pygeai 0.1.51b3__py3-none-any.whl → 0.6.0b15__py3-none-any.whl

pygeai/tests/snippets/embeddings/cache_example.py

@@ -0,0 +1,31 @@
+from pygeai.core.embeddings.managers import EmbeddingsManager
+from pygeai.core.embeddings.models import EmbeddingConfiguration
+from pygeai.core.services.llm.model import Model
+from pygeai.core.services.llm.providers import Provider
+import time
+
+manager = EmbeddingsManager()
+
+inputs = ["This text will be cached for faster repeated access"]
+
+configuration = EmbeddingConfiguration(
+    inputs=inputs,
+    model=f"{Provider.OPENAI}/{Model.OpenAI.TEXT_EMBEDDING_3_SMALL}",
+    cache=True
+)
+
+print("First request (not cached)...")
+start = time.time()
+embeddings1 = manager.generate_embeddings(configuration)
+elapsed1 = time.time() - start
+print(f"Time: {elapsed1:.3f}s")
+print(f"Cost: ${embeddings1.usage.total_cost} {embeddings1.usage.currency}")
+
+print("\nSecond request (should be cached)...")
+start = time.time()
+embeddings2 = manager.generate_embeddings(configuration)
+elapsed2 = time.time() - start
+print(f"Time: {elapsed2:.3f}s")
+print(f"Cost: ${embeddings2.usage.total_cost} {embeddings2.usage.currency}")
+
+print(f"\nCache speedup: {elapsed1/elapsed2:.2f}x faster")

pygeai/tests/snippets/embeddings/cohere_example.py

@@ -0,0 +1,41 @@
+from pygeai.core.embeddings.managers import EmbeddingsManager
+from pygeai.core.embeddings.models import EmbeddingConfiguration
+from pygeai.core.services.llm.model import Model
+from pygeai.core.services.llm.providers import Provider
+
+manager = EmbeddingsManager()
+
+document_inputs = [
+    "Machine learning is a subset of artificial intelligence",
+    "Neural networks are inspired by the human brain"
+]
+
+configuration = EmbeddingConfiguration(
+    inputs=document_inputs,
+    model=f"{Provider.COHERE}/{Model.Cohere.EMBED_ENGLISH_V3_0}",
+    input_type="search_document",
+    timeout=600,
+    cache=False
+)
+
+embeddings = manager.generate_embeddings(configuration)
+
+print(f"Model: {embeddings.model}")
+print(f"Input type: search_document")
+print(f"Documents embedded: {len(embeddings.data)}")
+print(f"Usage - Prompt tokens: {embeddings.usage.prompt_tokens}")
+print(f"Usage - Total cost: ${embeddings.usage.total_cost} {embeddings.usage.currency}")
+
+query_input = ["What is neural network?"]
+
+query_configuration = EmbeddingConfiguration(
+    inputs=query_input,
+    model=f"{Provider.COHERE}/{Model.Cohere.EMBED_ENGLISH_V3_0}",
+    input_type="search_query",
+    timeout=600,
+    cache=False
+)
+
+query_embeddings = manager.generate_embeddings(query_configuration)
+print(f"\nQuery embedded with input_type: search_query")
+print(f"Usage - Total cost: ${query_embeddings.usage.total_cost} {query_embeddings.usage.currency}")

pygeai/tests/snippets/embeddings/openai_base64_example.py

@@ -0,0 +1,27 @@
+from pygeai.core.embeddings.managers import EmbeddingsManager
+from pygeai.core.embeddings.models import EmbeddingConfiguration
+from pygeai.core.services.llm.model import Model
+from pygeai.core.services.llm.providers import Provider
+
+manager = EmbeddingsManager()
+
+inputs = ["Text to embed with base64 encoding"]
+
+configuration = EmbeddingConfiguration(
+    inputs=inputs,
+    model=f"{Provider.OPENAI}/{Model.OpenAI.TEXT_EMBEDDING_3_SMALL}",
+    encoding_format="base64",
+    dimensions=512,
+    user=None,
+    input_type=None,
+    timeout=600,
+    cache=False
+)
+
+embeddings = manager.generate_embeddings(configuration)
+
+print(f"Model: {embeddings.model}")
+print(f"Encoding format: base64")
+print(f"Embedding (base64 encoded, first 100 chars): {embeddings.data[0].embedding[:100]}")
+print(f"Type of embedding: {type(embeddings.data[0].embedding)}")
+print(f"Usage - Total cost: ${embeddings.usage.total_cost} {embeddings.usage.currency}")

