google-cloud-ces-v1beta 0.a → 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/.yardopts +12 -0
- data/AUTHENTICATION.md +122 -0
- data/README.md +154 -8
- data/lib/google/cloud/ces/v1beta/agent_pb.rb +30 -0
- data/lib/google/cloud/ces/v1beta/agent_service/client.rb +5635 -0
- data/lib/google/cloud/ces/v1beta/agent_service/credentials.rb +48 -0
- data/lib/google/cloud/ces/v1beta/agent_service/operations.rb +841 -0
- data/lib/google/cloud/ces/v1beta/agent_service/paths.rb +481 -0
- data/lib/google/cloud/ces/v1beta/agent_service/rest/client.rb +5243 -0
- data/lib/google/cloud/ces/v1beta/agent_service/rest/operations.rb +925 -0
- data/lib/google/cloud/ces/v1beta/agent_service/rest/service_stub.rb +3213 -0
- data/lib/google/cloud/ces/v1beta/agent_service/rest.rb +55 -0
- data/lib/google/cloud/ces/v1beta/agent_service.rb +57 -0
- data/lib/google/cloud/ces/v1beta/agent_service_pb.rb +131 -0
- data/lib/google/cloud/ces/v1beta/agent_service_services_pb.rb +148 -0
- data/lib/google/cloud/ces/v1beta/agent_tool_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/agent_transfers_pb.rb +29 -0
- data/lib/google/cloud/ces/v1beta/app_pb.rb +67 -0
- data/lib/google/cloud/ces/v1beta/app_version_pb.rb +32 -0
- data/lib/google/cloud/ces/v1beta/auth_pb.rb +34 -0
- data/lib/google/cloud/ces/v1beta/bigquery_export_pb.rb +23 -0
- data/lib/google/cloud/ces/v1beta/bindings_override.rb +75 -0
- data/lib/google/cloud/ces/v1beta/changelog_pb.rb +26 -0
- data/lib/google/cloud/ces/v1beta/client_function_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/common_pb.rb +46 -0
- data/lib/google/cloud/ces/v1beta/connector_tool_pb.rb +27 -0
- data/lib/google/cloud/ces/v1beta/connector_toolset_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/conversation_pb.rb +31 -0
- data/lib/google/cloud/ces/v1beta/data_store_pb.rb +27 -0
- data/lib/google/cloud/ces/v1beta/data_store_tool_pb.rb +40 -0
- data/lib/google/cloud/ces/v1beta/deployment_pb.rb +26 -0
- data/lib/google/cloud/ces/v1beta/evaluation_pb.rb +98 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/client.rb +3732 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/credentials.rb +48 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/operations.rb +841 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/paths.rb +306 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/rest/client.rb +3473 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/rest/operations.rb +925 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/rest/service_stub.rb +2047 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service/rest.rb +54 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service.rb +56 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service_pb.rb +86 -0
- data/lib/google/cloud/ces/v1beta/evaluation_service_services_pb.rb +110 -0
- data/lib/google/cloud/ces/v1beta/example_pb.rb +34 -0
- data/lib/google/cloud/ces/v1beta/fakes_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/file_context_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/file_search_tool_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/golden_run_pb.rb +21 -0
- data/lib/google/cloud/ces/v1beta/google_search_tool_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/guardrail_pb.rb +37 -0
- data/lib/google/cloud/ces/v1beta/mcp_tool_pb.rb +26 -0
- data/lib/google/cloud/ces/v1beta/mcp_toolset_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/mocks_pb.rb +26 -0
- data/lib/google/cloud/ces/v1beta/omnichannel_pb.rb +31 -0
- data/lib/google/cloud/ces/v1beta/omnichannel_service_pb.rb +27 -0
- data/lib/google/cloud/ces/v1beta/open_api_tool_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/open_api_toolset_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/python_function_pb.rb +23 -0
