You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardexpand all lines: docs/dyn/dialogflow_v2.projects.answerRecords.html
+4-4
Original file line number
Diff line number
Diff line change
@@ -98,7 +98,7 @@ <h3>Method Details</h3>
98
98
99
99
Args:
100
100
parent: string, Required. The project to list all answer records for in reverse chronological order. Format: `projects//locations/`. (required)
101
-
filter: string, Optional. Filters to restrict results to specific answer records. Marked deprecated as it hasn't been, and isn't currently, supported. For more information about filtering, see [API Filtering](https://aip.dev/160).
101
+
filter: string, Optional. Filters to restrict results to specific answer records. The expression has the following syntax: [AND ] ... The following fields and operators are supported: * conversation_id with equals(=) operator Examples: * "conversation_id=bar" matches answer records in the projects/foo/locations/global/conversations/bar conversation (assuming the parent is projects/foo/locations/global). For more information about filtering, see [API Filtering](https://aip.dev/160).
102
102
pageSize: integer, Optional. The maximum number of records to return in a single page. The server may return fewer records than this. If unspecified, we use 10. The maximum is 100.
103
103
pageToken: string, Optional. The ListAnswerRecordsResponse.next_page_token value returned from a previous list request used to continue listing on the next page.
104
104
x__xgafv: string, V1 error format.
@@ -603,7 +603,7 @@ <h3>Method Details</h3>
603
603
"answerFeedback": { # Represents feedback the customer has about the quality & correctness of a certain answer in a conversation. # Required. The AnswerFeedback for this record. You can set this with AnswerRecords.UpdateAnswerRecord in order to give us feedback about this answer.
604
604
"agentAssistantDetailFeedback": { # Detail feedback of Agent Assist result. # Detail feedback of agent assist suggestions.
605
605
"answerRelevance": "A String", # Optional. Whether or not the suggested answer is relevant. For example: * Query: "Can I change my mailing address?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * answer_relevance: AnswerRelevance.IRRELEVANT
606
-
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * [document_correctness]: INCORRECT
606
+
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * document_correctness: INCORRECT
607
607
"documentEfficiency": "A String", # Optional. Whether or not the suggested document is efficient. For example, if the document is poorly written, hard to understand, hard to use or too long to find useful information, document_efficiency is DocumentEfficiency.INEFFICIENT.
608
608
"knowledgeAssistFeedback": { # Feedback for knowledge assist. # Optional. Feedback for knowledge assist.
609
609
"answerCopied": True or False, # Whether the suggested answer was copied by the human agent. If the value is set to be true, AnswerFeedback.clicked will be updated to be true.
@@ -1154,7 +1154,7 @@ <h3>Method Details</h3>
1154
1154
"answerFeedback": { # Represents feedback the customer has about the quality & correctness of a certain answer in a conversation. # Required. The AnswerFeedback for this record. You can set this with AnswerRecords.UpdateAnswerRecord in order to give us feedback about this answer.
1155
1155
"agentAssistantDetailFeedback": { # Detail feedback of Agent Assist result. # Detail feedback of agent assist suggestions.
1156
1156
"answerRelevance": "A String", # Optional. Whether or not the suggested answer is relevant. For example: * Query: "Can I change my mailing address?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * answer_relevance: AnswerRelevance.IRRELEVANT
1157
-
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * [document_correctness]: INCORRECT
1157
+
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * document_correctness: INCORRECT
1158
1158
"documentEfficiency": "A String", # Optional. Whether or not the suggested document is efficient. For example, if the document is poorly written, hard to understand, hard to use or too long to find useful information, document_efficiency is DocumentEfficiency.INEFFICIENT.
1159
1159
"knowledgeAssistFeedback": { # Feedback for knowledge assist. # Optional. Feedback for knowledge assist.
1160
1160
"answerCopied": True or False, # Whether the suggested answer was copied by the human agent. If the value is set to be true, AnswerFeedback.clicked will be updated to be true.
@@ -1687,7 +1687,7 @@ <h3>Method Details</h3>
1687
1687
"answerFeedback": { # Represents feedback the customer has about the quality & correctness of a certain answer in a conversation. # Required. The AnswerFeedback for this record. You can set this with AnswerRecords.UpdateAnswerRecord in order to give us feedback about this answer.
1688
1688
"agentAssistantDetailFeedback": { # Detail feedback of Agent Assist result. # Detail feedback of agent assist suggestions.