pygeai/tests/snippets/embeddings/openai_example.py

@@ -0,0 +1,30 @@
+from pygeai.core.embeddings.managers import EmbeddingsManager
+from pygeai.core.embeddings.models import EmbeddingConfiguration
+from pygeai.core.services.llm.model import Model
+from pygeai.core.services.llm.providers import Provider
+
+manager = EmbeddingsManager()
+
+inputs = [
+    "The quick brown fox jumps over the lazy dog",
+    "Python is a high-level programming language"
+]
+
+configuration = EmbeddingConfiguration(
+    inputs=inputs,
+    model=f"{Provider.OPENAI}/{Model.OpenAI.TEXT_EMBEDDING_3_SMALL}",
+    encoding_format="float",
+    dimensions=1536,
+    user=None,
+    input_type=None,
+    timeout=600,
+    cache=False
+)
+
+embeddings = manager.generate_embeddings(configuration)
+
+print(f"Model: {embeddings.model}")
+print(f"Total embeddings generated: {len(embeddings.data)}")
+print(f"Embedding dimensions: {len(embeddings.data[0].embedding)}")
+print(f"Usage - Prompt tokens: {embeddings.usage.prompt_tokens}")
+print(f"Usage - Total cost: ${embeddings.usage.total_cost} {embeddings.usage.currency}")

pygeai/tests/snippets/embeddings/similarity_example.py

@@ -0,0 +1,42 @@
+from pygeai.core.embeddings.managers import EmbeddingsManager
+from pygeai.core.embeddings.models import EmbeddingConfiguration
+from pygeai.core.services.llm.model import Model
+from pygeai.core.services.llm.providers import Provider
+import numpy as np
+
+def cosine_similarity(vec1, vec2):
+    """Calculate cosine similarity between two vectors."""
+    vec1 = np.array(vec1)
+    vec2 = np.array(vec2)
+    return np.dot(vec1, vec2) / (np.linalg.norm(vec1) * np.linalg.norm(vec2))
+
+manager = EmbeddingsManager()
+
+texts = [
+    "The cat sits on the mat",
+    "A feline rests on a rug",
+    "Python is a programming language"
+]
+
+configuration = EmbeddingConfiguration(
+    inputs=texts,
+    model=f"{Provider.OPENAI}/{Model.OpenAI.TEXT_EMBEDDING_3_SMALL}",
+)
+
+embeddings = manager.generate_embeddings(configuration)
+
+embedding_vectors = [emb.embedding for emb in embeddings.data]
+
+print("Cosine Similarity Matrix:")
+print("=" * 60)
+for i, text1 in enumerate(texts):
+    for j, text2 in enumerate(texts):
+        if i <= j:
+            similarity = cosine_similarity(embedding_vectors[i], embedding_vectors[j])
+            print(f"\nText {i+1} vs Text {j+1}: {similarity:.4f}")
+            if i != j:
+                print(f"  '{text1[:40]}...'")
+                print(f"  '{text2[:40]}...'")
+
+print(f"\n{'='*60}")
+print(f"Total cost: ${embeddings.usage.total_cost} {embeddings.usage.currency}")

