1<html><body> 2<style> 3 4body, h1, h2, h3, div, span, p, pre, a { 5 margin: 0; 6 padding: 0; 7 border: 0; 8 font-weight: inherit; 9 font-style: inherit; 10 font-size: 100%; 11 font-family: inherit; 12 vertical-align: baseline; 13} 14 15body { 16 font-size: 13px; 17 padding: 1em; 18} 19 20h1 { 21 font-size: 26px; 22 margin-bottom: 1em; 23} 24 25h2 { 26 font-size: 24px; 27 margin-bottom: 1em; 28} 29 30h3 { 31 font-size: 20px; 32 margin-bottom: 1em; 33 margin-top: 1em; 34} 35 36pre, code { 37 line-height: 1.5; 38 font-family: Monaco, 'DejaVu Sans Mono', 'Bitstream Vera Sans Mono', 'Lucida Console', monospace; 39} 40 41pre { 42 margin-top: 0.5em; 43} 44 45h1, h2, h3, p { 46 font-family: Arial, sans serif; 47} 48 49h1, h2, h3 { 50 border-bottom: solid #CCC 1px; 51} 52 53.toc_element { 54 margin-top: 0.5em; 55} 56 57.firstline { 58 margin-left: 2 em; 59} 60 61.method { 62 margin-top: 1em; 63 border: solid 1px #CCC; 64 padding: 1em; 65 background: #EEE; 66} 67 68.details { 69 font-weight: bold; 70 font-size: 14px; 71} 72 73</style> 74 75<h1><a href="ml_v1.html">Google Cloud Machine Learning Engine</a> . <a href="ml_v1.projects.html">projects</a> . <a href="ml_v1.projects.models.html">models</a> . <a href="ml_v1.projects.models.versions.html">versions</a></h1> 76<h2>Instance Methods</h2> 77<p class="toc_element"> 78 <code><a href="#create">create(parent, body, x__xgafv=None)</a></code></p> 79<p class="firstline">Creates a new version of a model from a trained TensorFlow model.</p> 80<p class="toc_element"> 81 <code><a href="#delete">delete(name, x__xgafv=None)</a></code></p> 82<p class="firstline">Deletes a model version.</p> 83<p class="toc_element"> 84 <code><a href="#get">get(name, x__xgafv=None)</a></code></p> 85<p class="firstline">Gets information about a model version.</p> 86<p class="toc_element"> 87 <code><a href="#list">list(parent, pageToken=None, x__xgafv=None, pageSize=None)</a></code></p> 88<p class="firstline">Gets basic information about all the versions of a model.</p> 89<p class="toc_element"> 90 <code><a href="#list_next">list_next(previous_request, previous_response)</a></code></p> 91<p class="firstline">Retrieves the next page of results.</p> 92<p class="toc_element"> 93 <code><a href="#setDefault">setDefault(name, body, x__xgafv=None)</a></code></p> 94<p class="firstline">Designates a version to be the default for the model.</p> 95<h3>Method Details</h3> 96<div class="method"> 97 <code class="details" id="create">create(parent, body, x__xgafv=None)</code> 98 <pre>Creates a new version of a model from a trained TensorFlow model. 99 100If the version created in the cloud by this call is the first deployed 101version of the specified model, it will be made the default version of the 102model. When you add a version to a model that already has one or more 103versions, the default version does not automatically change. If you want a 104new version to be the default, you must call 105[projects.models.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 106 107Args: 108 parent: string, Required. The name of the model. 109 110Authorization: requires `Editor` role on the parent project. (required) 111 body: object, The request body. (required) 112 The object takes the form of: 113 114{ # Represents a version of the model. 115 # 116 # Each version is a trained model deployed in the cloud, ready to handle 117 # prediction requests. A model can have multiple versions. You can get 118 # information about all of the versions of a given model by calling 119 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 120 "description": "A String", # Optional. The description specified for the version when it was created. 121 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 122 # If not set, Google Cloud ML will choose a version. 123 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 124 # model. You should generally use `automatic_scaling` with an appropriate 125 # `min_nodes` instead, but this option is available if you want more 126 # predictable billing. Beware that latency and error rates will increase 127 # if the traffic exceeds that capability of the system to serve it based 128 # on the selected number of nodes. 129 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 130 # starting from the time the model is deployed, so the cost of operating 131 # this model will be proportional to `nodes` * number of hours since 132 # last billing cycle plus the cost for each prediction performed. 133 }, 134 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 135 # create the version. See the 136 # [overview of model 137 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 138 # informaiton. 139 # 140 # When passing Version to 141 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 142 # the model service uses the specified location as the source of the model. 143 # Once deployed, the model version is hosted by the prediction service, so 144 # this location is useful only as a historical record. 145 # The total number of model files can't exceed 1000. 146 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 147 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 148 # response to increases and decreases in traffic. Care should be 149 # taken to ramp up traffic according to the model's ability to scale 150 # or you will start seeing increases in latency and 429 response codes. 