chore(generative-ai): update gemini model version in count token examples (#12683)

* chore(generative-ai): update count token example to use Gemini Flash's latest version 002

* fix(generative-ai): function names in test cases & update requirements.txt

* chore: updates function name

function name `test_gemini_count_token_example` is already in use.

* 🦉 Updates from OwlBot post-processor

See https://github.com/googleapis/repo-automation-bots/blob/main/packages/owl-bot/README.md

* feat: clean up test-file

* feat: clean up test-file

* feat: clean up test-file

---------

Co-authored-by: Owl Bot <gcf-owl-bot[bot]@users.noreply.github.com>

Author: msampathkumar

generative_ai/embeddings/generate_embeddings_with_lower_dimension.py

@@ -41,7 +41,8 @@ def generate_embeddings_with_lower_dimension() -> MultiModalEmbeddingResponse:
 
     model = MultiModalEmbeddingModel.from_pretrained("multimodalembedding@001")
     image = Image.load_from_file(
-        "gs://cloud-samples-data/vertex-ai/llm/prompts/landmark1.png")
+        "gs://cloud-samples-data/vertex-ai/llm/prompts/landmark1.png"
+    )
 
     embeddings = model.get_embeddings(
         image=image,

generative_ai/embeddings/test_embeddings.py

@@ -69,7 +69,9 @@ def test_multimodal_embedding_image() -> None:
 
 @backoff.on_exception(backoff.expo, ResourceExhausted, max_time=10)
 def test_generate_embeddings_with_lower_dimension() -> None:
-    embeddings = generate_embeddings_with_lower_dimension.generate_embeddings_with_lower_dimension()
+    embeddings = (
+        generate_embeddings_with_lower_dimension.generate_embeddings_with_lower_dimension()
+    )
     assert embeddings is not None
     assert embeddings.image_embedding is not None
     assert len(embeddings.image_embedding) == 128

generative_ai/requirements.txt

@@ -3,7 +3,7 @@ pandas==2.0.3; python_version == '3.8'
 pandas==2.1.4; python_version > '3.8'
 pillow==10.3.0; python_version < '3.8'
 pillow==10.3.0; python_version >= '3.8'
-google-cloud-aiplatform[all]==1.64.0
+google-cloud-aiplatform[all]==1.69.0
 sentencepiece==0.2.0
 google-auth==2.29.0
 anthropic[vertex]==0.28.0

generative_ai/token_count/api_example.py

@@ -16,36 +16,35 @@
 PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
 
 
-def count_token_service() -> int:
+def count_token_api_example() -> int:
     # [START generativeaionvertexai_token_count_sample_with_genai]
     import vertexai
     from vertexai.generative_models import GenerativeModel
 
-    # TODO(developer): Update and un-comment below line
-    # PROJECT_ID = "your-project-id"
+    # TODO(developer): Update project & location
     vertexai.init(project=PROJECT_ID, location="us-central1")
 
     # using Vertex AI Model as tokenzier
-    model = GenerativeModel("gemini-1.5-flash")
+    model = GenerativeModel("gemini-1.5-flash-002")
 
     prompt = "hello world"
     response = model.count_tokens(prompt)
     print(f"Prompt Token Count: {response.total_tokens}")
     print(f"Prompt Character Count: {response.total_billable_characters}")
+    # Example response:
+    #     Prompt Token Count: 2
+    #     Prompt Token Count: 10
 
     prompt = ["hello world", "what's the weather today"]
     response = model.count_tokens(prompt)
     print(f"Prompt Token Count: {response.total_tokens}")
     print(f"Prompt Character Count: {response.total_billable_characters}")
     # Example response:
-    # Prompt Token Count: 2
-    # Prompt Character Count: 10
-    # Prompt Token Count: 8
-    # Prompt Character Count: 31
-
+    #     Prompt Token Count: 8
+    #     Prompt Token Count: 31
     # [END generativeaionvertexai_token_count_sample_with_genai]
     return response.total_tokens
 
 
 if __name__ == "__main__":
-    count_token_service()
+    count_token_api_example()

generative_ai/token_count/local_sdk_example.py

@@ -13,27 +13,28 @@
 # limitations under the License.
 