pygeai/tests/snippets/evaluation/dataset/complete_workflow_example.py

@@ -0,0 +1,195 @@
+"""
+Complete Dataset API Workflow Example
+
+This example demonstrates a full workflow using the Dataset API:
+1. Create a dataset
+2. Add rows
+3. Add expected sources
+4. Add filter variables
+5. Query and update
+6. Cleanup
+
+NOTE: API documentation at https://docs.globant.ai/en/wiki?839,DataSet+API contains a typo.
+The field 'dataSetExpectedSourceExtension' is spelled correctly in this code.
+"""
+
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+
+def main():
+    # Initialize client
+    client = EvaluationDatasetClient()
+    
+    print("="*60)
+    print("Dataset API Complete Workflow Example")
+    print("="*60)
+    
+    # Step 1: Create a dataset
+    print("\n[1/7] Creating dataset...")
+    dataset = client.create_dataset(
+        dataset_name="AI Knowledge Base - Example",
+        dataset_description="A sample dataset for AI-related questions",
+        dataset_type="T",
+        dataset_active=True,
+        rows=[]
+    )
+    dataset_id = dataset.get('dataSetId')
+    print(f"✓ Created dataset: {dataset_id}")
+    
+    # Step 2: Add rows to the dataset
+    print("\n[2/7] Adding rows...")
+    row1 = client.create_dataset_row(
+        dataset_id=dataset_id,
+        row={
+            "dataSetRowInput": "What is Machine Learning?",
+            "dataSetRowExpectedAnswer": "Machine Learning is a subset of AI that enables systems to learn from data.",
+            "dataSetRowContextDocument": "ML is a fundamental concept in artificial intelligence."
+        }
+    )
+    row1_id = row1.get('dataSetRowId')
+    print(f"✓ Created row 1: {row1_id}")
+    
+    row2 = client.create_dataset_row(
+        dataset_id=dataset_id,
+        row={
+            "dataSetRowInput": "What is Deep Learning?",
+            "dataSetRowExpectedAnswer": "Deep Learning is a subset of ML using neural networks with multiple layers.",
+            "dataSetRowContextDocument": "DL has revolutionized AI in recent years."
+        }
+    )
+    row2_id = row2.get('dataSetRowId')
+    print(f"✓ Created row 2: {row2_id}")
+    
+    # Step 3: Add expected sources to row 1
+    print("\n[3/7] Adding expected sources to row 1...")
+    source1 = client.create_expected_source(
+        dataset_id=dataset_id,
+        dataset_row_id=row1_id,
+        expected_source_name="Introduction to Machine Learning",
+        expected_source_value="Machine Learning is a method of data analysis that automates analytical model building.",
+        expected_source_extension="pdf"
+    )
+    source1_id = source1.get('dataSetExpectedSourceId')
+    print(f"✓ Created expected source: {source1_id}")
+    
+    # Step 4: Add filter variables to row 1
+    print("\n[4/7] Adding filter variables to row 1...")
+    filter1 = client.create_filter_variable(
+        dataset_id=dataset_id,
+        dataset_row_id=row1_id,
+        metadata_type="V",
+        filter_variable_key="category",
+        filter_variable_value="machine-learning",
+        filter_variable_operator="="
+    )
+    filter1_id = filter1.get('dataSetRowFilterVarId')
+    print(f"✓ Created filter variable: {filter1_id}")
+    
+    filter2 = client.create_filter_variable(
+        dataset_id=dataset_id,
+        dataset_row_id=row1_id,
+        metadata_type="V",
+        filter_variable_key="difficulty",
+        filter_variable_value="beginner",
+        filter_variable_operator="="
+    )
+    filter2_id = filter2.get('dataSetRowFilterVarId')
+    print(f"✓ Created filter variable: {filter2_id}")
+    
+    # Step 5: Query the dataset
+    print("\n[5/7] Querying dataset...")
+    
+    # Get full dataset
+    full_dataset = client.get_dataset(dataset_id=dataset_id)
+    print(f"✓ Dataset has {len(full_dataset.get('rows', []))} rows")
+    
+    # List all rows
+    all_rows = client.list_dataset_rows(dataset_id=dataset_id)
+    print(f"✓ Retrieved {len(all_rows) if isinstance(all_rows, list) else 'N/A'} rows")
+    
+    # Get specific row with all details
+    row1_details = client.get_dataset_row(dataset_id=dataset_id, dataset_row_id=row1_id)
+    print(f"✓ Row 1 has {len(row1_details.get('expectedSources', []))} expected sources")
+    print(f"✓ Row 1 has {len(row1_details.get('filterVariables', []))} filter variables")
+    
+    # Step 6: Update operations
+    print("\n[6/7] Updating resources...")
+    
+    # Update the dataset description
+    updated_dataset = client.update_dataset(
+        dataset_id=dataset_id,
+        dataset_name="AI Knowledge Base - Updated",
+        dataset_description="Updated description for AI dataset",
+        dataset_type="T",
+        dataset_active=True
+    )
+    print(f"✓ Updated dataset description")
+    
+    # Update a row
+    updated_row = client.update_dataset_row(
+        dataset_id=dataset_id,
+        dataset_row_id=row2_id,
+        row={
+            "dataSetRowInput": "What is Deep Learning? (Updated)",
+            "dataSetRowExpectedAnswer": "Deep Learning uses neural networks with many layers to learn complex patterns.",
+            "dataSetRowContextDocument": "DL is particularly effective for image and speech recognition."
+        }
+    )
+    print(f"✓ Updated row 2")
+    
+    # Update expected source
+    updated_source = client.update_expected_source(
+        dataset_id=dataset_id,
+        dataset_row_id=row1_id,
+        expected_source_id=source1_id,
+        expected_source_name="ML Fundamentals - Updated",
+        expected_source_value="Updated source content about Machine Learning.",
+        expected_source_extension="txt"
+    )
+    print(f"✓ Updated expected source")
+    
+    # Update filter variable
+    updated_filter = client.update_filter_variable(
+        dataset_id=dataset_id,
+        dataset_row_id=row1_id,
+        filter_variable_id=filter2_id,
+        metadata_type="V",
+        filter_variable_key="difficulty",
+        filter_variable_value="intermediate",
+        filter_variable_operator="="
+    )
+    print(f"✓ Updated filter variable")
+    
+    # Step 7: Cleanup
+    print("\n[7/7] Cleaning up...")
+    
+    # Delete filter variables
+    client.delete_filter_variable(dataset_id=dataset_id, dataset_row_id=row1_id, filter_variable_id=filter1_id)
+    client.delete_filter_variable(dataset_id=dataset_id, dataset_row_id=row1_id, filter_variable_id=filter2_id)
+    print(f"✓ Deleted filter variables")
+    
+    # Delete expected source
+    client.delete_expected_source(dataset_id=dataset_id, dataset_row_id=row1_id, expected_source_id=source1_id)
+    print(f"✓ Deleted expected source")
+    
+    # Delete rows
+    client.delete_dataset_row(dataset_id=dataset_id, dataset_row_id=row1_id)
+    client.delete_dataset_row(dataset_id=dataset_id, dataset_row_id=row2_id)
+    print(f"✓ Deleted rows")
+    
+    # Delete dataset
+    client.delete_dataset(dataset_id=dataset_id)
+    print(f"✓ Deleted dataset")
+    
+    print("\n" + "="*60)
+    print("Workflow completed successfully!")
+    print("="*60)
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except Exception as e:
+        print(f"\n❌ Error occurred: {e}")
+        import traceback
+        traceback.print_exc()