151 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 152 # nodes are always up, starting from the time the model is deployed, so the 153 # cost of operating this model will be at least 154 # `rate` * `min_nodes` * number of hours since last billing cycle, 155 # where `rate` is the cost per node-hour as documented in 156 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 157 # even if no predictions are performed. There is additional cost for each 158 # prediction performed. 159 # 160 # Unlike manual scaling, if the load gets too heavy for the nodes 161 # that are up, the service will automatically add nodes to handle the 162 # increased load as well as scale back as traffic drops, always maintaining 163 # at least `min_nodes`. You will be charged for the time in which additional 164 # nodes are used. 165 # 166 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 167 # to a model stops (and after a cool-down period), nodes will be shut down 168 # and no charges will be incurred until traffic to the model resumes. 169 }, 170 "createTime": "A String", # Output only. The time the version was created. 171 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 172 # requests that do not specify a version. 173 # 174 # You can change the default version by calling 175 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 176 "name": "A String", # Required.The name specified for the version when it was created. 177 # 178 # The version name must be unique within the model it is created in. 179} 180 181 x__xgafv: string, V1 error format. 182 Allowed values 183 1 - v1 error format 184 2 - v2 error format 185 186Returns: 187 An object of the form: 188 189 { # This resource represents a long-running operation that is the result of a 190 # network API call. 191 "metadata": { # Service-specific metadata associated with the operation. It typically 192 # contains progress information and common metadata such as create time. 193 # Some services might not provide such metadata. Any method that returns a 194 # long-running operation should document the metadata type, if any. 195 "a_key": "", # Properties of the object. Contains field @type with type URL. 196 }, 197 "error": { # The `Status` type defines a logical error model that is suitable for different # The error result of the operation in case of failure or cancellation. 198 # programming environments, including REST APIs and RPC APIs. It is used by 199 # [gRPC](https://github.com/grpc). The error model is designed to be: 200 # 201 # - Simple to use and understand for most users 202 # - Flexible enough to meet unexpected needs 203 # 204 # # Overview 205 # 206 # The `Status` message contains three pieces of data: error code, error message, 207 # and error details. The error code should be an enum value of 208 # google.rpc.Code, but it may accept additional error codes if needed. The 209 # error message should be a developer-facing English message that helps 210 # developers *understand* and *resolve* the error. If a localized user-facing 211 # error message is needed, put the localized message in the error details or 212 # localize it in the client. The optional error details may contain arbitrary 213 # information about the error. There is a predefined set of error detail types 214 # in the package `google.rpc` that can be used for common error conditions. 215 # 216 # # Language mapping 217 # 218 # The `Status` message is the logical representation of the error model, but it 219 # is not necessarily the actual wire format. When the `Status` message is 220 # exposed in different client libraries and different wire protocols, it can be 221 # mapped differently. For example, it will likely be mapped to some exceptions 222 # in Java, but more likely mapped to some error codes in C. 223 # 224 # # Other uses 225 # 226 # The error model and the `Status` message can be used in a variety of 227 # environments, either with or without APIs, to provide a 228 # consistent developer experience across different environments. 229 # 230 # Example uses of this error model include: 231 # 232 # - Partial errors. If a service needs to return partial errors to the client, 233 # it may embed the `Status` in the normal response to indicate the partial 234 # errors. 235 # 236 # - Workflow errors. A typical workflow has multiple steps. Each step may 237 # have a `Status` message for error reporting. 238 # 239 # - Batch operations. If a client uses batch request and batch response, the 240 # `Status` message should be used directly inside batch response, one for 241 # each error sub-response. 242 # 243 # - Asynchronous operations. If an API call embeds asynchronous operation 244 # results in its response, the status of those operations should be 245 # represented directly using the `Status` message. 246 # 247 # - Logging. If some API errors are stored in logs, the message `Status` could 248 # be used directly after any stripping needed for security/privacy reasons. 249 "message": "A String", # A developer-facing error message, which should be in English. Any 250 # user-facing error message should be localized and sent in the 251 # google.rpc.Status.details field, or localized by the client. 252 "code": 42, # The status code, which should be an enum value of google.rpc.Code. 253 "details": [ # A list of messages that carry the error details. There will be a 254 # common set of message types for APIs to use. 255 { 256 "a_key": "", # Properties of the object. Contains field @type with type URL. 257 }, 258 ], 259 }, 260 "done": True or False, # If the value is `false`, it means the operation is still in progress. 