1689
1689
"answerRelevance": "A String", # Optional. Whether or not the suggested answer is relevant. For example: * Query: "Can I change my mailing address?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * answer_relevance: AnswerRelevance.IRRELEVANT
1690
-
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * [document_correctness]: INCORRECT
1690
+
"documentCorrectness": "A String", # Optional. Whether or not the information in the document is correct. For example: * Query: "Can I return the package in 2 days once received?" * Suggested document says: "Items must be returned/exchanged within 60 days of the purchase date." * Ground truth: "No return or exchange is allowed." * document_correctness: INCORRECT
1691
1691
"documentEfficiency": "A String", # Optional. Whether or not the suggested document is efficient. For example, if the document is poorly written, hard to understand, hard to use or too long to find useful information, document_efficiency is DocumentEfficiency.INEFFICIENT.
1692
1692
"knowledgeAssistFeedback": { # Feedback for knowledge assist. # Optional. Feedback for knowledge assist.
1693
1693
"answerCopied": True or False, # Whether the suggested answer was copied by the human agent. If the value is set to be true, AnswerFeedback.clicked will be updated to be true.
"smartComposeConfig": { # Smart compose specific configuration for evaluation job. # Configuration for smart compose model evalution.
118
+
"smartComposeConfig": { # Smart compose specific configuration for evaluation job. # Configuration for smart compose model evaluation.
119
119
"allowlistDocument": "A String", # The allowlist document resource name. Format: `projects//knowledgeBases//documents/`. Only used for smart compose model.
120
120
"maxResultCount": 42, # Required. The model to be evaluated can return multiple results with confidence score on each query. These results will be sorted by the descending order of the scores and we only keep the first max_result_count results as the final results to evaluate.
121
121
},
122
-
"smartReplyConfig": { # Smart reply specific configuration for evaluation job. # Configuration for smart reply model evalution.
122
+
"smartReplyConfig": { # Smart reply specific configuration for evaluation job. # Configuration for smart reply model evaluation.
123
123
"allowlistDocument": "A String", # The allowlist document resource name. Format: `projects//knowledgeBases//documents/`. Only used for smart reply model.
124
124
"maxResultCount": 42, # Required. The model to be evaluated can return multiple results with confidence score on each query. These results will be sorted by the descending order of the scores and we only keep the first max_result_count results as the final results to evaluate.
125
125
},
126
126
},
127
127
"name": "A String", # The resource name of the evaluation. Format: `projects//conversationModels//evaluations/`
128
-
"rawHumanEvalTemplateCsv": "A String", # Output only. Human eval template in csv format. It tooks real-world conversations provided through input dataset, generates example suggestions for customer to verify quality of the model. For Smart Reply, the generated csv file contains columns of Context, (Suggestions,Q1,Q2)*3, Actual reply. Context contains at most 10 latest messages in the conversation prior to the current suggestion. Q1: "Would you send it as the next message of agent?" Evaluated based on whether the suggest is appropriate to be sent by agent in current context. Q2: "Does the suggestion move the conversation closer to resolution?" Evaluated based on whether the suggestion provide solutions, or answers customer's question or collect information from customer to resolve the customer's issue. Actual reply column contains the actual agent reply sent in the context.
128
+
"rawHumanEvalTemplateCsv": "A String", # Output only. Human eval template in csv format. It takes real-world conversations provided through input dataset, generates example suggestions for customer to verify quality of the model. For Smart Reply, the generated csv file contains columns of Context, (Suggestions,Q1,Q2)*3, Actual reply. Context contains at most 10 latest messages in the conversation prior to the current suggestion. Q1: "Would you send it as the next message of agent?" Evaluated based on whether the suggest is appropriate to be sent by agent in current context. Q2: "Does the suggestion move the conversation closer to resolution?" Evaluated based on whether the suggestion provide solutions, or answers customer's question or collect information from customer to resolve the customer's issue. Actual reply column contains the actual agent reply sent in the context.
129
129
"smartReplyMetrics": { # The evaluation metrics for smart reply model. # Output only. Only available when model is for smart reply.
130
130
"allowlistCoverage": 3.14, # Percentage of target participant messages in the evaluation dataset for which similar messages have appeared at least once in the allowlist. Should be [0, 1].
131
131
"conversationCount": "A String", # Total number of conversations used to generate this metric.