pygeai/tests/snippets/evaluation/dataset/create_dataset.py

@@ -0,0 +1,26 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+response = client.create_dataset(
+    dataset_name="My Test Dataset",
+    dataset_description="A dataset for testing purposes",
+    dataset_type="T",
+    dataset_active=True,
+    rows=[
+        {
+            "dataSetRowInput": "What is the capital of France?",
+            "dataSetRowExpectedAnswer": "Paris",
+            "dataSetRowContextDocument": "France is a country in Europe."
+        },
+        {
+            "dataSetRowInput": "What is 2+2?",
+            "dataSetRowExpectedAnswer": "4",
+            "dataSetRowContextDocument": ""
+        }
+    ]
+)
+
+print(f"Created dataset: {response}")
+dataset_id = response.get('dataSetId')
+print(f"Dataset ID: {dataset_id}")

pygeai/tests/snippets/evaluation/dataset/create_dataset_from_file.py

@@ -0,0 +1,11 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+file_path = "path/to/dataset.json"
+
+response = client.create_dataset_from_file(file_path=file_path)
+
+print(f"Created dataset from file: {response}")
+dataset_id = response.get('dataSetId')
+print(f"Dataset ID: {dataset_id}")

pygeai/tests/snippets/evaluation/dataset/create_dataset_row.py

@@ -0,0 +1,17 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+
+row = {
+    "dataSetRowInput": "What is AI?",
+    "dataSetRowExpectedAnswer": "Artificial Intelligence is the simulation of human intelligence by machines.",
+    "dataSetRowContextDocument": "AI enables computers to perform tasks that typically require human intelligence."
+}
+
+response = client.create_dataset_row(dataset_id=dataset_id, row=row)
+
+print(f"Created row: {response}")
+row_id = response.get('dataSetRowId')
+print(f"Row ID: {row_id}")

pygeai/tests/snippets/evaluation/dataset/create_expected_source.py

@@ -0,0 +1,18 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.create_expected_source(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    expected_source_name="Documentation PDF",
+    expected_source_value="This is the content from the source document...",
+    expected_source_extension="pdf"
+)
+
+print(f"Created expected source: {response}")
+source_id = response.get('dataSetExpectedSourceId')
+print(f"Expected Source ID: {source_id}")