261 # If true, the operation is completed, and either `error` or `response` is 262 # available. 263 "response": { # The normal response of the operation in case of success. If the original 264 # method returns no data on success, such as `Delete`, the response is 265 # `google.protobuf.Empty`. If the original method is standard 266 # `Get`/`Create`/`Update`, the response should be the resource. For other 267 # methods, the response should have the type `XxxResponse`, where `Xxx` 268 # is the original method name. For example, if the original method name 269 # is `TakeSnapshot()`, the inferred response type is 270 # `TakeSnapshotResponse`. 271 "a_key": "", # Properties of the object. Contains field @type with type URL. 272 }, 273 "name": "A String", # The server-assigned name, which is only unique within the same service that 274 # originally returns it. If you use the default HTTP mapping, the 275 # `name` should have the format of `operations/some/unique/name`. 276 }</pre> 277</div> 278 279<div class="method"> 280 <code class="details" id="delete">delete(name, x__xgafv=None)</code> 281 <pre>Deletes a model version. 282 283Each model can have multiple versions deployed and in use at any given 284time. Use this method to remove a single version. 285 286Note: You cannot delete the version that is set as the default version 287of the model unless it is the only remaining version. 288 289Args: 290 name: string, Required. The name of the version. You can get the names of all the 291versions of a model by calling 292[projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 293 294Authorization: requires `Editor` role on the parent project. (required) 295 x__xgafv: string, V1 error format. 296 Allowed values 297 1 - v1 error format 298 2 - v2 error format 299 300Returns: 301 An object of the form: 302 303 { # This resource represents a long-running operation that is the result of a 304 # network API call. 305 "metadata": { # Service-specific metadata associated with the operation. It typically 306 # contains progress information and common metadata such as create time. 307 # Some services might not provide such metadata. Any method that returns a 308 # long-running operation should document the metadata type, if any. 309 "a_key": "", # Properties of the object. Contains field @type with type URL. 310 }, 311 "error": { # The `Status` type defines a logical error model that is suitable for different # The error result of the operation in case of failure or cancellation. 312 # programming environments, including REST APIs and RPC APIs. It is used by 313 # [gRPC](https://github.com/grpc). The error model is designed to be: 314 # 315 # - Simple to use and understand for most users 316 # - Flexible enough to meet unexpected needs 317 # 318 # # Overview 319 # 320 # The `Status` message contains three pieces of data: error code, error message, 321 # and error details. The error code should be an enum value of 322 # google.rpc.Code, but it may accept additional error codes if needed. The 323 # error message should be a developer-facing English message that helps 324 # developers *understand* and *resolve* the error. If a localized user-facing 325 # error message is needed, put the localized message in the error details or 326 # localize it in the client. The optional error details may contain arbitrary 327 # information about the error. There is a predefined set of error detail types 328 # in the package `google.rpc` that can be used for common error conditions. 329 # 330 # # Language mapping 331 # 332 # The `Status` message is the logical representation of the error model, but it 333 # is not necessarily the actual wire format. When the `Status` message is 334 # exposed in different client libraries and different wire protocols, it can be 335 # mapped differently. For example, it will likely be mapped to some exceptions 336 # in Java, but more likely mapped to some error codes in C. 337 # 338 # # Other uses 339 # 340 # The error model and the `Status` message can be used in a variety of 341 # environments, either with or without APIs, to provide a 342 # consistent developer experience across different environments. 343 # 344 # Example uses of this error model include: 345 # 346 # - Partial errors. If a service needs to return partial errors to the client, 347 # it may embed the `Status` in the normal response to indicate the partial 348 # errors. 349 # 350 # - Workflow errors. A typical workflow has multiple steps. Each step may 351 # have a `Status` message for error reporting. 352 # 353 # - Batch operations. If a client uses batch request and batch response, the 354 # `Status` message should be used directly inside batch response, one for 355 # each error sub-response. 356 # 357 # - Asynchronous operations. If an API call embeds asynchronous operation 358 # results in its response, the status of those operations should be 359 # represented directly using the `Status` message. 360 # 361 # - Logging. If some API errors are stored in logs, the message `Status` could 362 # be used directly after any stripping needed for security/privacy reasons. 363 "message": "A String", # A developer-facing error message, which should be in English. Any 364 # user-facing error message should be localized and sent in the 365 # google.rpc.Status.details field, or localized by the client. 