- data/lib/google/cloud/ces/v1beta/rest.rb +42 -0
- data/lib/google/cloud/ces/v1beta/schema_pb.rb +25 -0
- data/lib/google/cloud/ces/v1beta/search_suggestions_pb.rb +22 -0
- data/lib/google/cloud/ces/v1beta/security_settings_pb.rb +27 -0
- data/lib/google/cloud/ces/v1beta/session_service/client.rb +760 -0
- data/lib/google/cloud/ces/v1beta/session_service/credentials.rb +48 -0
- data/lib/google/cloud/ces/v1beta/session_service/paths.rb +117 -0
- data/lib/google/cloud/ces/v1beta/session_service/rest/client.rb +560 -0
- data/lib/google/cloud/ces/v1beta/session_service/rest/service_stub.rb +199 -0
- data/lib/google/cloud/ces/v1beta/session_service/rest.rb +53 -0
- data/lib/google/cloud/ces/v1beta/session_service.rb +55 -0
- data/lib/google/cloud/ces/v1beta/session_service_pb.rb +53 -0
- data/lib/google/cloud/ces/v1beta/session_service_services_pb.rb +120 -0
- data/lib/google/cloud/ces/v1beta/system_tool_pb.rb +23 -0
- data/lib/google/cloud/ces/v1beta/tool_pb.rb +38 -0
- data/lib/google/cloud/ces/v1beta/tool_service/client.rb +727 -0
- data/lib/google/cloud/ces/v1beta/tool_service/credentials.rb +48 -0
- data/lib/google/cloud/ces/v1beta/tool_service/paths.rb +94 -0
- data/lib/google/cloud/ces/v1beta/tool_service/rest/client.rb +671 -0
- data/lib/google/cloud/ces/v1beta/tool_service/rest/service_stub.rb +267 -0
- data/lib/google/cloud/ces/v1beta/tool_service/rest.rb +53 -0
- data/lib/google/cloud/ces/v1beta/tool_service.rb +55 -0
- data/lib/google/cloud/ces/v1beta/tool_service_pb.rb +36 -0
- data/lib/google/cloud/ces/v1beta/tool_service_services_pb.rb +50 -0
- data/lib/google/cloud/ces/v1beta/toolset_pb.rb +30 -0
- data/lib/google/cloud/ces/v1beta/toolset_tool_pb.rb +24 -0
- data/lib/google/cloud/ces/v1beta/version.rb +7 -2
- data/lib/google/cloud/ces/v1beta/widget_service/client.rb +501 -0
- data/lib/google/cloud/ces/v1beta/widget_service/credentials.rb +48 -0
- data/lib/google/cloud/ces/v1beta/widget_service/paths.rb +75 -0
- data/lib/google/cloud/ces/v1beta/widget_service/rest/client.rb +459 -0
- data/lib/google/cloud/ces/v1beta/widget_service/rest/service_stub.rb +143 -0
- data/lib/google/cloud/ces/v1beta/widget_service/rest.rb +53 -0
- data/lib/google/cloud/ces/v1beta/widget_service.rb +55 -0
- data/lib/google/cloud/ces/v1beta/widget_service_pb.rb +28 -0
- data/lib/google/cloud/ces/v1beta/widget_service_services_pb.rb +46 -0
- data/lib/google/cloud/ces/v1beta/widget_tool_pb.rb +29 -0
- data/lib/google/cloud/ces/v1beta.rb +49 -0
- data/lib/google-cloud-ces-v1beta.rb +21 -0
- data/proto_docs/README.md +4 -0
- data/proto_docs/google/api/client.rb +593 -0
- data/proto_docs/google/api/field_behavior.rb +85 -0
- data/proto_docs/google/api/launch_stage.rb +71 -0
- data/proto_docs/google/api/resource.rb +227 -0
- data/proto_docs/google/cloud/ces/v1beta/agent.rb +224 -0
- data/proto_docs/google/cloud/ces/v1beta/agent_service.rb +1661 -0
- data/proto_docs/google/cloud/ces/v1beta/agent_tool.rb +48 -0
- data/proto_docs/google/cloud/ces/v1beta/agent_transfers.rb +111 -0
- data/proto_docs/google/cloud/ces/v1beta/app.rb +836 -0
- data/proto_docs/google/cloud/ces/v1beta/app_version.rb +83 -0
- data/proto_docs/google/cloud/ces/v1beta/auth.rb +214 -0
- data/proto_docs/google/cloud/ces/v1beta/bigquery_export.rb +45 -0
- data/proto_docs/google/cloud/ces/v1beta/changelog.rb +73 -0
- data/proto_docs/google/cloud/ces/v1beta/client_function.rb +48 -0
- data/proto_docs/google/cloud/ces/v1beta/common.rb +375 -0
- data/proto_docs/google/cloud/ces/v1beta/connector_tool.rb +122 -0
- data/proto_docs/google/cloud/ces/v1beta/connector_toolset.rb +52 -0
- data/proto_docs/google/cloud/ces/v1beta/conversation.rb +156 -0
- data/proto_docs/google/cloud/ces/v1beta/data_store.rb +102 -0
- data/proto_docs/google/cloud/ces/v1beta/data_store_tool.rb +328 -0
- data/proto_docs/google/cloud/ces/v1beta/deployment.rb +62 -0
- data/proto_docs/google/cloud/ces/v1beta/evaluation.rb +1618 -0
- data/proto_docs/google/cloud/ces/v1beta/evaluation_service.rb +1009 -0
- data/proto_docs/google/cloud/ces/v1beta/example.rb +245 -0
- data/proto_docs/google/cloud/ces/v1beta/fakes.rb +67 -0
- data/proto_docs/google/cloud/ces/v1beta/file_context.rb +50 -0
- data/proto_docs/google/cloud/ces/v1beta/file_search_tool.rb +60 -0
- data/proto_docs/google/cloud/ces/v1beta/golden_run.rb +40 -0
- data/proto_docs/google/cloud/ces/v1beta/google_search_tool.rb +76 -0
- data/proto_docs/google/cloud/ces/v1beta/guardrail.rb +324 -0
- data/proto_docs/google/cloud/ces/v1beta/mcp_tool.rb +85 -0
- data/proto_docs/google/cloud/ces/v1beta/mcp_toolset.rb +71 -0
- data/proto_docs/google/cloud/ces/v1beta/mocks.rb +60 -0
- data/proto_docs/google/cloud/ces/v1beta/omnichannel.rb +163 -0
- data/proto_docs/google/cloud/ces/v1beta/omnichannel_service.rb +45 -0
- data/proto_docs/google/cloud/ces/v1beta/open_api_tool.rb +66 -0
- data/proto_docs/google/cloud/ces/v1beta/open_api_toolset.rb +57 -0
- data/proto_docs/google/cloud/ces/v1beta/python_function.rb +44 -0
- data/proto_docs/google/cloud/ces/v1beta/schema.rb +162 -0
- data/proto_docs/google/cloud/ces/v1beta/search_suggestions.rb +55 -0
- data/proto_docs/google/cloud/ces/v1beta/security_settings.rb +80 -0
- data/proto_docs/google/cloud/ces/v1beta/session_service.rb +490 -0
- data/proto_docs/google/cloud/ces/v1beta/system_tool.rb +38 -0
- data/proto_docs/google/cloud/ces/v1beta/tool.rb +126 -0
- data/proto_docs/google/cloud/ces/v1beta/tool_service.rb +173 -0
- data/proto_docs/google/cloud/ces/v1beta/toolset.rb +80 -0
- data/proto_docs/google/cloud/ces/v1beta/toolset_tool.rb +40 -0
- data/proto_docs/google/cloud/ces/v1beta/widget_service.rb +62 -0
- data/proto_docs/google/cloud/ces/v1beta/widget_tool.rb +146 -0
- data/proto_docs/google/longrunning/operations.rb +191 -0
- data/proto_docs/google/protobuf/any.rb +145 -0
- data/proto_docs/google/protobuf/duration.rb +98 -0
- data/proto_docs/google/protobuf/empty.rb +34 -0
- data/proto_docs/google/protobuf/field_mask.rb +229 -0
- data/proto_docs/google/protobuf/struct.rb +108 -0
- data/proto_docs/google/protobuf/timestamp.rb +127 -0
- data/proto_docs/google/rpc/status.rb +48 -0
- metadata +204 -9
|
@@ -0,0 +1,1618 @@
|
|
|
1
|
+
# frozen_string_literal: true
|
|
2
|
+
|
|
3
|
+
# Copyright 2026 Google LLC
|
|
4
|
+
#
|
|
5
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
6
|
+
# you may not use this file except in compliance with the License.
|
|
7
|
+
# You may obtain a copy of the License at
|
|
8
|
+
#
|
|
9
|
+
# https://www.apache.org/licenses/LICENSE-2.0
|
|
10
|
+
#
|
|
11
|
+
# Unless required by applicable law or agreed to in writing, software
|
|
12
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
13
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
14
|
+
# See the License for the specific language governing permissions and
|
|
15
|
+
# limitations under the License.
|
|
16
|
+
|
|
17
|
+
# Auto-generated by gapic-generator-ruby. DO NOT EDIT!
|
|
18
|
+
|
|
19
|
+
|
|
20
|
+
module Google
|
|
21
|
+
module Cloud
|
|
22
|
+
module Ces
|
|
23
|
+
module V1beta
|
|
24
|
+
# Aggregated metrics for an evaluation or evaluation dataset.
|
|
25
|
+
# @!attribute [r] metrics_by_app_version
|
|
26
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::MetricsByAppVersion>]
|
|
27
|
+
# Output only. Aggregated metrics, grouped by app version ID.
|
|
28
|
+
class AggregatedMetrics
|
|
29
|
+
include ::Google::Protobuf::MessageExts
|
|
30
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
31
|
+
|
|
32
|
+
# Metrics for a single tool.
|
|
33
|
+
# @!attribute [r] tool
|
|
34
|
+
# @return [::String]
|
|
35
|
+
# Output only. The name of the tool.
|
|
36
|
+
# @!attribute [r] pass_count
|
|
37
|
+
# @return [::Integer]
|
|
38
|
+
# Output only. The number of times the tool passed.
|
|
39
|
+
# @!attribute [r] fail_count
|
|
40
|
+
# @return [::Integer]
|
|
41
|
+
# Output only. The number of times the tool failed.
|
|
42
|
+
class ToolMetrics
|
|
43
|
+
include ::Google::Protobuf::MessageExts
|
|
44
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
45
|
+
end
|
|
46
|
+
|
|
47
|
+
# Metrics for turn latency.
|
|
48
|
+
# @!attribute [r] average_latency
|
|
49
|
+
# @return [::Google::Protobuf::Duration]
|
|
50
|
+
# Output only. The average latency of the turns.
|
|
51
|
+
class TurnLatencyMetrics
|
|
52
|
+
include ::Google::Protobuf::MessageExts
|
|
53
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
54
|
+
end
|
|
55
|
+
|
|
56
|
+
# Metrics for tool call latency.
|
|
57
|
+
# @!attribute [r] tool
|
|
58
|
+
# @return [::String]
|
|
59
|
+
# Output only. The name of the tool.
|
|
60
|
+
# @!attribute [r] average_latency
|
|
61
|
+
# @return [::Google::Protobuf::Duration]
|
|
62
|
+
# Output only. The average latency of the tool calls.
|
|
63
|
+
class ToolCallLatencyMetrics
|
|
64
|
+
include ::Google::Protobuf::MessageExts
|
|
65
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
66
|
+
end
|
|
67
|
+
|
|
68
|
+
# Metrics for semantic similarity results.
|
|
69
|
+
# @!attribute [r] score
|
|
70
|
+
# @return [::Float]
|
|
71
|
+
# Output only. The average semantic similarity score (0-4).
|
|
72
|
+
class SemanticSimilarityMetrics
|
|
73
|
+
include ::Google::Protobuf::MessageExts
|
|
74
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
75
|
+
end
|
|
76
|
+
|
|
77
|
+
# Metrics for hallucination results.
|
|
78
|
+
# @!attribute [r] score
|
|
79
|
+
# @return [::Float]
|
|
80
|
+
# Output only. The average hallucination score (0 to 1).
|
|
81
|
+
class HallucinationMetrics
|
|
82
|
+
include ::Google::Protobuf::MessageExts
|
|
83
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
84
|
+
end
|
|
85
|
+
|
|
86
|
+
# Metrics aggregated per app version.
|
|
87
|
+
# @!attribute [r] app_version_id
|
|
88
|
+
# @return [::String]
|
|
89
|
+
# Output only. The app version ID.
|
|
90
|
+
# @!attribute [r] tool_metrics
|
|
91
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::ToolMetrics>]
|
|
92
|
+
# Output only. Metrics for each tool within this app version.
|
|
93
|
+
# @!attribute [r] semantic_similarity_metrics
|
|
94
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::SemanticSimilarityMetrics>]
|
|
95
|
+
# Output only. Metrics for semantic similarity within this app version.
|
|
96
|
+
# @!attribute [r] hallucination_metrics
|
|
97
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::HallucinationMetrics>]
|
|
98
|
+
# Output only. Metrics for hallucination within this app version.
|
|
99
|
+
# @!attribute [r] tool_call_latency_metrics
|
|
100
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::ToolCallLatencyMetrics>]
|
|
101
|
+
# Output only. Metrics for tool call latency within this app version.
|
|
102
|
+
# @!attribute [r] turn_latency_metrics
|
|
103
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::TurnLatencyMetrics>]
|
|
104
|
+
# Output only. Metrics for turn latency within this app version.
|
|
105
|
+
# @!attribute [r] pass_count
|
|
106
|
+
# @return [::Integer]
|
|
107
|
+
# Output only. The number of times the evaluation passed.
|
|
108
|
+
# @!attribute [r] fail_count
|
|
109
|
+
# @return [::Integer]
|
|
110
|
+
# Output only. The number of times the evaluation failed.
|
|
111
|
+
# @!attribute [r] metrics_by_turn
|
|
112
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::MetricsByTurn>]
|
|
113
|
+
# Output only. Metrics aggregated per turn within this app version.
|
|
114
|
+
class MetricsByAppVersion
|
|
115
|
+
include ::Google::Protobuf::MessageExts
|
|
116
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
117
|
+
end
|
|
118
|
+
|
|
119
|
+
# Metrics aggregated per turn.
|
|
120
|
+
# @!attribute [r] turn_index
|
|
121
|
+
# @return [::Integer]
|
|
122
|
+
# Output only. The turn index (0-based).
|
|
123
|
+
# @!attribute [r] tool_metrics
|
|
124
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::ToolMetrics>]
|
|
125
|
+
# Output only. Metrics for each tool within this turn.
|
|
126
|
+
# @!attribute [r] semantic_similarity_metrics
|
|
127
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::SemanticSimilarityMetrics>]
|
|
128
|
+
# Output only. Metrics for semantic similarity within this turn.
|
|
129
|
+
# @!attribute [r] hallucination_metrics
|
|
130
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::HallucinationMetrics>]
|
|
131
|
+
# Output only. Metrics for hallucination within this turn.
|
|
132
|
+
# @!attribute [r] tool_call_latency_metrics
|
|
133
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::ToolCallLatencyMetrics>]
|
|
134
|
+
# Output only. Metrics for tool call latency within this turn.
|
|
135
|
+
# @!attribute [r] turn_latency_metrics
|
|
136
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::AggregatedMetrics::TurnLatencyMetrics>]
|
|
137
|
+
# Output only. Metrics for turn latency within this turn.
|
|
138
|
+
class MetricsByTurn
|
|
139
|
+
include ::Google::Protobuf::MessageExts
|
|
140
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
141
|
+
end
|
|
142
|
+
end
|
|
143
|
+
|
|
144
|
+
# An evaluation represents all of the information needed to simulate and
|
|
145
|
+
# evaluate an agent.
|
|
146
|
+
# @!attribute [rw] golden
|
|
147
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::Golden]
|
|
148
|
+
# Optional. The golden steps to be evaluated.
|
|
149
|
+
#
|
|
150
|
+
# Note: The following fields are mutually exclusive: `golden`, `scenario`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
151
|
+
# @!attribute [rw] scenario
|
|
152
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::Scenario]
|
|
153
|
+
# Optional. The config for a scenario.
|
|
154
|
+
#
|
|
155
|
+
# Note: The following fields are mutually exclusive: `scenario`, `golden`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
156
|
+
# @!attribute [rw] name
|
|
157
|
+
# @return [::String]
|
|
158
|
+
# Identifier. The unique identifier of this evaluation.
|
|
159
|
+
# Format:
|
|
160
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluations/{evaluation}`
|
|
161
|
+
# @!attribute [rw] display_name
|
|
162
|
+
# @return [::String]
|
|
163
|
+
# Required. User-defined display name of the evaluation. Unique within an
|
|
164
|
+
# App.
|
|
165
|
+
# @!attribute [rw] description
|
|
166
|
+
# @return [::String]
|
|
167
|
+
# Optional. User-defined description of the evaluation.
|
|
168
|
+
# @!attribute [rw] tags
|
|
169
|
+
# @return [::Array<::String>]
|
|
170
|
+
# Optional. User defined tags to categorize the evaluation.
|
|
171
|
+
# @!attribute [r] evaluation_datasets
|
|
172
|
+
# @return [::Array<::String>]
|
|
173
|
+
# Output only. List of evaluation datasets the evaluation belongs to.
|
|
174
|
+
# Format:
|
|
175
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationDatasets/{evaluationDataset}`
|
|
176
|
+
# @!attribute [r] create_time
|
|
177
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
178
|
+
# Output only. Timestamp when the evaluation was created.
|
|
179
|
+
# @!attribute [r] created_by
|
|
180
|
+
# @return [::String]
|
|
181
|
+
# Output only. The user who created the evaluation.
|
|
182
|
+
# @!attribute [r] update_time
|
|
183
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
184
|
+
# Output only. Timestamp when the evaluation was last updated.
|
|
185
|
+
# @!attribute [r] last_updated_by
|
|
186
|
+
# @return [::String]
|
|
187
|
+
# Output only. The user who last updated the evaluation.
|
|
188
|
+
# @!attribute [r] evaluation_runs
|
|
189
|
+
# @return [::Array<::String>]
|
|
190
|
+
# Output only. The EvaluationRuns that this Evaluation is associated with.
|
|
191
|
+
# @!attribute [r] etag
|
|
192
|
+
# @return [::String]
|
|
193
|
+
# Output only. Etag used to ensure the object hasn't changed during a
|
|
194
|
+
# read-modify-write operation. If the etag is empty, the update will
|
|
195
|
+
# overwrite any concurrent changes.
|
|
196
|
+
# @!attribute [r] aggregated_metrics
|
|
197
|
+
# @return [::Google::Cloud::Ces::V1beta::AggregatedMetrics]
|
|
198
|
+
# Output only. The aggregated metrics for this evaluation across all runs.
|
|
199
|
+
# @!attribute [r] last_completed_result
|
|
200
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult]
|
|
201
|
+
# Output only. The latest evaluation result for this evaluation.
|
|
202
|
+
# @!attribute [r] invalid
|
|
203
|
+
# @return [::Boolean]
|
|
204
|
+
# Output only. Whether the evaluation is invalid. This can happen if an
|
|
205
|
+
# evaluation is referencing a tool, toolset, or agent that has since been
|
|
206
|
+
# deleted.
|
|
207
|
+
# @!attribute [r] last_ten_results
|
|
208
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult>]
|
|
209
|
+
# Output only. The last 10 evaluation results for this evaluation. This is
|
|
210
|
+
# only populated if include_last_ten_results is set to true in the
|
|
211
|
+
# ListEvaluationsRequest or GetEvaluationRequest.
|
|
212
|
+
class Evaluation
|
|
213
|
+
include ::Google::Protobuf::MessageExts
|
|
214
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
215
|
+
|
|
216
|
+
# Represents a single, checkable requirement.
|
|
217
|
+
# @!attribute [rw] tool_call
|
|
218
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolCall]
|
|
219
|
+
# Optional. Check that a specific tool was called with the parameters.
|
|
220
|
+
#
|
|
221
|
+
# Note: The following fields are mutually exclusive: `tool_call`, `tool_response`, `agent_response`, `agent_transfer`, `updated_variables`, `mock_tool_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
222
|
+
# @!attribute [rw] tool_response
|
|
223
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolResponse]
|
|
224
|
+
# Optional. Check that a specific tool had the expected response.
|
|
225
|
+
#
|
|
226
|
+
# Note: The following fields are mutually exclusive: `tool_response`, `tool_call`, `agent_response`, `agent_transfer`, `updated_variables`, `mock_tool_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
227
|
+
# @!attribute [rw] agent_response
|
|
228
|
+
# @return [::Google::Cloud::Ces::V1beta::Message]
|
|
229
|
+
# Optional. Check that the agent responded with the correct response. The
|
|
230
|
+
# role "agent" is implied.
|
|
231
|
+
#
|
|
232
|
+
# Note: The following fields are mutually exclusive: `agent_response`, `tool_call`, `tool_response`, `agent_transfer`, `updated_variables`, `mock_tool_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
233
|
+
# @!attribute [rw] agent_transfer
|
|
234
|
+
# @return [::Google::Cloud::Ces::V1beta::AgentTransfer]
|
|
235
|
+
# Optional. Check that the agent transferred the conversation to a
|
|
236
|
+
# different agent.
|
|
237
|
+
#
|
|
238
|
+
# Note: The following fields are mutually exclusive: `agent_transfer`, `tool_call`, `tool_response`, `agent_response`, `updated_variables`, `mock_tool_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
239
|
+
# @!attribute [rw] updated_variables
|
|
240
|
+
# @return [::Google::Protobuf::Struct]
|
|
241
|
+
# Optional. Check that the agent updated the session variables to the
|
|
242
|
+
# expected values. Used to also capture agent variable updates for golden
|
|
243
|
+
# evals.
|
|
244
|
+
#
|
|
245
|
+
# Note: The following fields are mutually exclusive: `updated_variables`, `tool_call`, `tool_response`, `agent_response`, `agent_transfer`, `mock_tool_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
246
|
+
# @!attribute [rw] mock_tool_response
|
|
247
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolResponse]
