Merge pull request #177 from GoogleCloudPlatform/@invertase/add-filte…

…r-params-GL

feat(firestore-palm-gen-text): add config params for filters

firestore-palm-gen-text/README.md

@@ -116,6 +116,20 @@ Additionally, this extension uses the PaLM API, which is currently in public pre
 
 * Maximum number of tokens: If you have selected the Vertex AI service as your PaLM API provider, this parameter will be used to set the max_tokens parameter in the Vertex API request. It should be an integer in the range [1,1024]. The default value for the extension is 100.
 
+* Derogatory Content Threshold: Threshold for derogatory content. Specify what level of derogatory content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Toxicity Threshold: Threshold for toxic content. Specify what level of toxic content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Sexual Content Threshold: Threshold for sexual content. Specify what level of sexual content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Violent Content Threshold: Threshold for violent content. Specify what level of violent content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Medical Content Threshold: Threshold for medical content. Specify what level of medical content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Dangerous Content Threshold: Threshold for dangerous content. Specify what level of dangerous content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
+* Unspecified Harm Threshold: Threshold for non-specific harmful content. Specify what level of non-specific harmful content is blocked by the PaLM provider. This threshold is applicable only to the Generative Language PaLM API.
+
 
 
 **Cloud Functions:**

firestore-palm-gen-text/extension.yaml

@@ -279,3 +279,136 @@ params:
     required: false
     default: 100
     immutable: false
+
+  - param: DEROGATORY_THRESHOLD
+    label: Derogatory Content Threshold
+    description: >-
+      Threshold for derogatory content. Specify what level of derogatory content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: TOXICITY_THRESHOLD
+    label: Toxicity Threshold
+    description: >-
+      Threshold for toxic content. Specify what level of toxic content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: SEXUAL_THRESHOLD
+    label: Sexual Content Threshold
+    description: >-
+      Threshold for sexual content. Specify what level of sexual content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: VIOLENCE_THRESHOLD
+    label: Violent Content Threshold
+    description: >-
+      Threshold for violent content. Specify what level of violent content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: MEDICAL_THRESHOLD
+    label: Medical Content Threshold
+    description: >-
+      Threshold for medical content. Specify what level of medical content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: DANGEROUS_THRESHOLD
+    label: Dangerous Content Threshold 
+    description: >-
+      Threshold for dangerous content. Specify what level of dangerous content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false
+
+  - param: UNSPECIFIED_THRESHOLD
+    label: Unspecified Harm Threshold
+    description: >-
+      Threshold for non-specific harmful content. Specify what level of non-specific harmful content is blocked by the PaLM provider.
+      This threshold is applicable only to the Generative Language PaLM API.
+    type: select
+    options:
+      - label: Harm block threshold unspecified
+        value: HARM_BLOCK_THRESHOLD_UNSPECIFIED
+      - label: Block low and above
+        value: BLOCK_LOW_AND_ABOVE
+      - label: Block medium and above
+        value: BLOCK_MEDIUM_AND_ABOVE
+      - label: Block only high
+        value: BLOCK_ONLY_HIGH
+    required: false
+    default: BLOCK_ONLY_HIGH
+    immutable: false