366 "code": 42, # The status code, which should be an enum value of google.rpc.Code. 367 "details": [ # A list of messages that carry the error details. There will be a 368 # common set of message types for APIs to use. 369 { 370 "a_key": "", # Properties of the object. Contains field @type with type URL. 371 }, 372 ], 373 }, 374 "done": True or False, # If the value is `false`, it means the operation is still in progress. 375 # If true, the operation is completed, and either `error` or `response` is 376 # available. 377 "response": { # The normal response of the operation in case of success. If the original 378 # method returns no data on success, such as `Delete`, the response is 379 # `google.protobuf.Empty`. If the original method is standard 380 # `Get`/`Create`/`Update`, the response should be the resource. For other 381 # methods, the response should have the type `XxxResponse`, where `Xxx` 382 # is the original method name. For example, if the original method name 383 # is `TakeSnapshot()`, the inferred response type is 384 # `TakeSnapshotResponse`. 385 "a_key": "", # Properties of the object. Contains field @type with type URL. 386 }, 387 "name": "A String", # The server-assigned name, which is only unique within the same service that 388 # originally returns it. If you use the default HTTP mapping, the 389 # `name` should have the format of `operations/some/unique/name`. 390 }</pre> 391</div> 392 393<div class="method"> 394 <code class="details" id="get">get(name, x__xgafv=None)</code> 395 <pre>Gets information about a model version. 396 397Models can have multiple versions. You can call 398[projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list) 399to get the same information that this method returns for all of the 400versions of a model. 401 402Args: 403 name: string, Required. The name of the version. 404 405Authorization: requires `Viewer` role on the parent project. (required) 406 x__xgafv: string, V1 error format. 407 Allowed values 408 1 - v1 error format 409 2 - v2 error format 410 411Returns: 412 An object of the form: 413 414 { # Represents a version of the model. 415 # 416 # Each version is a trained model deployed in the cloud, ready to handle 417 # prediction requests. A model can have multiple versions. You can get 418 # information about all of the versions of a given model by calling 419 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 420 "description": "A String", # Optional. The description specified for the version when it was created. 421 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 422 # If not set, Google Cloud ML will choose a version. 423 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 424 # model. You should generally use `automatic_scaling` with an appropriate 425 # `min_nodes` instead, but this option is available if you want more 426 # predictable billing. Beware that latency and error rates will increase 427 # if the traffic exceeds that capability of the system to serve it based 428 # on the selected number of nodes. 429 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 430 # starting from the time the model is deployed, so the cost of operating 431 # this model will be proportional to `nodes` * number of hours since 432 # last billing cycle plus the cost for each prediction performed. 433 }, 434 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 435 # create the version. See the 436 # [overview of model 437 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 438 # informaiton. 439 # 440 # When passing Version to 441 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 442 # the model service uses the specified location as the source of the model. 443 # Once deployed, the model version is hosted by the prediction service, so 444 # this location is useful only as a historical record. 445 # The total number of model files can't exceed 1000. 446 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 447 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 448 # response to increases and decreases in traffic. Care should be 449 # taken to ramp up traffic according to the model's ability to scale 450 # or you will start seeing increases in latency and 429 response codes. 451 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 452 # nodes are always up, starting from the time the model is deployed, so the 453 # cost of operating this model will be at least 454 # `rate` * `min_nodes` * number of hours since last billing cycle, 455 # where `rate` is the cost per node-hour as documented in 456 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 457 # even if no predictions are performed. There is additional cost for each 458 # prediction performed. 459 # 460 # Unlike manual scaling, if the load gets too heavy for the nodes 461 # that are up, the service will automatically add nodes to handle the 462 # increased load as well as scale back as traffic drops, always maintaining 463 # at least `min_nodes`. You will be charged for the time in which additional 464 # nodes are used. 465 # 466 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 467 # to a model stops (and after a cool-down period), nodes will be shut down 468 # and no charges will be incurred until traffic to the model resumes. 469 }, 470 "createTime": "A String", # Output only. The time the version was created. 