|
|
248
|
+
# Optional. The tool response to mock, with the parameters of interest
|
|
249
|
+
# specified. Any parameters not specified will be hallucinated by the
|
|
250
|
+
# LLM.
|
|
251
|
+
#
|
|
252
|
+
# Note: The following fields are mutually exclusive: `mock_tool_response`, `tool_call`, `tool_response`, `agent_response`, `agent_transfer`, `updated_variables`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
253
|
+
# @!attribute [rw] note
|
|
254
|
+
# @return [::String]
|
|
255
|
+
# Optional. A note for this requirement, useful in reporting when specific
|
|
256
|
+
# checks fail. E.g., "Check_Payment_Tool_Called".
|
|
257
|
+
class GoldenExpectation
|
|
258
|
+
include ::Google::Protobuf::MessageExts
|
|
259
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
260
|
+
end
|
|
261
|
+
|
|
262
|
+
# A step defines a singular action to happen during the evaluation.
|
|
263
|
+
# @!attribute [rw] user_input
|
|
264
|
+
# @return [::Google::Cloud::Ces::V1beta::SessionInput]
|
|
265
|
+
# Optional. User input for the conversation.
|
|
266
|
+
#
|
|
267
|
+
# Note: The following fields are mutually exclusive: `user_input`, `agent_transfer`, `expectation`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
268
|
+
# @!attribute [rw] agent_transfer
|
|
269
|
+
# @return [::Google::Cloud::Ces::V1beta::AgentTransfer]
|
|
270
|
+
# Optional. Transfer the conversation to a different agent.
|
|
271
|
+
#
|
|
272
|
+
# Note: The following fields are mutually exclusive: `agent_transfer`, `user_input`, `expectation`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
273
|
+
# @!attribute [rw] expectation
|
|
274
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::GoldenExpectation]
|
|
275
|
+
# Optional. Executes an expectation on the current turn.
|
|
276
|
+
#
|
|
277
|
+
# Note: The following fields are mutually exclusive: `expectation`, `user_input`, `agent_transfer`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
278
|
+
class Step
|
|
279
|
+
include ::Google::Protobuf::MessageExts
|
|
280
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
281
|
+
end
|
|
282
|
+
|
|
283
|
+
# A golden turn defines a single turn in a golden conversation.
|
|
284
|
+
# @!attribute [rw] steps
|
|
285
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::Evaluation::Step>]
|
|
286
|
+
# Required. The steps required to replay a golden conversation.
|
|
287
|
+
# @!attribute [rw] root_span
|
|
288
|
+
# @return [::Google::Cloud::Ces::V1beta::Span]
|
|
289
|
+
# Optional. The root span of the golden turn for processing and maintaining
|
|
290
|
+
# audio information.
|
|
291
|
+
class GoldenTurn
|
|
292
|
+
include ::Google::Protobuf::MessageExts
|
|
293
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
294
|
+
end
|
|
295
|
+
|
|
296
|
+
# The steps required to replay a golden conversation.
|
|
297
|
+
# @!attribute [rw] turns
|
|
298
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::Evaluation::GoldenTurn>]
|
|
299
|
+
# Required. The golden turns required to replay a golden conversation.
|
|
300
|
+
# @!attribute [rw] evaluation_expectations
|
|
301
|
+
# @return [::Array<::String>]
|
|
302
|
+
# Optional. The evaluation expectations to evaluate the replayed
|
|
303
|
+
# conversation against. Format:
|
|
304
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationExpectations/{evaluationExpectation}`
|
|
305
|
+
class Golden
|
|
306
|
+
include ::Google::Protobuf::MessageExts
|
|
307
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
308
|
+
end
|
|
309
|
+
|
|
310
|
+
# The expectation to evaluate the conversation produced by the simulation.
|
|
311
|
+
# @!attribute [rw] tool_expectation
|
|
312
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::ScenarioExpectation::ToolExpectation]
|
|
313
|
+
# Optional. The tool call and response pair to be evaluated.
|
|
314
|
+
#
|
|
315
|
+
# Note: The following fields are mutually exclusive: `tool_expectation`, `agent_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
316
|
+
# @!attribute [rw] agent_response
|
|
317
|
+
# @return [::Google::Cloud::Ces::V1beta::Message]
|
|
318
|
+
# Optional. The agent response to be evaluated.
|
|
319
|
+
#
|
|
320
|
+
# Note: The following fields are mutually exclusive: `agent_response`, `tool_expectation`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
321
|
+
class ScenarioExpectation
|
|
322
|
+
include ::Google::Protobuf::MessageExts
|
|
323
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
324
|
+
|
|
325
|
+
# The tool call and response pair to be evaluated.
|
|
326
|
+
# @!attribute [rw] expected_tool_call
|
|
327
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolCall]
|
|
328
|
+
# Required. The expected tool call, with the parameters of interest
|
|
329
|
+
# specified. Any parameters not specified will be hallucinated by the
|
|
330
|
+
# LLM.
|
|
331
|
+
# @!attribute [rw] mock_tool_response
|
|
332
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolResponse]
|
|
333
|
+
# Required. The tool response to mock, with the parameters of interest
|
|
334
|
+
# specified. Any parameters not specified will be hallucinated by the
|
|
335
|
+
# LLM.
|
|
336
|
+
class ToolExpectation
|
|
337
|
+
include ::Google::Protobuf::MessageExts
|
|
338
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
339
|
+
end
|
|
340
|
+
end
|
|
341
|
+
|
|
342
|
+
# The config for a scenario
|
|
343
|
+
# @!attribute [rw] task
|
|
344
|
+
# @return [::String]
|
|
345
|
+
# Required. The task to be targeted by the scenario.
|
|
346
|
+
# @!attribute [rw] user_facts
|
|
347
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::Evaluation::Scenario::UserFact>]
|
|
348
|
+
# Optional. The user facts to be used by the scenario.
|
|
349
|
+
# @!attribute [rw] max_turns
|
|
350
|
+
# @return [::Integer]
|
|
351
|
+
# Optional. The maximum number of turns to simulate. If not specified, the
|
|
352
|
+
# simulation will continue until the task is complete.
|
|
353
|
+
# @!attribute [rw] rubrics
|
|
354
|
+
# @return [::Array<::String>]
|
|
355
|
+
# Required. The rubrics to score the scenario against.
|
|
356
|
+
# @!attribute [rw] scenario_expectations
|
|
357
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::Evaluation::ScenarioExpectation>]
|
|
358
|
+
# Required. The ScenarioExpectations to evaluate the conversation produced
|
|
359
|
+
# by the user simulation.
|
|
360
|
+
# @!attribute [rw] variable_overrides
|
|
361
|
+
# @return [::Google::Protobuf::Struct]
|
|
362
|
+
# Optional. Variables / Session Parameters as context for the session,
|
|
363
|
+
# keyed by variable names. Members of this struct will override any default
|
|
364
|
+
# values set by the system.
|
|
365
|
+
#
|
|
366
|
+
# Note, these are different from user facts, which are facts known to the
|
|
367
|
+
# user. Variables are parameters known to the agent: i.e. MDN (phone
|
|
368
|
+
# number) passed by the telephony system.
|
|
369
|
+
# @!attribute [rw] task_completion_behavior
|
|
370
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
371
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::Scenario::TaskCompletionBehavior]
|
|
372
|
+
# Optional. Deprecated. Use user_goal_behavior instead.
|
|
373
|
+
# @!attribute [rw] user_goal_behavior
|
|
374
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::Scenario::UserGoalBehavior]
|
|
375
|
+
# Optional. The expected behavior of the user goal.
|
|
376
|
+
# @!attribute [rw] evaluation_expectations
|
|
377
|
+
# @return [::Array<::String>]
|
|
378
|
+
# Optional. The evaluation expectations to evaluate the conversation
|
|
379
|
+
# produced by the simulation against. Format:
|
|
380
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationExpectations/{evaluationExpectation}`
|
|
381
|
+
class Scenario
|
|
382
|
+
include ::Google::Protobuf::MessageExts
|
|
383
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
384
|
+
|
|
385
|
+
# Facts about the user as a key value pair.
|
|
386
|
+
# @!attribute [rw] name
|
|
387
|
+
# @return [::String]
|
|
388
|
+
# Required. The name of the user fact.
|
|
389
|
+
# @!attribute [rw] value
|
|
390
|
+
# @return [::String]
|
|
391
|
+
# Required. The value of the user fact.
|
|
392
|
+
class UserFact
|
|
393
|
+
include ::Google::Protobuf::MessageExts
|
|
394
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
395
|
+
end
|
|
396
|
+
|
|
397
|
+
# The expected behavior of the user task. This is used to determine whether
|
|
398
|
+
# the scenario is successful.
|
|
399
|
+
module TaskCompletionBehavior
|
|
400
|
+
# Behavior unspecified. Will default to TASK_SATISFIED.
|
|
401
|
+
TASK_COMPLETION_BEHAVIOR_UNSPECIFIED = 0
|
|
402
|
+
|
|
403
|
+
# The user task should be completed successfully.
|
|
404
|
+
TASK_SATISFIED = 1
|
|
405
|
+
|
|
406
|
+
# The user task should be rejected.
|
|
407
|
+
TASK_REJECTED = 2
|
|
408
|
+
end
|
|
409
|
+
|
|
410
|
+
# The expected behavior of the user goal. This is used to determine whether
|
|
411
|
+
# the scenario is successful.
|
|
412
|
+
module UserGoalBehavior
|
|
413
|
+
# Behavior unspecified. Will default to USER_GOAL_SATISFIED.
|
|
414
|
+
USER_GOAL_BEHAVIOR_UNSPECIFIED = 0
|
|
415
|
+
|
|
416
|
+
# The user goal should be completed successfully.
|
|
417
|
+
USER_GOAL_SATISFIED = 1
|
|
418
|
+
|
|
419
|
+
# The user goal should be rejected.
|
|
420
|
+
USER_GOAL_REJECTED = 2
|
|
421
|
+
|
|
422
|
+
# Ignore the user goal status.
|
|
423
|
+
USER_GOAL_IGNORED = 3
|
|
424
|
+
end
|
|
425
|
+
end
|
|
426
|
+
end
|
|
427
|
+
|
|
428
|
+
# An evaluation dataset represents a set of evaluations that are grouped
|
|
429
|
+
# together basaed on shared tags.
|
|
430
|
+
# @!attribute [rw] name
|
|
431
|
+
# @return [::String]
|
|
432
|
+
# Identifier. The unique identifier of this evaluation dataset.
|
|
433
|
+
# Format:
|
|
434
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationDatasets/{evaluationDataset}`
|
|
435
|
+
# @!attribute [rw] display_name
|
|
436
|
+
# @return [::String]
|
|
437
|
+
# Required. User-defined display name of the evaluation dataset. Unique
|
|
438
|
+
# within an App.
|
|
439
|
+
# @!attribute [rw] evaluations
|
|
440
|
+
# @return [::Array<::String>]
|
|
441
|
+
# Optional. Evaluations that are included in this dataset.
|
|
442
|
+
# @!attribute [r] create_time
|
|
443
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
444
|
+
# Output only. Timestamp when the evaluation dataset was created.
|
|
445
|
+
# @!attribute [r] update_time
|
|
446
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
447
|
+
# Output only. Timestamp when the evaluation dataset was last updated.
|
|
448
|
+
# @!attribute [r] etag
|
|
449
|
+
# @return [::String]
|
|
450
|
+
# Output only. Etag used to ensure the object hasn't changed during a
|
|
451
|
+
# read-modify-write operation. If the etag is empty, the update will
|
|
452
|
+
# overwrite any concurrent changes.
|
|
453
|
+
# @!attribute [r] created_by
|
|
454
|
+
# @return [::String]
|
|
455
|
+
# Output only. The user who created the evaluation dataset.
|
|
456
|
+
# @!attribute [r] last_updated_by
|
|
457
|
+
# @return [::String]
|
|
458
|
+
# Output only. The user who last updated the evaluation dataset.
|
|
459
|
+
# @!attribute [r] aggregated_metrics
|
|
460
|
+
# @return [::Google::Cloud::Ces::V1beta::AggregatedMetrics]
|
|
461
|
+
# Output only. The aggregated metrics for this evaluation dataset across all
|
|
462
|
+
# runs.
|
|
463
|
+
class EvaluationDataset
|
|
464
|
+
include ::Google::Protobuf::MessageExts
|
|
465
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
466
|
+
end
|
|
467
|
+
|
|
468
|
+
# An evaluation result represents the output of running an Evaluation.
|
|
469
|
+
# @!attribute [r] golden_result
|
|
470
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::GoldenResult]
|
|
471
|
+
# Output only. The outcome of a golden evaluation.
|
|
472
|
+
#
|
|
473
|
+
# Note: The following fields are mutually exclusive: `golden_result`, `scenario_result`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
474
|
+
# @!attribute [r] scenario_result
|
|
475
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::ScenarioResult]
|
|
476
|
+
# Output only. The outcome of a scenario evaluation.
|
|
477
|
+
#
|
|
478
|
+
# Note: The following fields are mutually exclusive: `scenario_result`, `golden_result`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