"smartComposeConfig": { # Smart compose specific configuration for evaluation job. # Configuration for smart compose model evalution.
169
+
"smartComposeConfig": { # Smart compose specific configuration for evaluation job. # Configuration for smart compose model evaluation.
170
170
"allowlistDocument": "A String", # The allowlist document resource name. Format: `projects//knowledgeBases//documents/`. Only used for smart compose model.
171
171
"maxResultCount": 42, # Required. The model to be evaluated can return multiple results with confidence score on each query. These results will be sorted by the descending order of the scores and we only keep the first max_result_count results as the final results to evaluate.
172
172
},
173
-
"smartReplyConfig": { # Smart reply specific configuration for evaluation job. # Configuration for smart reply model evalution.
173
+
"smartReplyConfig": { # Smart reply specific configuration for evaluation job. # Configuration for smart reply model evaluation.
174
174
"allowlistDocument": "A String", # The allowlist document resource name. Format: `projects//knowledgeBases//documents/`. Only used for smart reply model.
175
175
"maxResultCount": 42, # Required. The model to be evaluated can return multiple results with confidence score on each query. These results will be sorted by the descending order of the scores and we only keep the first max_result_count results as the final results to evaluate.
176
176
},
177
177
},
178
178
"name": "A String", # The resource name of the evaluation. Format: `projects//conversationModels//evaluations/`
179
-
"rawHumanEvalTemplateCsv": "A String", # Output only. Human eval template in csv format. It tooks real-world conversations provided through input dataset, generates example suggestions for customer to verify quality of the model. For Smart Reply, the generated csv file contains columns of Context, (Suggestions,Q1,Q2)*3, Actual reply. Context contains at most 10 latest messages in the conversation prior to the current suggestion. Q1: "Would you send it as the next message of agent?" Evaluated based on whether the suggest is appropriate to be sent by agent in current context. Q2: "Does the suggestion move the conversation closer to resolution?" Evaluated based on whether the suggestion provide solutions, or answers customer's question or collect information from customer to resolve the customer's issue. Actual reply column contains the actual agent reply sent in the context.
179
+
"rawHumanEvalTemplateCsv": "A String", # Output only. Human eval template in csv format. It takes real-world conversations provided through input dataset, generates example suggestions for customer to verify quality of the model. For Smart Reply, the generated csv file contains columns of Context, (Suggestions,Q1,Q2)*3, Actual reply. Context contains at most 10 latest messages in the conversation prior to the current suggestion. Q1: "Would you send it as the next message of agent?" Evaluated based on whether the suggest is appropriate to be sent by agent in current context. Q2: "Does the suggestion move the conversation closer to resolution?" Evaluated based on whether the suggestion provide solutions, or answers customer's question or collect information from customer to resolve the customer's issue. Actual reply column contains the actual agent reply sent in the context.
180
180
"smartReplyMetrics": { # The evaluation metrics for smart reply model. # Output only. Only available when model is for smart reply.
181
181
"allowlistCoverage": 3.14, # Percentage of target participant messages in the evaluation dataset for which similar messages have appeared at least once in the allowlist. Should be [0, 1].
182
182
"conversationCount": "A String", # Total number of conversations used to generate this metric.
Copy file name to clipboardexpand all lines: docs/dyn/dialogflow_v2.projects.conversations.participants.suggestions.html
+1-1
Original file line number
Diff line number
Diff line change
@@ -199,7 +199,7 @@ <h3>Method Details</h3>
199
199
{ # The request message for Participants.SuggestKnowledgeAssist.
200
200
"contextSize": 42, # Optional. Max number of messages prior to and including latest_message to use as context when compiling the suggestion. The context size is by default 100 and at most 100.
201
201
"latestMessage": "A String", # Optional. The name of the latest conversation message to compile suggestions for. If empty, it will be the latest message of the conversation. Format: `projects//locations//conversations//messages/`.
202
-
"previousSuggestedQuery": "A String", # Optional. The previously suggested query for the given conversation. This helps identify whether the next suggestion we generate is resonably different from the previous one. This is useful to avoid similar suggestions within the conversation.
202
+
"previousSuggestedQuery": "A String", # Optional. The previously suggested query for the given conversation. This helps identify whether the next suggestion we generate is reasonably different from the previous one. This is useful to avoid similar suggestions within the conversation.
0 commit comments