471 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 472 # requests that do not specify a version. 473 # 474 # You can change the default version by calling 475 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 476 "name": "A String", # Required.The name specified for the version when it was created. 477 # 478 # The version name must be unique within the model it is created in. 479 }</pre> 480</div> 481 482<div class="method"> 483 <code class="details" id="list">list(parent, pageToken=None, x__xgafv=None, pageSize=None)</code> 484 <pre>Gets basic information about all the versions of a model. 485 486If you expect that a model has a lot of versions, or if you need to handle 487only a limited number of results at a time, you can request that the list 488be retrieved in batches (called pages): 489 490Args: 491 parent: string, Required. The name of the model for which to list the version. 492 493Authorization: requires `Viewer` role on the parent project. (required) 494 pageToken: string, Optional. A page token to request the next page of results. 495 496You get the token from the `next_page_token` field of the response from 497the previous call. 498 x__xgafv: string, V1 error format. 499 Allowed values 500 1 - v1 error format 501 2 - v2 error format 502 pageSize: integer, Optional. The number of versions to retrieve per "page" of results. If 503there are more remaining results than this number, the response message 504will contain a valid value in the `next_page_token` field. 505 506The default value is 20, and the maximum page size is 100. 507 508Returns: 509 An object of the form: 510 511 { # Response message for the ListVersions method. 512 "nextPageToken": "A String", # Optional. Pass this token as the `page_token` field of the request for a 513 # subsequent call. 514 "versions": [ # The list of versions. 515 { # Represents a version of the model. 516 # 517 # Each version is a trained model deployed in the cloud, ready to handle 518 # prediction requests. A model can have multiple versions. You can get 519 # information about all of the versions of a given model by calling 520 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 521 "description": "A String", # Optional. The description specified for the version when it was created. 522 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 523 # If not set, Google Cloud ML will choose a version. 524 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 525 # model. You should generally use `automatic_scaling` with an appropriate 526 # `min_nodes` instead, but this option is available if you want more 527 # predictable billing. Beware that latency and error rates will increase 528 # if the traffic exceeds that capability of the system to serve it based 529 # on the selected number of nodes. 530 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 531 # starting from the time the model is deployed, so the cost of operating 532 # this model will be proportional to `nodes` * number of hours since 533 # last billing cycle plus the cost for each prediction performed. 534 }, 535 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 536 # create the version. See the 537 # [overview of model 538 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 539 # informaiton. 540 # 541 # When passing Version to 542 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 543 # the model service uses the specified location as the source of the model. 544 # Once deployed, the model version is hosted by the prediction service, so 545 # this location is useful only as a historical record. 546 # The total number of model files can't exceed 1000. 547 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 548 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 549 # response to increases and decreases in traffic. Care should be 550 # taken to ramp up traffic according to the model's ability to scale 551 # or you will start seeing increases in latency and 429 response codes. 552 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 553 # nodes are always up, starting from the time the model is deployed, so the 554 # cost of operating this model will be at least 555 # `rate` * `min_nodes` * number of hours since last billing cycle, 556 # where `rate` is the cost per node-hour as documented in 557 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 558 # even if no predictions are performed. There is additional cost for each 559 # prediction performed. 560 # 561 # Unlike manual scaling, if the load gets too heavy for the nodes 562 # that are up, the service will automatically add nodes to handle the 563 # increased load as well as scale back as traffic drops, always maintaining 564 # at least `min_nodes`. You will be charged for the time in which additional 565 # nodes are used. 566 # 567 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 568 # to a model stops (and after a cool-down period), nodes will be shut down 569 # and no charges will be incurred until traffic to the model resumes. 570 }, 571 "createTime": "A String", # Output only. The time the version was created. 572 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 573 # requests that do not specify a version. 574 # 575 # You can change the default version by calling 576 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 577 "name": "A String", # Required.The name specified for the version when it was created. 