pygeai/tests/snippets/evaluation/dataset/create_filter_variable.py

@@ -0,0 +1,19 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.create_filter_variable(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    metadata_type="V",
+    filter_variable_key="category",
+    filter_variable_value="science",
+    filter_variable_operator="="
+)
+
+print(f"Created filter variable: {response}")
+filter_var_id = response.get('dataSetRowFilterVarId')
+print(f"Filter Variable ID: {filter_var_id}")

pygeai/tests/snippets/evaluation/dataset/delete_dataset.py

@@ -0,0 +1,9 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+
+response = client.delete_dataset(dataset_id=dataset_id)
+
+print(f"Deleted dataset: {response}")

pygeai/tests/snippets/evaluation/dataset/delete_dataset_row.py

@@ -0,0 +1,10 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.delete_dataset_row(dataset_id=dataset_id, dataset_row_id=row_id)
+
+print(f"Deleted row: {response}")

pygeai/tests/snippets/evaluation/dataset/delete_expected_source.py

@@ -0,0 +1,15 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+source_id = "your-expected-source-id-here"
+
+response = client.delete_expected_source(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    expected_source_id=source_id
+)
+
+print(f"Deleted expected source: {response}")

pygeai/tests/snippets/evaluation/dataset/delete_filter_variable.py

@@ -0,0 +1,15 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+filter_var_id = "your-filter-variable-id-here"
+
+response = client.delete_filter_variable(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    filter_variable_id=filter_var_id
+)
+
+print(f"Deleted filter variable: {response}")

pygeai/tests/snippets/evaluation/dataset/get_dataset.py

@@ -0,0 +1,9 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+
+response = client.get_dataset(dataset_id=dataset_id)
+
+print(f"Dataset details: {response}")

pygeai/tests/snippets/evaluation/dataset/get_dataset_row.py

@@ -0,0 +1,10 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.get_dataset_row(dataset_id=dataset_id, dataset_row_id=row_id)
+
+print(f"Row details: {response}")

pygeai/tests/snippets/evaluation/dataset/get_expected_source.py

@@ -0,0 +1,15 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+source_id = "your-expected-source-id-here"
+
+response = client.get_expected_source(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    expected_source_id=source_id
+)
+
+print(f"Expected source details: {response}")

pygeai/tests/snippets/evaluation/dataset/get_filter_variable.py

@@ -0,0 +1,15 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+filter_var_id = "your-filter-variable-id-here"
+
+response = client.get_filter_variable(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    filter_variable_id=filter_var_id
+)
+
+print(f"Filter variable details: {response}")

pygeai/tests/snippets/evaluation/dataset/list_dataset_rows.py

@@ -0,0 +1,9 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+
+response = client.list_dataset_rows(dataset_id=dataset_id)
+
+print(f"Dataset rows: {response}")

pygeai/tests/snippets/evaluation/dataset/list_datasets.py

@@ -0,0 +1,6 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+response = client.list_datasets()
+print(f"Datasets: {response}")

pygeai/tests/snippets/evaluation/dataset/list_expected_sources.py

@@ -0,0 +1,10 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.list_expected_sources(dataset_id=dataset_id, dataset_row_id=row_id)
+
+print(f"Expected sources: {response}")

pygeai/tests/snippets/evaluation/dataset/list_filter_variables.py

@@ -0,0 +1,10 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+response = client.list_filter_variables(dataset_id=dataset_id, dataset_row_id=row_id)
+
+print(f"Filter variables: {response}")

pygeai/tests/snippets/evaluation/dataset/update_dataset.py

@@ -0,0 +1,15 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+
+response = client.update_dataset(
+    dataset_id=dataset_id,
+    dataset_name="Updated Dataset Name",
+    dataset_description="Updated description",
+    dataset_type="E",
+    dataset_active=False
+)
+
+print(f"Updated dataset: {response}")

pygeai/tests/snippets/evaluation/dataset/update_dataset_row.py

@@ -0,0 +1,20 @@
+from pygeai.evaluation.dataset.clients import EvaluationDatasetClient
+
+client = EvaluationDatasetClient()
+
+dataset_id = "your-dataset-id-here"
+row_id = "your-row-id-here"
+
+updated_row = {
+    "dataSetRowInput": "Updated question: What is Machine Learning?",
+    "dataSetRowExpectedAnswer": "Updated answer: ML is a subset of AI",
+    "dataSetRowContextDocument": "Updated context about ML"
+}
+
+response = client.update_dataset_row(
+    dataset_id=dataset_id,
+    dataset_row_id=row_id,
+    row=updated_row
+)
+
+print(f"Updated row: {response}")