fix(generative-ai): update text-bison stable version to 002 (GoogleCloudPlatform#11229)

gericdong · web-flow · commit 18632ae947d9 · 2024-02-22T15:46:06.000+01:00
diff --git a/generative_ai/classify_news_items.py b/generative_ai/classify_news_items.py
@@ -27,7 +27,7 @@ def classify_news_items(temperature: float = 0.2) -> None:
         "top_k": 1,  # A top_k of 1 means the selected token is the most probable among all tokens.
     }
 
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     response = model.predict(
         """What is the topic for a given news headline?
 - business
diff --git a/generative_ai/evaluate_model.py b/generative_ai/evaluate_model.py
@@ -34,7 +34,7 @@ def evaluate_model(
     vertexai.init(project=project_id, location=location, credentials=credentials)
 
     # Create a reference to a generative AI model
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
 
     # Define the evaluation specification for a text classification task
     task_spec = EvaluationTextClassificationSpec(
diff --git a/generative_ai/extraction.py b/generative_ai/extraction.py
@@ -33,7 +33,7 @@ def extractive_question_answering(
         "top_k": 1,  # A top_k of 1 means the selected token is the most probable among all tokens.
     }
 
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     response = model.predict(
         """Background: There is evidence that there have been significant changes \
 in Amazon rainforest vegetation over the last 21,000 years through the Last \
diff --git a/generative_ai/ideation.py b/generative_ai/ideation.py
@@ -33,7 +33,7 @@ def interview(
         "top_k": 40,  # A top_k of 1 means the selected token is the most probable among all tokens.
     }
 
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     response = model.predict(
         "Give me ten interview questions for the role of program manager.",
         **parameters,
diff --git a/generative_ai/list_tuned_models.py b/generative_ai/list_tuned_models.py
@@ -25,7 +25,7 @@ def list_tuned_models(
     """List tuned models."""
 
     vertexai.init(project=project_id, location=location)
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     tuned_model_names = model.list_tuned_model_names()
     print(tuned_model_names)
 
diff --git a/generative_ai/sentiment_analysis.py b/generative_ai/sentiment_analysis.py
@@ -33,7 +33,7 @@ def sentiment_analysis(
         "top_k": 1,  # A top_k of 1 means the selected token is the most probable among all tokens.
     }
 
-    model = TextGenerationModel.from_pretrained("google/text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     response = model.predict(
         """I had to compare two versions of Hamlet for my Shakespeare class and \
 unfortunately I picked this version. Everything from the acting (the actors \
diff --git a/generative_ai/summarization.py b/generative_ai/summarization.py
@@ -33,7 +33,7 @@ def text_summarization(
         "top_k": 40,  # A top_k of 1 means the selected token is the most probable among all tokens.
     }
 
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
     response = model.predict(
         """Provide a summary with about two sentences for the following article:
 The efficient-market hypothesis (EMH) is a hypothesis in financial \
diff --git a/generative_ai/tuning.py b/generative_ai/tuning.py
@@ -69,7 +69,7 @@ def tuning(
     eval_spec.tensorboard = aiplatform.Tensorboard(
         tensorboard_name=tensorboard_instance_name
     )
-    model = TextGenerationModel.from_pretrained("text-bison@001")
+    model = TextGenerationModel.from_pretrained("text-bison@002")
 
     model.tune_model(
         training_data=training_data,

Original file line number	Diff line number	Diff line change
`@@ -27,7 +27,7 @@ def classify_news_items(temperature: float = 0.2) -> None:`
`27`	`27`	`"top_k": 1, # A top_k of 1 means the selected token is the most probable among all tokens.`
`28`	`28`	`}`
`29`	`29`
`30`		`- model = TextGenerationModel.from_pretrained("text-bison@001")`
	`30`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`31`	`31`	`response = model.predict(`
`32`	`32`	`"""What is the topic for a given news headline?`
`33`	`33`	`- business`
Original file line number	Diff line number	Diff line change
`@@ -33,7 +33,7 @@ def extractive_question_answering(`
`33`	`33`	`"top_k": 1, # A top_k of 1 means the selected token is the most probable among all tokens.`
`34`	`34`	`}`
`35`	`35`
`36`		`- model = TextGenerationModel.from_pretrained("text-bison@001")`
	`36`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`37`	`37`	`response = model.predict(`
`38`	`38`	`"""Background: There is evidence that there have been significant changes \`
`39`	`39`	`in Amazon rainforest vegetation over the last 21,000 years through the Last \`
Original file line number	Diff line number	Diff line change
`@@ -33,7 +33,7 @@ def interview(`
`33`	`33`	`"top_k": 40, # A top_k of 1 means the selected token is the most probable among all tokens.`
`34`	`34`	`}`
`35`	`35`
`36`		`- model = TextGenerationModel.from_pretrained("text-bison@001")`
	`36`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`37`	`37`	`response = model.predict(`
`38`	`38`	`"Give me ten interview questions for the role of program manager.",`
`39`	`39`	`**parameters,`
Original file line number	Diff line number	Diff line change
`@@ -33,7 +33,7 @@ def sentiment_analysis(`
`33`	`33`	`"top_k": 1, # A top_k of 1 means the selected token is the most probable among all tokens.`
`34`	`34`	`}`
`35`	`35`
`36`		`- model = TextGenerationModel.from_pretrained("google/text-bison@001")`
	`36`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`37`	`37`	`response = model.predict(`
`38`	`38`	`"""I had to compare two versions of Hamlet for my Shakespeare class and \`
`39`	`39`	`unfortunately I picked this version. Everything from the acting (the actors \`
Original file line number	Diff line number	Diff line change
`@@ -33,7 +33,7 @@ def text_summarization(`
`33`	`33`	`"top_k": 40, # A top_k of 1 means the selected token is the most probable among all tokens.`
`34`	`34`	`}`
`35`	`35`
`36`		`- model = TextGenerationModel.from_pretrained("text-bison@001")`
	`36`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`37`	`37`	`response = model.predict(`
`38`	`38`	`"""Provide a summary with about two sentences for the following article:`
`39`	`39`	`The efficient-market hypothesis (EMH) is a hypothesis in financial \`
Original file line number	Diff line number	Diff line change
`@@ -69,7 +69,7 @@ def tuning(`
`69`	`69`	`eval_spec.tensorboard = aiplatform.Tensorboard(`
`70`	`70`	`tensorboard_name=tensorboard_instance_name`
`71`	`71`	`)`
`72`		`- model = TextGenerationModel.from_pretrained("text-bison@001")`
	`72`	`+ model = TextGenerationModel.from_pretrained("text-bison@002")`
`73`	`73`
`74`	`74`	`model.tune_model(`
`75`	`75`	`training_data=training_data,`