChatBlock),
-            ("text_parser", TextParserBlock),
-            ("filter_block", ColumnValueFilterBlock),
-        ]:
-            if hasattr(self, block_attr) and name in block_class.model_fields:
-                setattr(getattr(self, block_attr), name, value)
-
-    def _reinitialize_client_manager(self) -> None:
-        """Reinitialize internal LLM block's client manager."""
-        if hasattr(self.llm_chat, "_reinitialize_client_manager"):
-            self.llm_chat._reinitialize_client_manager()
-
-    def get_internal_blocks_info(self) -> dict[str, Any]:
-        """Get information about internal blocks."""
-        return {
-            "prompt_builder": self.prompt_builder.get_info(),
-            "llm_chat": self.llm_chat.get_info(),
-            "text_parser": self.text_parser.get_info(),
-            "filter": self.filter_block.get_info(),
-        }
-
-    def __repr__(self) -> str:
-        """String representation of the block."""
-        filter_value = (
-            getattr(self.filter_block, "filter_value", "YES")
-            if hasattr(self, "filter_block")
-            else "YES"
-        )
-        return (
-            f"EvaluateFaithfulnessBlock(name='{self.block_name}', "
-            f"model='{self.model}', filter_value='{filter_value}')"
-        )