479
|
+
# @!attribute [rw] name
|
|
480
|
+
# @return [::String]
|
|
481
|
+
# Identifier. The unique identifier of the evaluation result.
|
|
482
|
+
# Format:
|
|
483
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluations/{evaluation}/results/{result}`
|
|
484
|
+
# @!attribute [rw] display_name
|
|
485
|
+
# @return [::String]
|
|
486
|
+
# Required. Display name of the Evaluation Result. Unique within an
|
|
487
|
+
# Evaluation. By default, it has the following format:
|
|
488
|
+
# "<evaluation-display-name> result
|
|
489
|
+
# - <timestamp>".
|
|
490
|
+
# @!attribute [r] create_time
|
|
491
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
492
|
+
# Output only. Timestamp when the evaluation result was created.
|
|
493
|
+
# @!attribute [r] evaluation_status
|
|
494
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
495
|
+
# Output only. The outcome of the evaluation. Only populated if
|
|
496
|
+
# execution_state is COMPLETE.
|
|
497
|
+
# @!attribute [r] evaluation_run
|
|
498
|
+
# @return [::String]
|
|
499
|
+
# Output only. The evaluation run that produced this result.
|
|
500
|
+
# Format:
|
|
501
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationRuns/{evaluationRun}`
|
|
502
|
+
# @!attribute [r] persona
|
|
503
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationPersona]
|
|
504
|
+
# Output only. The persona used to generate the conversation for the
|
|
505
|
+
# evaluation result.
|
|
506
|
+
# @!attribute [r] error_info
|
|
507
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationErrorInfo]
|
|
508
|
+
# Output only. Error information for the evaluation result.
|
|
509
|
+
# @!attribute [r] error
|
|
510
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
511
|
+
# @return [::Google::Rpc::Status]
|
|
512
|
+
# Output only. Deprecated: Use `error_info` instead. Errors encountered
|
|
513
|
+
# during execution.
|
|
514
|
+
# @!attribute [r] initiated_by
|
|
515
|
+
# @return [::String]
|
|
516
|
+
# Output only. The user who initiated the evaluation run that resulted in
|
|
517
|
+
# this result.
|
|
518
|
+
# @!attribute [r] app_version
|
|
519
|
+
# @return [::String]
|
|
520
|
+
# Output only. The app version used to generate the conversation that
|
|
521
|
+
# resulted in this result. Format:
|
|
522
|
+
# `projects/{project}/locations/{location}/apps/{app}/versions/{version}`
|
|
523
|
+
# @!attribute [r] app_version_display_name
|
|
524
|
+
# @return [::String]
|
|
525
|
+
# Output only. The display name of the `app_version` that the evaluation ran
|
|
526
|
+
# against.
|
|
527
|
+
# @!attribute [r] changelog
|
|
528
|
+
# @return [::String]
|
|
529
|
+
# Output only. The changelog of the app version that the evaluation ran
|
|
530
|
+
# against. This is populated if user runs evaluation on latest/draft.
|
|
531
|
+
# @!attribute [r] changelog_create_time
|
|
532
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
533
|
+
# Output only. The create time of the changelog of the app version that the
|
|
534
|
+
# evaluation ran against. This is populated if user runs evaluation on
|
|
535
|
+
# latest/draft.
|
|
536
|
+
# @!attribute [r] execution_state
|
|
537
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::ExecutionState]
|
|
538
|
+
# Output only. The state of the evaluation result execution.
|
|
539
|
+
# @!attribute [r] evaluation_metrics_thresholds
|
|
540
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationMetricsThresholds]
|
|
541
|
+
# Output only. The evaluation thresholds for the result.
|
|
542
|
+
# @!attribute [r] config
|
|
543
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationConfig]
|
|
544
|
+
# Output only. The configuration used in the evaluation run that resulted in
|
|
545
|
+
# this result.
|
|
546
|
+
# @!attribute [r] golden_run_method
|
|
547
|
+
# @return [::Google::Cloud::Ces::V1beta::GoldenRunMethod]
|
|
548
|
+
# Output only. The method used to run the golden evaluation.
|
|
549
|
+
class EvaluationResult
|
|
550
|
+
include ::Google::Protobuf::MessageExts
|
|
551
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
552
|
+
|
|
553
|
+
# Specifies the expectation and the result of that expectation.
|
|
554
|
+
# @!attribute [r] observed_tool_call
|
|
555
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolCall]
|
|
556
|
+
# Output only. The result of the tool call expectation.
|
|
557
|
+
#
|
|
558
|
+
# Note: The following fields are mutually exclusive: `observed_tool_call`, `observed_tool_response`, `observed_agent_response`, `observed_agent_transfer`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
559
|
+
# @!attribute [r] observed_tool_response
|
|
560
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolResponse]
|
|
561
|
+
# Output only. The result of the tool response expectation.
|
|
562
|
+
#
|
|
563
|
+
# Note: The following fields are mutually exclusive: `observed_tool_response`, `observed_tool_call`, `observed_agent_response`, `observed_agent_transfer`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
564
|
+
# @!attribute [r] observed_agent_response
|
|
565
|
+
# @return [::Google::Cloud::Ces::V1beta::Message]
|
|
566
|
+
# Output only. The result of the agent response expectation.
|
|
567
|
+
#
|
|
568
|
+
# Note: The following fields are mutually exclusive: `observed_agent_response`, `observed_tool_call`, `observed_tool_response`, `observed_agent_transfer`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
569
|
+
# @!attribute [r] observed_agent_transfer
|
|
570
|
+
# @return [::Google::Cloud::Ces::V1beta::AgentTransfer]
|
|
571
|
+
# Output only. The result of the agent transfer expectation.
|
|
572
|
+
#
|
|
573
|
+
# Note: The following fields are mutually exclusive: `observed_agent_transfer`, `observed_tool_call`, `observed_tool_response`, `observed_agent_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
574
|
+
# @!attribute [r] expectation
|
|
575
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::GoldenExpectation]
|
|
576
|
+
# Output only. The expectation that was evaluated.
|
|
577
|
+
# @!attribute [r] outcome
|
|
578
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
579
|
+
# Output only. The outcome of the expectation.
|
|
580
|
+
# @!attribute [r] semantic_similarity_result
|
|
581
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
582
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::SemanticSimilarityResult]
|
|
583
|
+
# Output only. The result of the semantic similarity check.
|
|
584
|
+
# @!attribute [r] tool_invocation_result
|
|
585
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::GoldenExpectationOutcome::ToolInvocationResult]
|
|
586
|
+
# Output only. The result of the tool invocation check.
|
|
587
|
+
class GoldenExpectationOutcome
|
|
588
|
+
include ::Google::Protobuf::MessageExts
|
|
589
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
590
|
+
|
|
591
|
+
# The result of the tool invocation check.
|
|
592
|
+
# @!attribute [r] parameter_correctness_score
|
|
593
|
+
# @return [::Float]
|
|
594
|
+
# Output only. The tool invocation parameter correctness score. This
|
|
595
|
+
# indicates the percent of parameters from the expected tool call that
|
|
596
|
+
# were also present in the actual tool call.
|
|
597
|
+
# @!attribute [r] outcome
|
|
598
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
599
|
+
# Output only. The outcome of the tool invocation check. This is
|
|
600
|
+
# determined by comparing the parameter_correctness_score to the
|
|
601
|
+
# threshold. If the score is equal to or above the threshold, the outcome
|
|
602
|
+
# will be PASS. Otherwise, the outcome will be FAIL.
|
|
603
|
+
# @!attribute [r] explanation
|
|
604
|
+
# @return [::String]
|
|
605
|
+
# Output only. A free text explanation for the tool invocation result.
|
|
606
|
+
class ToolInvocationResult
|
|
607
|
+
include ::Google::Protobuf::MessageExts
|
|
608
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
609
|
+
end
|
|
610
|
+
end
|
|
611
|
+
|
|
612
|
+
# The result of a single evaluation expectation.
|
|
613
|
+
# @!attribute [r] evaluation_expectation
|
|
614
|
+
# @return [::String]
|
|
615
|
+
# Output only. The evaluation expectation.
|
|
616
|
+
# Format:
|
|
617
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationExpectations/{evaluation_expectation}`
|
|
618
|
+
# @!attribute [r] prompt
|
|
619
|
+
# @return [::String]
|
|
620
|
+
# Output only. The prompt that was used for the evaluation.
|
|
621
|
+
# @!attribute [r] outcome
|
|
622
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
623
|
+
# Output only. The outcome of the evaluation expectation.
|
|
624
|
+
# @!attribute [r] explanation
|
|
625
|
+
# @return [::String]
|
|
626
|
+
# Output only. The explanation for the result.
|
|
627
|
+
class EvaluationExpectationResult
|
|
628
|
+
include ::Google::Protobuf::MessageExts
|
|
629
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
630
|
+
end
|
|
631
|
+
|
|
632
|
+
# The result of a golden evaluation.
|
|
633
|
+
# @!attribute [r] turn_replay_results
|
|
634
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::GoldenResult::TurnReplayResult>]
|
|
635
|
+
# Output only. The result of running each turn of the golden conversation.
|
|
636
|
+
# @!attribute [r] evaluation_expectation_results
|
|
637
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::EvaluationExpectationResult>]
|
|
638
|
+
# Output only. The results of the evaluation expectations.
|
|
639
|
+
class GoldenResult
|
|
640
|
+
include ::Google::Protobuf::MessageExts
|
|
641
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
642
|
+
|
|
643
|
+
# The result of running a single turn of the golden conversation.
|
|
644
|
+
# @!attribute [r] conversation
|
|
645
|
+
# @return [::String]
|
|
646
|
+
# Output only. The conversation that was generated for this turn.
|
|
647
|
+
# @!attribute [r] expectation_outcome
|
|
648
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::GoldenExpectationOutcome>]
|
|
649
|
+
# Output only. The outcome of each expectation.
|
|
650
|
+
# @!attribute [r] hallucination_result
|
|
651
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::HallucinationResult]
|
|
652
|
+
# Output only. The result of the hallucination check.
|
|
653
|
+
# @!attribute [r] tool_invocation_score
|
|
654
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
655
|
+
# @return [::Float]
|
|
656
|
+
# Output only. Deprecated. Use OverallToolInvocationResult instead.
|
|
657
|
+
# @!attribute [r] tool_ordered_invocation_score
|
|
658
|
+
# @return [::Float]
|
|
659
|
+
# Output only. The overall tool ordered invocation score for this turn.
|
|
660
|
+
# This indicates the overall percent of tools from the expected turn that
|
|
661
|
+
# were actually invoked in the expected order.
|
|
662
|
+
# @!attribute [r] turn_latency
|
|
663
|
+
# @return [::Google::Protobuf::Duration]
|
|
664
|
+
# Output only. Duration of the turn.
|
|
665
|
+
# @!attribute [r] tool_call_latencies
|
|
666
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::ToolCallLatency>]
|
|
667
|
+
# Output only. The latency of each tool call in the turn.
|
|
668
|
+
# @!attribute [r] semantic_similarity_result
|
|
669
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::SemanticSimilarityResult]
|
|
670
|
+
# Output only. The result of the semantic similarity check.
|
|
671
|
+
# @!attribute [r] overall_tool_invocation_result
|
|
672
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::OverallToolInvocationResult]
|
|
673
|
+
# Output only. The result of the overall tool invocation check.
|
|
674
|
+
# @!attribute [r] error_info
|
|
675
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationErrorInfo]
|
|
676
|
+
# Output only. Information about the error that occurred during this
|
|
677
|
+
# turn.
|
|
678
|
+
# @!attribute [r] span_latencies
|
|
679
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::SpanLatency>]
|
|
680
|
+
# Output only. The latency of spans in the turn.
|
|
681
|
+
class TurnReplayResult
|
|
682
|
+
include ::Google::Protobuf::MessageExts
|
|
683
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
684
|
+
end
|
|
685
|
+
end
|
|
686
|
+
|
|
687
|
+
# The outcome of the evaluation against the rubric.
|
|
688
|
+
# @!attribute [r] rubric
|
|
689
|
+
# @return [::String]
|
|
690
|
+
# Output only. The rubric that was used to evaluate the conversation.
|
|
691
|
+
# @!attribute [r] score
|
|
692
|
+
# @return [::Float]
|
|
693
|
+
# Output only. The score of the conversation against the rubric.
|
|
694
|
+
# @!attribute [r] score_explanation
|
|
695
|
+
# @return [::String]
|
|
696
|
+
# Output only. The rater's response to the rubric.
|
|
697
|
+
class ScenarioRubricOutcome
|
|
698
|
+
include ::Google::Protobuf::MessageExts
|
|
699
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
700
|
+
end
|
|
701
|
+
|
|
702
|
+
# The outcome of a scenario expectation.
|
|
703
|
+
# @!attribute [r] observed_tool_call
|
|
704
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::ScenarioExpectationOutcome::ObservedToolCall]
|
|
705
|
+
# Output only. The observed tool call.
|
|
706
|
+
#
|
|
707
|
+
# Note: The following fields are mutually exclusive: `observed_tool_call`, `observed_agent_response`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