578 # 579 # The version name must be unique within the model it is created in. 580 }, 581 ], 582 }</pre> 583</div> 584 585<div class="method"> 586 <code class="details" id="list_next">list_next(previous_request, previous_response)</code> 587 <pre>Retrieves the next page of results. 588 589Args: 590 previous_request: The request for the previous page. (required) 591 previous_response: The response from the request for the previous page. (required) 592 593Returns: 594 A request object that you can call 'execute()' on to request the next 595 page. Returns None if there are no more items in the collection. 596 </pre> 597</div> 598 599<div class="method"> 600 <code class="details" id="setDefault">setDefault(name, body, x__xgafv=None)</code> 601 <pre>Designates a version to be the default for the model. 602 603The default version is used for prediction requests made against the model 604that don't specify a version. 605 606The first version to be created for a model is automatically set as the 607default. You must make any subsequent changes to the default version 608setting manually using this method. 609 610Args: 611 name: string, Required. The name of the version to make the default for the model. You 612can get the names of all the versions of a model by calling 613[projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 614 615Authorization: requires `Editor` role on the parent project. (required) 616 body: object, The request body. (required) 617 The object takes the form of: 618 619{ # Request message for the SetDefaultVersion request. 620 } 621 622 x__xgafv: string, V1 error format. 623 Allowed values 624 1 - v1 error format 625 2 - v2 error format 626 627Returns: 628 An object of the form: 629 630 { # Represents a version of the model. 631 # 632 # Each version is a trained model deployed in the cloud, ready to handle 633 # prediction requests. A model can have multiple versions. You can get 634 # information about all of the versions of a given model by calling 635 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 636 "description": "A String", # Optional. The description specified for the version when it was created. 637 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 638 # If not set, Google Cloud ML will choose a version. 639 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 640 # model. You should generally use `automatic_scaling` with an appropriate 641 # `min_nodes` instead, but this option is available if you want more 642 # predictable billing. Beware that latency and error rates will increase 643 # if the traffic exceeds that capability of the system to serve it based 644 # on the selected number of nodes. 645 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 646 # starting from the time the model is deployed, so the cost of operating 647 # this model will be proportional to `nodes` * number of hours since 648 # last billing cycle plus the cost for each prediction performed. 649 }, 650 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 651 # create the version. See the 652 # [overview of model 653 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 654 # informaiton. 655 # 656 # When passing Version to 657 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 658 # the model service uses the specified location as the source of the model. 659 # Once deployed, the model version is hosted by the prediction service, so 660 # this location is useful only as a historical record. 661 # The total number of model files can't exceed 1000. 662 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 663 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 664 # response to increases and decreases in traffic. Care should be 665 # taken to ramp up traffic according to the model's ability to scale 666 # or you will start seeing increases in latency and 429 response codes. 667 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 668 # nodes are always up, starting from the time the model is deployed, so the 669 # cost of operating this model will be at least 670 # `rate` * `min_nodes` * number of hours since last billing cycle, 671 # where `rate` is the cost per node-hour as documented in 672 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 673 # even if no predictions are performed. There is additional cost for each 674 # prediction performed. 675 # 676 # Unlike manual scaling, if the load gets too heavy for the nodes 677 # that are up, the service will automatically add nodes to handle the 678 # increased load as well as scale back as traffic drops, always maintaining 679 # at least `min_nodes`. You will be charged for the time in which additional 680 # nodes are used. 681 # 682 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 683 # to a model stops (and after a cool-down period), nodes will be shut down 684 # and no charges will be incurred until traffic to the model resumes. 685 }, 686 "createTime": "A String", # Output only. The time the version was created. 687 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 688 # requests that do not specify a version. 689 # 690 # You can change the default version by calling 691 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 692 "name": "A String", # Required.The name specified for the version when it was created. 693 # 694 # The version name must be unique within the model it is created in. 695 }</pre> 696</div> 697 698</body></html>