 
-def count_token_locally() -> int:
+def local_tokenizer_example() -> int:
     # [START generativeaionvertexai_token_count_sample_with_local_sdk]
     from vertexai.preview.tokenization import get_tokenizer_for_model
 
     # Using local tokenzier
-    tokenizer = get_tokenizer_for_model("gemini-1.5-flash")
+    tokenizer = get_tokenizer_for_model("gemini-1.5-flash-002")
 
     prompt = "hello world"
     response = tokenizer.count_tokens(prompt)
     print(f"Prompt Token Count: {response.total_tokens}")
+    # Example response:
+    # Prompt Token Count: 2
 
     prompt = ["hello world", "what's the weather today"]
     response = tokenizer.count_tokens(prompt)
     print(f"Prompt Token Count: {response.total_tokens}")
     # Example response:
-    # Prompt Token Count: 2
     # Prompt Token Count: 8
 
     # [END generativeaionvertexai_token_count_sample_with_local_sdk]
     return response.total_tokens
 
 
 if __name__ == "__main__":
-    count_token_locally()
+    local_tokenizer_example()

generative_ai/token_count/multimodal_token_count_example.py

@@ -18,7 +18,7 @@
 PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
 
 
-def count_tokens_multimodal() -> GenerationResponse:
+def count_tokens_multimodal_example() -> GenerationResponse:
     # [START generativeaionvertexai_gemini_token_count_multimodal]
     import vertexai
     from vertexai.generative_models import GenerativeModel, Part
@@ -37,29 +37,30 @@ def count_tokens_multimodal() -> GenerationResponse:
         "Provide a description of the video.",
     ]
 
-    # Prompt tokens count
+    # tokens count for user prompt
     response = model.count_tokens(contents)
     print(f"Prompt Token Count: {response.total_tokens}")
     print(f"Prompt Character Count: {response.total_billable_characters}")
+    # Example response:
+    #     Prompt Token Count: 16822
+    #     Prompt Character Count: 30
 
     # Send text to Gemini
     response = model.generate_content(contents)
     usage_metadata = response.usage_metadata
 
-    # Response tokens count
+    # tokens count for model response
     print(f"Prompt Token Count: {usage_metadata.prompt_token_count}")
     print(f"Candidates Token Count: {usage_metadata.candidates_token_count}")
     print(f"Total Token Count: {usage_metadata.total_token_count}")
     # Example response:
-    # Prompt Token Count: 16822
-    # Prompt Character Count: 30
-    # Prompt Token Count: 16822
-    # Candidates Token Count: 71
-    # Total Token Count: 16893
+    #     Prompt Token Count: 16822
+    #     Candidates Token Count: 71
+    #     Total Token Count: 16893
 
     # [END generativeaionvertexai_gemini_token_count_multimodal]
     return response
 
 
 if __name__ == "__main__":
-    count_tokens_multimodal()
+    count_tokens_multimodal_example()

generative_ai/token_count/simple_example.py

@@ -18,7 +18,7 @@
 PROJECT_ID = os.getenv("GOOGLE_CLOUD_PROJECT")
 
 
-def count_tokens() -> GenerationResponse:
+def count_token_example() -> GenerationResponse:
     # [START generativeaionvertexai_gemini_token_count]
     import vertexai
     from vertexai.generative_models import GenerativeModel
@@ -55,4 +55,4 @@ def count_tokens() -> GenerationResponse:
 
 
 if __name__ == "__main__":
-    count_tokens()
+    count_token_example()

generative_ai/token_count/token_count_test.py renamed to generative_ai/token_count/test_token_count_examples.py

@@ -18,16 +18,19 @@
 import simple_example
 
 
-def test_count_token() -> None:
-    assert local_sdk_example.count_token_locally()
-    assert api_example.count_token_service()
+def test_local_sdk_example() -> None:
+    assert local_sdk_example.local_tokenizer_example()
+    assert api_example.count_token_api_example()
 
 
-def test_gemini_count_token_example() -> None:
-    response = simple_example.count_tokens()
+def test_simple_example() -> None:
+    response = simple_example.count_token_example()
     assert response
     assert response.usage_metadata
 
-    response = multimodal_token_count_example.count_tokens_multimodal()
+
+def test_multimodal_example() -> None:
+    print(dir(multimodal_token_count_example))
+    response = multimodal_token_count_example.count_tokens_multimodal_example()
     assert response
     assert response.usage_metadata