708
|
+
# @!attribute [r] observed_agent_response
|
|
709
|
+
# @return [::Google::Cloud::Ces::V1beta::Message]
|
|
710
|
+
# Output only. The observed agent response.
|
|
711
|
+
#
|
|
712
|
+
# Note: The following fields are mutually exclusive: `observed_agent_response`, `observed_tool_call`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
713
|
+
# @!attribute [r] expectation
|
|
714
|
+
# @return [::Google::Cloud::Ces::V1beta::Evaluation::ScenarioExpectation]
|
|
715
|
+
# Output only. The expectation that was evaluated.
|
|
716
|
+
# @!attribute [r] outcome
|
|
717
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
718
|
+
# Output only. The outcome of the ScenarioExpectation.
|
|
719
|
+
class ScenarioExpectationOutcome
|
|
720
|
+
include ::Google::Protobuf::MessageExts
|
|
721
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
722
|
+
|
|
723
|
+
# The observed tool call and response.
|
|
724
|
+
# @!attribute [r] tool_call
|
|
725
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolCall]
|
|
726
|
+
# Output only. The observed tool call.
|
|
727
|
+
# @!attribute [r] tool_response
|
|
728
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolResponse]
|
|
729
|
+
# Output only. The observed tool response.
|
|
730
|
+
class ObservedToolCall
|
|
731
|
+
include ::Google::Protobuf::MessageExts
|
|
732
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
733
|
+
end
|
|
734
|
+
end
|
|
735
|
+
|
|
736
|
+
# The outcome of a scenario evaluation.
|
|
737
|
+
# @!attribute [r] conversation
|
|
738
|
+
# @return [::String]
|
|
739
|
+
# Output only. The conversation that was generated in the scenario.
|
|
740
|
+
# @!attribute [r] task
|
|
741
|
+
# @return [::String]
|
|
742
|
+
# Output only. The task that was used when running the scenario for this
|
|
743
|
+
# result.
|
|
744
|
+
# @!attribute [r] user_facts
|
|
745
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::Evaluation::Scenario::UserFact>]
|
|
746
|
+
# Output only. The user facts that were used by the scenario for this
|
|
747
|
+
# result.
|
|
748
|
+
# @!attribute [r] expectation_outcomes
|
|
749
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::ScenarioExpectationOutcome>]
|
|
750
|
+
# Output only. The outcome of each expectation.
|
|
751
|
+
# @!attribute [r] rubric_outcomes
|
|
752
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::ScenarioRubricOutcome>]
|
|
753
|
+
# Output only. The outcome of the rubric.
|
|
754
|
+
# @!attribute [r] hallucination_result
|
|
755
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::HallucinationResult>]
|
|
756
|
+
# Output only. The result of the hallucination check.
|
|
757
|
+
# There will be one hallucination result for each turn in the conversation.
|
|
758
|
+
# @!attribute [r] task_completion_result
|
|
759
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
760
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::TaskCompletionResult]
|
|
761
|
+
# Output only. The result of the task completion check.
|
|
762
|
+
# @!attribute [r] tool_call_latencies
|
|
763
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::ToolCallLatency>]
|
|
764
|
+
# Output only. The latency of each tool call execution in the conversation.
|
|
765
|
+
# @!attribute [r] user_goal_satisfaction_result
|
|
766
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::UserGoalSatisfactionResult]
|
|
767
|
+
# Output only. The result of the user goal satisfaction check.
|
|
768
|
+
# @!attribute [r] all_expectations_satisfied
|
|
769
|
+
# @return [::Boolean]
|
|
770
|
+
# Output only. Whether all expectations were satisfied for this turn.
|
|
771
|
+
# @!attribute [r] task_completed
|
|
772
|
+
# @return [::Boolean]
|
|
773
|
+
# Output only. Whether the task was completed for this turn. This is a
|
|
774
|
+
# composite of all expectations satisfied, no hallucinations, and user goal
|
|
775
|
+
# satisfaction.
|
|
776
|
+
# @!attribute [r] span_latencies
|
|
777
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::SpanLatency>]
|
|
778
|
+
# Output only. The latency of spans in the conversation.
|
|
779
|
+
# @!attribute [r] evaluation_expectation_results
|
|
780
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::EvaluationResult::EvaluationExpectationResult>]
|
|
781
|
+
# Output only. The results of the evaluation expectations.
|
|
782
|
+
class ScenarioResult
|
|
783
|
+
include ::Google::Protobuf::MessageExts
|
|
784
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
785
|
+
end
|
|
786
|
+
|
|
787
|
+
# The latency of a span execution.
|
|
788
|
+
# @!attribute [r] resource
|
|
789
|
+
# @return [::String]
|
|
790
|
+
# Output only. The resource name of the guardrail or tool spans.
|
|
791
|
+
#
|
|
792
|
+
# Note: The following fields are mutually exclusive: `resource`, `toolset`, `model`, `callback`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
793
|
+
# @!attribute [r] toolset
|
|
794
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolsetTool]
|
|
795
|
+
# Output only. The toolset tool identifier.
|
|
796
|
+
#
|
|
797
|
+
# Note: The following fields are mutually exclusive: `toolset`, `resource`, `model`, `callback`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
798
|
+
# @!attribute [r] model
|
|
799
|
+
# @return [::String]
|
|
800
|
+
# Output only. The name of the LLM span.
|
|
801
|
+
#
|
|
802
|
+
# Note: The following fields are mutually exclusive: `model`, `resource`, `toolset`, `callback`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
803
|
+
# @!attribute [r] callback
|
|
804
|
+
# @return [::String]
|
|
805
|
+
# Output only. The name of the user callback span.
|
|
806
|
+
#
|
|
807
|
+
# Note: The following fields are mutually exclusive: `callback`, `resource`, `toolset`, `model`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
808
|
+
# @!attribute [r] type
|
|
809
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::SpanLatency::Type]
|
|
810
|
+
# Output only. The type of span.
|
|
811
|
+
# @!attribute [r] display_name
|
|
812
|
+
# @return [::String]
|
|
813
|
+
# Output only. The display name of the span. Applicable to tool and
|
|
814
|
+
# guardrail spans.
|
|
815
|
+
# @!attribute [r] start_time
|
|
816
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
817
|
+
# Output only. The start time of span.
|
|
818
|
+
# @!attribute [r] end_time
|
|
819
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
820
|
+
# Output only. The end time of span.
|
|
821
|
+
# @!attribute [r] execution_latency
|
|
822
|
+
# @return [::Google::Protobuf::Duration]
|
|
823
|
+
# Output only. The latency of span.
|
|
824
|
+
class SpanLatency
|
|
825
|
+
include ::Google::Protobuf::MessageExts
|
|
826
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
827
|
+
|
|
828
|
+
# The type of span.
|
|
829
|
+
# Additional values may be added in the future.
|
|
830
|
+
module Type
|
|
831
|
+
# Default value. This value is unused.
|
|
832
|
+
TYPE_UNSPECIFIED = 0
|
|
833
|
+
|
|
834
|
+
# Tool call span.
|
|
835
|
+
TOOL = 1
|
|
836
|
+
|
|
837
|
+
# User callback span.
|
|
838
|
+
USER_CALLBACK = 2
|
|
839
|
+
|
|
840
|
+
# Guardrail span.
|
|
841
|
+
GUARDRAIL = 3
|
|
842
|
+
|
|
843
|
+
# LLM span.
|
|
844
|
+
LLM = 4
|
|
845
|
+
end
|
|
846
|
+
end
|
|
847
|
+
|
|
848
|
+
# The latency of a tool call execution.
|
|
849
|
+
# @!attribute [r] tool
|
|
850
|
+
# @return [::String]
|
|
851
|
+
# Output only. The name of the tool that got executed.
|
|
852
|
+
# Format:
|
|
853
|
+
# `projects/{project}/locations/{location}/apps/{app}/tools/{tool}`.
|
|
854
|
+
# @!attribute [r] display_name
|
|
855
|
+
# @return [::String]
|
|
856
|
+
# Output only. The display name of the tool.
|
|
857
|
+
# @!attribute [r] start_time
|
|
858
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
859
|
+
# Output only. The start time of the tool call execution.
|
|
860
|
+
# @!attribute [r] end_time
|
|
861
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
862
|
+
# Output only. The end time of the tool call execution.
|
|
863
|
+
# @!attribute [r] execution_latency
|
|
864
|
+
# @return [::Google::Protobuf::Duration]
|
|
865
|
+
# Output only. The latency of the tool call execution.
|
|
866
|
+
class ToolCallLatency
|
|
867
|
+
include ::Google::Protobuf::MessageExts
|
|
868
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
869
|
+
end
|
|
870
|
+
|
|
871
|
+
# The result of the hallucination check for a single turn.
|
|
872
|
+
# @!attribute [r] score
|
|
873
|
+
# @return [::Integer]
|
|
874
|
+
# Output only. The hallucination score. Can be -1, 0, 1.
|
|
875
|
+
# @!attribute [r] label
|
|
876
|
+
# @return [::String]
|
|
877
|
+
# Output only. The label associated with each score.
|
|
878
|
+
# Score 1: Justified
|
|
879
|
+
# Score 0: Not Justified
|
|
880
|
+
# Score -1: No Claim To Assess
|
|
881
|
+
# @!attribute [r] explanation
|
|
882
|
+
# @return [::String]
|
|
883
|
+
# Output only. The explanation for the hallucination score.
|
|
884
|
+
class HallucinationResult
|
|
885
|
+
include ::Google::Protobuf::MessageExts
|
|
886
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
887
|
+
end
|
|
888
|
+
|
|
889
|
+
# The result of a user goal satisfaction check for a conversation.
|
|
890
|
+
# @!attribute [r] score
|
|
891
|
+
# @return [::Integer]
|
|
892
|
+
# Output only. The user task satisfaction score. Can be -1, 0, 1.
|
|
893
|
+
# @!attribute [r] label
|
|
894
|
+
# @return [::String]
|
|
895
|
+
# Output only. The label associated with each score.
|
|
896
|
+
# Score 1: User Task Satisfied
|
|
897
|
+
# Score 0: User Task Not Satisfied
|
|
898
|
+
# Score -1: User Task Unspecified
|
|
899
|
+
# @!attribute [r] explanation
|
|
900
|
+
# @return [::String]
|
|
901
|
+
# Output only. The explanation for the user task satisfaction score.
|
|
902
|
+
class UserGoalSatisfactionResult
|
|
903
|
+
include ::Google::Protobuf::MessageExts
|
|
904
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
905
|
+
end
|
|
906
|
+
|
|
907
|
+
# The result of the semantic similarity check.
|
|
908
|
+
# @!attribute [r] score
|
|
909
|
+
# @return [::Integer]
|
|
910
|
+
# Output only. The semantic similarity score. Can be 0, 1, 2, 3, or 4.
|
|
911
|
+
# @!attribute [r] label
|
|
912
|
+
# @return [::String]
|
|
913
|
+
# Output only. The label associated with each score.
|
|
914
|
+
# Score 4: Fully Consistent
|
|
915
|
+
# Score 3: Mostly Consistent
|
|
916
|
+
# Score 2: Partially Consistent (Minor Omissions)
|
|
917
|
+
# Score 1: Largely Inconsistent (Major Omissions)
|
|
918
|
+
# Score 0: Completely Inconsistent / Contradictory
|
|
919
|
+
# @!attribute [r] explanation
|
|
920
|
+
# @return [::String]
|
|
921
|
+
# Output only. The explanation for the semantic similarity score.
|
|
922
|
+
# @!attribute [r] outcome
|
|
923
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
924
|
+
# Output only. The outcome of the semantic similarity check. This is
|
|
925
|
+
# determined by comparing the score to the
|
|
926
|
+
# semantic_similarity_success_threshold. If the score is equal to or above
|
|
927
|
+
# the threshold, the outcome will be PASS. Otherwise, the outcome will be
|
|
928
|
+
# FAIL.
|
|
929
|
+
class SemanticSimilarityResult
|
|
930
|
+
include ::Google::Protobuf::MessageExts
|
|
931
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
932
|
+
end
|
|
933
|
+
|
|
934
|
+
# The result of the overall tool invocation check.
|
|
935
|
+
# @!attribute [rw] tool_invocation_score
|
|
936
|
+
# @return [::Float]
|
|
937
|
+
# The overall tool invocation score for this turn. This indicates the
|
|
938
|
+
# overall percent of tools from the expected turn that were actually
|
|
939
|
+
# invoked.
|
|
940
|
+
# @!attribute [r] outcome
|
|
941
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationResult::Outcome]
|
|
942
|
+
# Output only. The outcome of the tool invocation check. This is determined
|
|
943
|
+
# by comparing the tool_invocation_score to the
|
|
944
|
+
# overall_tool_invocation_correctness_threshold. If the
|
|
945
|
+
# score is equal to or above the threshold, the outcome will be PASS.
|
|
946
|
+
# Otherwise, the outcome will be FAIL.
|
|
947
|
+
class OverallToolInvocationResult
|
|
948
|
+
include ::Google::Protobuf::MessageExts
|
|
949
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
950
|
+
end
|
|
951
|
+
|
|
952
|
+
# The result of the task completion check for the conversation.
|
|
953
|
+
# @!attribute [r] score
|
|
954
|
+
# @return [::Integer]
|
|
955
|
+
# Output only. The task completion score. Can be -1, 0, 1
|
|
956
|
+
# @!attribute [r] label
|
|
957
|
+
# @return [::String]
|
|
958
|
+
# Output only. The label associated with each score.
|
|
959
|
+
# Score 1: Task Completed
|
|
960
|
+
# Score 0: Task Not Completed
|
|
961
|
+
# Score -1: User Goal Undefined
|
|
962
|
+
# @!attribute [r] explanation
|
|
963
|
+
# @return [::String]
|
|
964
|
+
# Output only. The explanation for the task completion score.
|
|
965
|
+
class TaskCompletionResult
|
|
966
|
+
include ::Google::Protobuf::MessageExts
|
|
967
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
968
|
+
end
|
|
969
|
+
|
|
970
|
+
# The outcome of the evaluation or expectation.
|
|
971
|
+
module Outcome
|
|
972
|
+
# Evaluation outcome is not specified.
|
|
973
|
+
OUTCOME_UNSPECIFIED = 0
|
|
974
|
+
|
|
975
|
+
# Evaluation/Expectation passed. In the case of an evaluation, this means
|
|
976
|
+
# that all expectations were met.
|
|
977
|
+
PASS = 1
|
|
978
|
+
|
|
979
|
+
# Evaluation/Expectation failed. In the case of an evaluation, this means
|
|
980
|
+
# that at least one expectation was not met.
|
|
981
|
+
FAIL = 2
|
|
982
|
+
|
|
983
|
+
# Evaluation/Expectation was skipped.
|
|
984
|
+
SKIPPED = 3
|
|
985
|
+
end
|
|
986
|
+
|
|
987
|
+
# The state of the evaluation result execution.
|
|
988
|
+
module ExecutionState
|
|
989
|
+
# Evaluation result execution state is not specified.
|
|
990
|
+
EXECUTION_STATE_UNSPECIFIED = 0
|
|
991
|
+
|
|
992
|
+
# Evaluation result execution is running.
|
|
993
|
+
RUNNING = 1
|
|
994
|
+
|
|
995
|
+
# Evaluation result execution has completed.
|
|
996
|
+
COMPLETED = 2
|
|
997
|
+
|
|
998
|
+
# Evaluation result execution failed due to an internal error.
|
|
999
|
+
ERROR = 3
|
|
1000
|
+
end
|
|
1001
|
+
end
|
|
1002
|
+
|
|
1003
|
+
# An evaluation run represents an all the evaluation results from an
|
|
1004
|
+
# evaluation execution.
|
|
1005
|
+
# @!attribute [rw] name
|
|
1006
|
+
# @return [::String]
|
|
1007
|
+
# Identifier. The unique identifier of the evaluation run.
|
|
1008
|
+
# Format:
|
|
1009
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationRuns/{evaluationRun}`
|
|
1010
|
+
# @!attribute [rw] display_name
|
|
1011
|
+
# @return [::String]
|
|
1012
|
+
# Optional. User-defined display name of the evaluation run.
|
|
1013
|
+
# default: "<evaluation-dataset-display-name> run - <timestamp>".
|
|
1014
|
+
# @!attribute [r] evaluation_results
|
|
1015
|
+
# @return [::Array<::String>]
|
|
1016
|
+
# Output only. The evaluation results that are part of this run.
|
|
1017
|
+
# Format:
|
|
1018
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluations/{evaluation}/results/{result}`
|
|
1019
|
+
# @!attribute [r] create_time
|
|
1020
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1021
|
+
# Output only. Timestamp when the evaluation run was created.
|
|
1022
|
+
# @!attribute [r] initiated_by
|
|
1023
|
+
# @return [::String]
|
|
1024
|
+
# Output only. The user who initiated the evaluation run.
|
|
1025
|
+
# @!attribute [r] app_version
|
|
1026
|
+
# @return [::String]
|
|
1027
|
+
# Output only. The app version to evaluate.
|
|
1028
|
+
# Format:
|
|
1029
|
+
# `projects/{project}/locations/{location}/apps/{app}/versions/{version}`
|
|
1030
|
+
# @!attribute [r] app_version_display_name
|
|
1031
|
+
# @return [::String]
|
|
1032
|
+
# Output only. The display name of the `app_version` that the evaluation ran
|
|
1033
|
+
# against.
|
|
1034
|
+
# @!attribute [r] changelog
|
|
1035
|
+
# @return [::String]
|
|
1036
|
+
# Output only. The changelog of the app version that the evaluation ran
|
|
1037
|
+
# against. This is populated if user runs evaluation on latest/draft.
|
|
1038
|
+
# @!attribute [r] changelog_create_time
|
|
1039
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1040
|
+
# Output only. The create time of the changelog of the app version that the
|
|
1041
|
+
# evaluation ran against. This is populated if user runs evaluation on
|
|
1042
|
+
# latest/draft.
|
|
1043
|
+
# @!attribute [r] evaluations
|
|
1044
|
+
# @return [::Array<::String>]
|
|
1045
|
+
# Output only. The evaluations that are part of this run. The list may
|
|
1046
|
+
# contain evaluations of either type. This field is mutually exclusive with
|
|
1047
|
+
# `evaluation_dataset`.
|
|
1048
|
+
# Format:
|
|
1049
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluations/{evaluation}`
|
|
1050
|
+
# @!attribute [r] evaluation_dataset
|
|
1051
|
+
# @return [::String]
|
|
1052
|
+
# Output only. The evaluation dataset that this run is associated with. This
|
|
1053
|
+
# field is mutually exclusive with `evaluations`. Format:
|
|
1054
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationDatasets/{evaluationDataset}`
|
|
1055
|
+
# @!attribute [r] evaluation_type
|
|
1056
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationRun::EvaluationType]
|
|
1057
|
+
# Output only. The type of the evaluations in this run.
|
|
1058
|
+
# @!attribute [r] state
|
|
1059
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationRun::EvaluationRunState]
|
|
1060
|
+
# Output only. The state of the evaluation run.
|
|
1061
|
+
# @!attribute [r] progress
|
|
1062
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationRun::Progress]
|
|
1063
|
+
# Output only. The progress of the evaluation run.
|
|
1064
|
+
# @!attribute [r] config
|
|
1065
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationConfig]
|
|
1066
|
+
# Output only. The configuration used in the run.
|
|
1067
|
+
# @!attribute [r] error
|
|
1068
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
1069
|
+
# @return [::Google::Rpc::Status]
|
|
1070
|
+
# Output only. Deprecated: Use error_info instead. Errors encountered during
|
|
1071
|
+
# execution.
|
|
1072
|
+
# @!attribute [r] error_info
|
|
1073
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationErrorInfo]
|
|
1074
|
+
# Output only. Error information for the evaluation run.
|
|
1075
|
+
# @!attribute [r] evaluation_run_summaries
|
|
1076
|
+
# @return [::Google::Protobuf::Map{::String => ::Google::Cloud::Ces::V1beta::EvaluationRun::EvaluationRunSummary}]
|
|
1077
|
+
# Output only. Map of evaluation name to EvaluationRunSummary.
|
|
1078
|
+
# @!attribute [r] latency_report
|
|
1079
|
+
# @return [::Google::Cloud::Ces::V1beta::LatencyReport]
|
|
1080
|
+
# Output only. Latency report for the evaluation run.
|
|
1081
|
+
# @!attribute [r] run_count
|
|
1082
|
+
# @return [::Integer]
|
|
1083
|
+
# Output only. The number of times the evaluations inside the run were run.
|
|
1084
|
+
# @!attribute [r] persona_run_configs
|
|
1085
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::PersonaRunConfig>]
|
|
1086
|
+
# Output only. The configuration to use for the run per persona.
|
|
1087
|
+
# @!attribute [rw] optimization_config
|
|
1088
|
+
# @return [::Google::Cloud::Ces::V1beta::OptimizationConfig]
|
|
1089
|
+
# Optional. Configuration for running the optimization step after the
|
|
1090
|
+
# evaluation run. If not set, the optimization step will not be run.
|
|
1091
|
+
# @!attribute [r] scheduled_evaluation_run
|
|
1092
|
+
# @return [::String]
|
|
1093
|
+
# Output only. The scheduled evaluation run resource name that created this
|
|
1094
|
+
# evaluation run. This field is only set if the evaluation run was created
|
|
1095
|
+
# by a scheduled evaluation run.
|
|
1096
|
+
# Format:
|
|
1097
|
+
# `projects/{project}/locations/{location}/apps/{app}/scheduledEvaluationRuns/{scheduled_evaluation_run}`
|
|
1098
|
+
# @!attribute [r] golden_run_method
|
|
1099
|
+
# @return [::Google::Cloud::Ces::V1beta::GoldenRunMethod]
|
|
1100
|
+
# Output only. The method used to run the evaluation.
|
|
1101
|
+
class EvaluationRun
|
|
1102
|
+
include ::Google::Protobuf::MessageExts
|
|
1103
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1104
|
+
|
|
1105
|
+
# The progress of the evaluation run.
|
|
1106
|
+
# @!attribute [r] total_count
|
|
1107
|
+
# @return [::Integer]
|
|
1108
|
+
# Output only. Total number of evaluation results in this run.
|
|
1109
|
+
# @!attribute [r] failed_count
|
|
1110
|
+
# @return [::Integer]
|
|
1111
|
+
# Output only. Number of completed evaluation results with an outcome of
|
|
1112
|
+
# FAIL. (EvaluationResult.execution_state is COMPLETED and
|
|
1113
|
+
# EvaluationResult.evaluation_status is FAIL).
|
|
1114
|
+
# @!attribute [r] error_count
|
|
1115
|
+
# @return [::Integer]
|
|
1116
|
+
# Output only. Number of evaluation results that failed to execute.
|
|
1117
|
+
# (EvaluationResult.execution_state is ERROR).
|
|
1118
|
+
# @!attribute [r] completed_count
|
|
1119
|
+
# @return [::Integer]
|
|
1120
|
+
# Output only. Number of evaluation results that finished successfully.
|
|
1121
|
+
# (EvaluationResult.execution_state is COMPLETED).
|
|
1122
|
+
# @!attribute [r] passed_count
|
|
1123
|
+
# @return [::Integer]
|
|
1124
|
+
# Output only. Number of completed evaluation results with an outcome of
|
|
1125
|
+
# PASS. (EvaluationResult.execution_state is COMPLETED and
|
|
1126
|
+
# EvaluationResult.evaluation_status is PASS).
|
|
1127
|
+
class Progress
|
|
1128
|
+
include ::Google::Protobuf::MessageExts
|
|
1129
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1130
|
+
end
|
|
1131
|
+
|
|
1132
|
+
# Contains the summary of passed and failed result counts for a specific
|
|
1133
|
+
# evaluation in an evaluation run.
|
|
1134
|
+
# @!attribute [r] passed_count
|
|
1135
|
+
# @return [::Integer]
|
|
1136
|
+
# Output only. Number of passed results for the associated Evaluation in
|
|
1137
|
+
# this run.
|
|
1138
|
+
# @!attribute [r] failed_count
|
|
1139
|
+
# @return [::Integer]
|
|
1140
|
+
# Output only. Number of failed results for the associated Evaluation in
|
|
1141
|
+
# this run.
|
|
1142
|
+
# @!attribute [r] error_count
|
|
1143
|
+
# @return [::Integer]
|
|
1144
|
+
# Output only. Number of error results for the associated Evaluation in
|
|
1145
|
+
# this run.
|
|
1146
|
+
class EvaluationRunSummary
|
|
1147
|
+
include ::Google::Protobuf::MessageExts
|
|
1148
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1149
|
+
end
|
|
1150
|
+
|
|
1151
|
+
# @!attribute [rw] key
|
|
1152
|
+
# @return [::String]
|
|
1153
|
+
# @!attribute [rw] value
|
|
1154
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationRun::EvaluationRunSummary]
|
|
1155
|
+
class EvaluationRunSummariesEntry
|
|
1156
|
+
include ::Google::Protobuf::MessageExts
|
|
1157
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1158
|
+
end
|
|
1159
|
+
|
|
1160
|
+
# The type of the evaluations in this run.
|
|
1161
|
+
# Additional values may be added in the future.
|
|
1162
|
+
module EvaluationType
|
|
1163
|
+
# Evaluation type is not specified.
|
|
1164
|
+
EVALUATION_TYPE_UNSPECIFIED = 0
|
|
1165
|
+
|
|
1166
|
+
# Golden evaluation.
|
|
1167
|
+
GOLDEN = 1
|
|
1168
|
+
|
|
1169
|
+
# Scenario evaluation.
|
|
1170
|
+
SCENARIO = 2
|
|
1171
|
+
|
|
1172
|
+
# Indicates the run includes a mix of golden and scenario evaluations.
|
|
1173
|
+
MIXED = 3
|
|
1174
|
+
end
|
|
1175
|
+
|
|
1176
|
+
# The state of the evaluation run.
|
|
1177
|
+
module EvaluationRunState
|
|
1178
|
+
# Evaluation run state is not specified.
|
|
1179
|
+
EVALUATION_RUN_STATE_UNSPECIFIED = 0
|
|
1180
|
+
|
|
1181
|
+
# Evaluation run is running.
|
|
1182
|
+
RUNNING = 1
|
|
1183
|
+
|
|
1184
|
+
# Evaluation run has completed.
|
|
1185
|
+
COMPLETED = 2
|
|
1186
|
+
|
|
1187
|
+
# The evaluation run has an error.
|
|
1188
|
+
ERROR = 3
|
|
1189
|
+
end
|
|
1190
|
+
end
|
|
1191
|
+
|
|
1192
|
+
# Latency report for the evaluation run.
|
|
1193
|
+
# @!attribute [r] tool_latencies
|
|
1194
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::LatencyReport::ToolLatency>]
|
|
1195
|
+
# Output only. Unordered list. Latency metrics for each tool.
|
|
1196
|
+
# @!attribute [r] callback_latencies
|
|
1197
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::LatencyReport::CallbackLatency>]
|
|
1198
|
+
# Output only. Unordered list. Latency metrics for each callback.
|
|
1199
|
+
# @!attribute [r] guardrail_latencies
|
|
1200
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::LatencyReport::GuardrailLatency>]
|
|
1201
|
+
# Output only. Unordered list. Latency metrics for each guardrail.
|
|
1202
|
+
# @!attribute [r] llm_call_latencies
|
|
1203
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::LatencyReport::LlmCallLatency>]
|
|
1204
|
+
# Output only. Unordered list. Latency metrics for each LLM call.
|
|
1205
|
+
# @!attribute [r] session_count
|
|
1206
|
+
# @return [::Integer]
|
|
1207
|
+
# Output only. The total number of sessions considered in the latency report.
|
|
1208
|
+
class LatencyReport
|
|
1209
|
+
include ::Google::Protobuf::MessageExts
|
|
1210
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1211
|
+
|
|
1212
|
+
# Latency metrics for a component.
|
|
1213
|
+
# @!attribute [r] p50_latency
|
|
1214
|
+
# @return [::Google::Protobuf::Duration]
|
|
1215
|
+
# Output only. The 50th percentile latency.
|
|
1216
|
+
# @!attribute [r] p90_latency
|
|
1217
|
+
# @return [::Google::Protobuf::Duration]
|
|
1218
|
+
# Output only. The 90th percentile latency.
|
|
1219
|
+
# @!attribute [r] p99_latency
|
|
1220
|
+
# @return [::Google::Protobuf::Duration]
|
|
1221
|
+
# Output only. The 99th percentile latency.
|
|
1222
|
+
# @!attribute [r] call_count
|
|
1223
|
+
# @return [::Integer]
|
|
1224
|
+
# Output only. The number of times the resource was called.
|
|
1225
|
+
class LatencyMetrics
|
|
1226
|
+
include ::Google::Protobuf::MessageExts
|
|
1227
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1228
|
+
end
|
|
1229
|
+
|
|
1230
|
+
# Latency metrics for a single tool.
|
|
1231
|
+
# @!attribute [r] tool
|
|
1232
|
+
# @return [::String]
|
|
1233
|
+
# Output only. Format:
|
|
1234
|
+
# `projects/{project}/locations/{location}/apps/{app}/tools/{tool}`.
|
|
1235
|
+
#
|
|
1236
|
+
# Note: The following fields are mutually exclusive: `tool`, `toolset_tool`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
1237
|
+
# @!attribute [r] toolset_tool
|
|
1238
|
+
# @return [::Google::Cloud::Ces::V1beta::ToolsetTool]
|
|
1239
|
+
# Output only. The toolset tool identifier.
|
|
1240
|
+
#
|
|
1241
|
+
# Note: The following fields are mutually exclusive: `toolset_tool`, `tool`. If a field in that set is populated, all other fields in the set will automatically be cleared.
|
|
1242
|
+
# @!attribute [r] tool_display_name
|
|
1243
|
+
# @return [::String]
|
|
1244
|
+
# Output only. The display name of the tool.
|
|
1245
|
+
# @!attribute [r] latency_metrics
|
|
1246
|
+
# @return [::Google::Cloud::Ces::V1beta::LatencyReport::LatencyMetrics]
|
|
1247
|
+
# Output only. The latency metrics for the tool.
|
|
1248
|
+
class ToolLatency
|
|
1249
|
+
include ::Google::Protobuf::MessageExts
|
|
1250
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1251
|
+
end
|
|
1252
|
+
|
|
1253
|
+
# Latency metrics for a single callback.
|
|
1254
|
+
# @!attribute [r] stage
|
|
1255
|
+
# @return [::String]
|
|
1256
|
+
# Output only. The stage of the callback.
|
|
1257
|
+
# @!attribute [r] latency_metrics
|
|
1258
|
+
# @return [::Google::Cloud::Ces::V1beta::LatencyReport::LatencyMetrics]
|
|
1259
|
+
# Output only. The latency metrics for the callback.
|
|
1260
|
+
class CallbackLatency
|
|
1261
|
+
include ::Google::Protobuf::MessageExts
|
|
1262
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1263
|
+
end
|
|
1264
|
+
|
|
1265
|
+
# Latency metrics for a single guardrail.
|
|
1266
|
+
# @!attribute [r] guardrail
|
|
1267
|
+
# @return [::String]
|
|
1268
|
+
# Output only. The name of the guardrail.
|
|
1269
|
+
# Format:
|
|
1270
|
+
# `projects/{project}/locations/{location}/apps/{app}/guardrails/{guardrail}`.
|
|
1271
|
+
# @!attribute [r] guardrail_display_name
|
|
1272
|
+
# @return [::String]
|
|
1273
|
+
# Output only. The display name of the guardrail.
|
|
1274
|
+
# @!attribute [r] latency_metrics
|
|
1275
|
+
# @return [::Google::Cloud::Ces::V1beta::LatencyReport::LatencyMetrics]
|
|
1276
|
+
# Output only. The latency metrics for the guardrail.
|
|
1277
|
+
class GuardrailLatency
|
|
1278
|
+
include ::Google::Protobuf::MessageExts
|
|
1279
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1280
|
+
end
|
|
1281
|
+
|
|
1282
|
+
# Latency metrics for a single LLM call.
|
|
1283
|
+
# @!attribute [r] model
|
|
1284
|
+
# @return [::String]
|
|
1285
|
+
# Output only. The name of the model.
|
|
1286
|
+
# @!attribute [r] latency_metrics
|
|
1287
|
+
# @return [::Google::Cloud::Ces::V1beta::LatencyReport::LatencyMetrics]
|
|
1288
|
+
# Output only. The latency metrics for the LLM call.
|
|
1289
|
+
class LlmCallLatency
|
|
1290
|
+
include ::Google::Protobuf::MessageExts
|
|
1291
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1292
|
+
end
|
|
1293
|
+
end
|
|
1294
|
+
|
|
1295
|
+
# An evaluation expectation represents a specific criteria to evaluate against.
|
|
1296
|
+
# @!attribute [rw] llm_criteria
|
|
1297
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationExpectation::LlmCriteria]
|
|
1298
|
+
# Optional. Evaluation criteria based on an LLM prompt.
|
|
1299
|
+
# @!attribute [rw] name
|
|
1300
|
+
# @return [::String]
|
|
1301
|
+
# Identifier. The unique identifier of this evaluation expectation.
|
|
1302
|
+
# Format:
|
|
1303
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationExpectations/{evaluation_expectation}`
|
|
1304
|
+
# @!attribute [rw] display_name
|
|
1305
|
+
# @return [::String]
|
|
1306
|
+
# Required. User-defined display name. Must be unique within the app.
|
|
1307
|
+
# @!attribute [rw] tags
|
|
1308
|
+
# @return [::Array<::String>]
|
|
1309
|
+
# Optional. User-defined tags for expectations. Can be used to filter
|
|
1310
|
+
# expectations.
|
|
1311
|
+
# @!attribute [r] create_time
|
|
1312
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1313
|
+
# Output only. Timestamp when the evaluation expectation was created.
|
|
1314
|
+
# @!attribute [r] update_time
|
|
1315
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1316
|
+
# Output only. Timestamp when the evaluation expectation was last updated.
|
|
1317
|
+
# @!attribute [r] etag
|
|
1318
|
+
# @return [::String]
|
|
1319
|
+
# Output only. Etag used to ensure the object hasn't changed during a
|
|
1320
|
+
# read-modify-write operation. If the etag is empty, the update will
|
|
1321
|
+
# overwrite any concurrent changes.
|
|
1322
|
+
class EvaluationExpectation
|
|
1323
|
+
include ::Google::Protobuf::MessageExts
|
|
1324
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1325
|
+
|
|
1326
|
+
# Configuration for LLM-based evaluation criteria.
|
|
1327
|
+
# @!attribute [rw] prompt
|
|
1328
|
+
# @return [::String]
|
|
1329
|
+
# Required. The prompt/instructions provided to the LLM judge.
|
|
1330
|
+
class LlmCriteria
|
|
1331
|
+
include ::Google::Protobuf::MessageExts
|
|
1332
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1333
|
+
end
|
|
1334
|
+
end
|
|
1335
|
+
|
|
1336
|
+
# EvaluationConfig configures settings for running the evaluation.
|
|
1337
|
+
# @!attribute [rw] input_audio_config
|
|
1338
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
1339
|
+
# @return [::Google::Cloud::Ces::V1beta::InputAudioConfig]
|
|
1340
|
+
# Optional. Configuration for processing the input audio.
|
|
1341
|
+
# @!attribute [rw] output_audio_config
|
|
1342
|
+
# @deprecated This field is deprecated and may be removed in the next major version update.
|
|
1343
|
+
# @return [::Google::Cloud::Ces::V1beta::OutputAudioConfig]
|
|
1344
|
+
# Optional. Configuration for generating the output audio.
|
|
1345
|
+
# @!attribute [rw] evaluation_channel
|
|
1346
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationConfig::EvaluationChannel]
|
|
1347
|
+
# Optional. The channel to evaluate.
|
|
1348
|
+
# @!attribute [rw] tool_call_behaviour
|
|
1349
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationToolCallBehaviour]
|
|
1350
|
+
# Optional. Specifies whether the evaluation should use real tool calls or
|
|
1351
|
+
# fake tools.
|
|
1352
|
+
class EvaluationConfig
|
|
1353
|
+
include ::Google::Protobuf::MessageExts
|
|
1354
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1355
|
+
|
|
1356
|
+
# The channel to evaluate.
|
|
1357
|
+
module EvaluationChannel
|
|
1358
|
+
# Unspecified evaluation channel.
|
|
1359
|
+
EVALUATION_CHANNEL_UNSPECIFIED = 0
|
|
1360
|
+
|
|
1361
|
+
# Text-only evaluation channel.
|
|
1362
|
+
TEXT = 1
|
|
1363
|
+
|
|
1364
|
+
# Audio evaluation channel.
|
|
1365
|
+
AUDIO = 2
|
|
1366
|
+
end
|
|
1367
|
+
end
|
|
1368
|
+
|
|
1369
|
+
# Information about an error encountered during an evaluation execution.
|
|
1370
|
+
# @!attribute [r] error_type
|
|
1371
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationErrorInfo::ErrorType]
|
|
1372
|
+
# Output only. The type of error.
|
|
1373
|
+
# @!attribute [r] error_message
|
|
1374
|
+
# @return [::String]
|
|
1375
|
+
# Output only. The error message.
|
|
1376
|
+
# @!attribute [r] session_id
|
|
1377
|
+
# @return [::String]
|
|
1378
|
+
# Output only. The session ID for the conversation that caused the error.
|
|
1379
|
+
class EvaluationErrorInfo
|
|
1380
|
+
include ::Google::Protobuf::MessageExts
|
|
1381
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1382
|
+
|
|
1383
|
+
# The type of error
|
|
1384
|
+
module ErrorType
|
|
1385
|
+
# Unspecified error type.
|
|
1386
|
+
ERROR_TYPE_UNSPECIFIED = 0
|
|
1387
|
+
|
|
1388
|
+
# Failure during runtime execution.
|
|
1389
|
+
RUNTIME_FAILURE = 1
|
|
1390
|
+
|
|
1391
|
+
# Failure to retrieve conversation from CES Runtime.
|
|
1392
|
+
CONVERSATION_RETRIEVAL_FAILURE = 2
|
|
1393
|
+
|
|
1394
|
+
# Failure to calculate a metric / outcome.
|
|
1395
|
+
METRIC_CALCULATION_FAILURE = 3
|
|
1396
|
+
|
|
1397
|
+
# Failure to update the evaluation.
|
|
1398
|
+
EVALUATION_UPDATE_FAILURE = 4
|
|
1399
|
+
|
|
1400
|
+
# Ran out of quota.
|
|
1401
|
+
QUOTA_EXHAUSTED = 5
|
|
1402
|
+
|
|
1403
|
+
# Failure during user simulation.
|
|
1404
|
+
USER_SIMULATION_FAILURE = 6
|
|
1405
|
+
end
|
|
1406
|
+
end
|
|
1407
|
+
|
|
1408
|
+
# Request message for
|
|
1409
|
+
# {::Google::Cloud::Ces::V1beta::EvaluationService::Client#run_evaluation EvaluationService.RunEvaluation}.
|
|
1410
|
+
# @!attribute [rw] app
|
|
1411
|
+
# @return [::String]
|
|
1412
|
+
# Required. The app to evaluate.
|
|
1413
|
+
# Format:
|
|
1414
|
+
# `projects/{project}/locations/{location}/apps/{app}`
|
|
1415
|
+
# @!attribute [rw] evaluations
|
|
1416
|
+
# @return [::Array<::String>]
|
|
1417
|
+
# Optional. List of evaluations to run.
|
|
1418
|
+
# Format:
|
|
1419
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluations/{evaluation}`
|
|
1420
|
+
# @!attribute [rw] evaluation_dataset
|
|
1421
|
+
# @return [::String]
|
|
1422
|
+
# Optional. An evaluation dataset to run.
|
|
1423
|
+
# Format:
|
|
1424
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationDatasets/{evaluationDataset}`
|
|
1425
|
+
# @!attribute [rw] display_name
|
|
1426
|
+
# @return [::String]
|
|
1427
|
+
# Optional. The display name of the evaluation run.
|
|
1428
|
+
# @!attribute [rw] app_version
|
|
1429
|
+
# @return [::String]
|
|
1430
|
+
# Optional. The app version to evaluate.
|
|
1431
|
+
# Format:
|
|
1432
|
+
# `projects/{project}/locations/{location}/apps/{app}/versions/{version}`
|
|
1433
|
+
# @!attribute [rw] config
|
|
1434
|
+
# @return [::Google::Cloud::Ces::V1beta::EvaluationConfig]
|
|
1435
|
+
# Optional. The configuration to use for the run.
|
|
1436
|
+
# @!attribute [rw] run_count
|
|
1437
|
+
# @return [::Integer]
|
|
1438
|
+
# Optional. The number of times to run the evaluation. If not set, the
|
|
1439
|
+
# default value is 1 per golden, and 5 per scenario.
|
|
1440
|
+
# @!attribute [rw] persona_run_configs
|
|
1441
|
+
# @return [::Array<::Google::Cloud::Ces::V1beta::PersonaRunConfig>]
|
|
1442
|
+
# Optional. The configuration to use for the run per persona.
|
|
1443
|
+
# @!attribute [rw] optimization_config
|
|
1444
|
+
# @return [::Google::Cloud::Ces::V1beta::OptimizationConfig]
|
|
1445
|
+
# Optional. Configuration for running the optimization step after the
|
|
1446
|
+
# evaluation run. If not set, the optimization step will not be run.
|
|
1447
|
+
# @!attribute [rw] scheduled_evaluation_run
|
|
1448
|
+
# @return [::String]
|
|
1449
|
+
# Optional. The resource name of the `ScheduledEvaluationRun` that is
|
|
1450
|
+
# triggering this evaluation run.
|
|
1451
|
+
#
|
|
1452
|
+
# If this field is set, the `scheduled_evaluation_run` field on the created
|
|
1453
|
+
# `EvaluationRun` resource will be populated from this value.
|
|
1454
|
+
# Format:
|
|
1455
|
+
# `projects/{project}/locations/{location}/apps/{app}/scheduledEvaluationRuns/{scheduled_evaluation_run}`
|
|
1456
|
+
# @!attribute [rw] golden_run_method
|
|
1457
|
+
# @return [::Google::Cloud::Ces::V1beta::GoldenRunMethod]
|
|
1458
|
+
# Optional. The method to run the evaluation if it is a golden evaluation. If
|
|
1459
|
+
# not set, default to STABLE.
|
|
1460
|
+
# @!attribute [rw] generate_latency_report
|
|
1461
|
+
# @return [::Boolean]
|
|
1462
|
+
# Optional. Whether to generate a latency report for the evaluation run.
|
|
1463
|
+
class RunEvaluationRequest
|
|
1464
|
+
include ::Google::Protobuf::MessageExts
|
|
1465
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1466
|
+
end
|
|
1467
|
+
|
|
1468
|
+
# Represents a scheduled evaluation run configuration.
|
|
1469
|
+
# @!attribute [rw] name
|
|
1470
|
+
# @return [::String]
|
|
1471
|
+
# Identifier. The unique identifier of the scheduled evaluation run config.
|
|
1472
|
+
# Format:
|
|
1473
|
+
# projects/\\{projectId}/locations/\\{locationId}/apps/\\{appId}/scheduledEvaluationRuns/\\{scheduledEvaluationRunId}
|
|
1474
|
+
# @!attribute [rw] display_name
|
|
1475
|
+
# @return [::String]
|
|
1476
|
+
# Required. User-defined display name of the scheduled evaluation run config.
|
|
1477
|
+
# @!attribute [rw] request
|
|
1478
|
+
# @return [::Google::Cloud::Ces::V1beta::RunEvaluationRequest]
|
|
1479
|
+
# Required. The RunEvaluationRequest to schedule
|
|
1480
|
+
# @!attribute [rw] description
|
|
1481
|
+
# @return [::String]
|
|
1482
|
+
# Optional. User-defined description of the scheduled evaluation run.
|
|
1483
|
+
# @!attribute [rw] scheduling_config
|
|
1484
|
+
# @return [::Google::Cloud::Ces::V1beta::ScheduledEvaluationRun::SchedulingConfig]
|
|
1485
|
+
# Required. Configuration for the timing and frequency with which to execute
|
|
1486
|
+
# the evaluations.
|
|
1487
|
+
# @!attribute [rw] active
|
|
1488
|
+
# @return [::Boolean]
|
|
1489
|
+
# Optional. Whether this config is active
|
|
1490
|
+
# @!attribute [r] last_completed_run
|
|
1491
|
+
# @return [::String]
|
|
1492
|
+
# Output only. The last successful EvaluationRun of this scheduled execution.
|
|
1493
|
+
# Format:
|
|
1494
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationRuns/{evaluationRun}`
|
|
1495
|
+
# @!attribute [r] total_executions
|
|
1496
|
+
# @return [::Integer]
|
|
1497
|
+
# Output only. The total number of times this run has been executed
|
|
1498
|
+
# @!attribute [r] next_scheduled_execution_time
|
|
1499
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1500
|
+
# Output only. The next time this is scheduled to execute
|
|
1501
|
+
# @!attribute [r] create_time
|
|
1502
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1503
|
+
# Output only. Timestamp when the scheduled evaluation run was created.
|
|
1504
|
+
# @!attribute [r] created_by
|
|
1505
|
+
# @return [::String]
|
|
1506
|
+
# Output only. The user who created the scheduled evaluation run.
|
|
1507
|
+
# @!attribute [r] update_time
|
|
1508
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1509
|
+
# Output only. Timestamp when the evaluation was last updated.
|
|
1510
|
+
# @!attribute [r] last_updated_by
|
|
1511
|
+
# @return [::String]
|
|
1512
|
+
# Output only. The user who last updated the evaluation.
|
|
1513
|
+
# @!attribute [r] etag
|
|
1514
|
+
# @return [::String]
|
|
1515
|
+
# Output only. Etag used to ensure the object hasn't changed during a
|
|
1516
|
+
# read-modify-write operation. If the etag is empty, the update will
|
|
1517
|
+
# overwrite any concurrent changes.
|
|
1518
|
+
class ScheduledEvaluationRun
|
|
1519
|
+
include ::Google::Protobuf::MessageExts
|
|
1520
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1521
|
+
|
|
1522
|
+
# Eval scheduling configuration details
|
|
1523
|
+
# @!attribute [rw] frequency
|
|
1524
|
+
# @return [::Google::Cloud::Ces::V1beta::ScheduledEvaluationRun::SchedulingConfig::Frequency]
|
|
1525
|
+
# Required. The frequency with which to run the eval
|
|
1526
|
+
# @!attribute [rw] start_time
|
|
1527
|
+
# @return [::Google::Protobuf::Timestamp]
|
|
1528
|
+
# Required. Timestamp when the eval should start.
|
|
1529
|
+
# @!attribute [rw] days_of_week
|
|
1530
|
+
# @return [::Array<::Integer>]
|
|
1531
|
+
# Optional. The days of the week to run the eval. Applicable only for
|
|
1532
|
+
# Weekly and Biweekly frequencies. 1 is Monday, 2 is Tuesday, ..., 7 is
|
|
1533
|
+
# Sunday.
|
|
1534
|
+
class SchedulingConfig
|
|
1535
|
+
include ::Google::Protobuf::MessageExts
|
|
1536
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1537
|
+
|
|
1538
|
+
# The frequencies evals can be run at
|
|
1539
|
+
module Frequency
|
|
1540
|
+
# The frequency is unspecified.
|
|
1541
|
+
FREQUENCY_UNSPECIFIED = 0
|
|
1542
|
+
|
|
1543
|
+
# Indicates a one-time scheduled run that should not repeat
|
|
1544
|
+
NONE = 1
|
|
1545
|
+
|
|
1546
|
+
# Run the evaluation daily.
|
|
1547
|
+
DAILY = 2
|
|
1548
|
+
|
|
1549
|
+
# Run the evaluation weekly.
|
|
1550
|
+
WEEKLY = 3
|
|
1551
|
+
|
|
1552
|
+
# Run the evaluation biweekly.
|
|
1553
|
+
BIWEEKLY = 4
|
|
1554
|
+
end
|
|
1555
|
+
end
|
|
1556
|
+
end
|
|
1557
|
+
|
|
1558
|
+
# Configuration for running an evaluation for a specific persona.
|
|
1559
|
+
# @!attribute [rw] persona
|
|
1560
|
+
# @return [::String]
|
|
1561
|
+
# Optional. The persona to use for the evaluation.
|
|
1562
|
+
# Format:
|
|
1563
|
+
# `projects/{project}/locations/{location}/apps/{app}/evaluationPersonas/{evaluationPersona}`
|
|
1564
|
+
# @!attribute [rw] task_count
|
|
1565
|
+
# @return [::Integer]
|
|
1566
|
+
# Optional. The number of tasks to run for the persona.
|
|
1567
|
+
class PersonaRunConfig
|
|
1568
|
+
include ::Google::Protobuf::MessageExts
|
|
1569
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1570
|
+
end
|
|
1571
|
+
|
|
1572
|
+
# Configuration for running the optimization step after the evaluation run.
|
|
1573
|
+
# @!attribute [rw] generate_loss_report
|
|
1574
|
+
# @return [::Boolean]
|
|
1575
|
+
# Optional. Whether to generate a loss report.
|
|
1576
|
+
# @!attribute [r] assistant_session
|
|
1577
|
+
# @return [::String]
|
|
1578
|
+
# Output only. The assistant session to use for the optimization based on
|
|
1579
|
+
# this evaluation run. Format:
|
|
1580
|
+
# `projects/{project}/locations/{location}/apps/{app}/assistantSessions/{assistantSession}`
|
|
1581
|
+
# @!attribute [r] report_summary
|
|
1582
|
+
# @return [::String]
|
|
1583
|
+
# Output only. The summary of the loss report.
|
|
1584
|
+
# @!attribute [r] should_suggest_fix
|
|
1585
|
+
# @return [::Boolean]
|
|
1586
|
+
# Output only. Whether to suggest a fix for the losses.
|
|
1587
|
+
# @!attribute [r] status
|
|
1588
|
+
# @return [::Google::Cloud::Ces::V1beta::OptimizationConfig::OptimizationStatus]
|
|
1589
|
+
# Output only. The status of the optimization run.
|
|
1590
|
+
# @!attribute [r] error_message
|
|
1591
|
+
# @return [::String]
|
|
1592
|
+
# Output only. The error message if the optimization run failed.
|
|
1593
|
+
# @!attribute [r] loss_report
|
|
1594
|
+
# @return [::Google::Protobuf::Struct]
|
|
1595
|
+
# Output only. The generated loss report.
|
|
1596
|
+
class OptimizationConfig
|
|
1597
|
+
include ::Google::Protobuf::MessageExts
|
|
1598
|
+
extend ::Google::Protobuf::MessageExts::ClassMethods
|
|
1599
|
+
|
|
1600
|
+
# The status of the optimization run.
|
|
1601
|
+
module OptimizationStatus
|
|
1602
|
+
# Optimization status is not specified.
|
|
1603
|
+
OPTIMIZATION_STATUS_UNSPECIFIED = 0
|
|
1604
|
+
|
|
1605
|
+
# Optimization is running.
|
|
1606
|
+
RUNNING = 1
|
|
1607
|
+
|
|
1608
|
+
# Optimization has completed.
|
|
1609
|
+
COMPLETED = 2
|
|
1610
|
+
|
|
1611
|
+
# Optimization failed due to an internal error.
|
|
1612
|
+
ERROR = 3
|
|
1613
|
+
end
|
|
1614
|
+
end
|
|
1615
|
+
end
|
|
1616
|
+
end
|
|
1617
|
+
end
|
|
1618
|
+
end
|