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></h1> 76<h2>Instance Methods</h2> 77<p class="toc_element"> 78 <code><a href="ml_v1.projects.models.versions.html">versions()</a></code> 79</p> 80<p class="firstline">Returns the versions Resource.</p> 81 82<p class="toc_element"> 83 <code><a href="#create">create(parent, body, x__xgafv=None)</a></code></p> 84<p class="firstline">Creates a model which will later contain one or more versions.</p> 85<p class="toc_element"> 86 <code><a href="#delete">delete(name, x__xgafv=None)</a></code></p> 87<p class="firstline">Deletes a model.</p> 88<p class="toc_element"> 89 <code><a href="#get">get(name, x__xgafv=None)</a></code></p> 90<p class="firstline">Gets information about a model, including its name, the description (if</p> 91<p class="toc_element"> 92 <code><a href="#list">list(parent, pageToken=None, x__xgafv=None, pageSize=None)</a></code></p> 93<p class="firstline">Lists the models in a project.</p> 94<p class="toc_element"> 95 <code><a href="#list_next">list_next(previous_request, previous_response)</a></code></p> 96<p class="firstline">Retrieves the next page of results.</p> 97<h3>Method Details</h3> 98<div class="method"> 99 <code class="details" id="create">create(parent, body, x__xgafv=None)</code> 100 <pre>Creates a model which will later contain one or more versions. 101 102You must add at least one version before you can request predictions from 103the model. Add versions by calling 104[projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create). 105 106Args: 107 parent: string, Required. The project name. 108 109Authorization: requires `Editor` role on the specified project. (required) 110 body: object, The request body. (required) 111 The object takes the form of: 112 113{ # Represents a machine learning solution. 114 # 115 # A model can have multiple versions, each of which is a deployed, trained 116 # model ready to receive prediction requests. The model itself is just a 117 # container. 118 "regions": [ # Optional. The list of regions where the model is going to be deployed. 119 # Currently only one region per model is supported. 120 # Defaults to 'us-central1' if nothing is set. 121 # Note: 122 # * No matter where a model is deployed, it can always be accessed by 123 # users from anywhere, both for online and batch prediction. 124 # * The region for a batch prediction job is set by the region field when 125 # submitting the batch prediction job and does not take its value from 126 # this field. 127 "A String", 128 ], 129 "defaultVersion": { # Represents a version of the model. # Output only. The default version of the model. This version will be used to 130 # handle prediction requests that do not specify a version. 131 # 132 # You can change the default version by calling 133 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 134 # 135 # Each version is a trained model deployed in the cloud, ready to handle 136 # prediction requests. A model can have multiple versions. You can get 137 # information about all of the versions of a given model by calling 138 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 139 "description": "A String", # Optional. The description specified for the version when it was created. 140 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 141 # If not set, Google Cloud ML will choose a version. 142 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 143 # model. You should generally use `automatic_scaling` with an appropriate 144 # `min_nodes` instead, but this option is available if you want more 145 # predictable billing. Beware that latency and error rates will increase 146 # if the traffic exceeds that capability of the system to serve it based 147 # on the selected number of nodes. 148 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 149 # starting from the time the model is deployed, so the cost of operating 150 # this model will be proportional to `nodes` * number of hours since 151 # last billing cycle plus the cost for each prediction performed. 152 }, 153 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 154 # create the version. See the 155 # [overview of model 156 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 157 # informaiton. 158 # 159 # When passing Version to 160 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 161 # the model service uses the specified location as the source of the model. 162 # Once deployed, the model version is hosted by the prediction service, so 163 # this location is useful only as a historical record. 164 # The total number of model files can't exceed 1000. 165 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 166 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 167 # response to increases and decreases in traffic. Care should be 168 # taken to ramp up traffic according to the model's ability to scale 169 # or you will start seeing increases in latency and 429 response codes. 170 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 171 # nodes are always up, starting from the time the model is deployed, so the 172 # cost of operating this model will be at least 173 # `rate` * `min_nodes` * number of hours since last billing cycle, 174 # where `rate` is the cost per node-hour as documented in 175 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 176 # even if no predictions are performed. There is additional cost for each 177 # prediction performed. 178 # 179 # Unlike manual scaling, if the load gets too heavy for the nodes 180 # that are up, the service will automatically add nodes to handle the 181 # increased load as well as scale back as traffic drops, always maintaining 182 # at least `min_nodes`. You will be charged for the time in which additional 183 # nodes are used. 184 # 185 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 186 # to a model stops (and after a cool-down period), nodes will be shut down 187 # and no charges will be incurred until traffic to the model resumes. 188 }, 189 "createTime": "A String", # Output only. The time the version was created. 190 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 191 # requests that do not specify a version. 192 # 193 # You can change the default version by calling 194 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 195 "name": "A String", # Required.The name specified for the version when it was created. 196 # 197 # The version name must be unique within the model it is created in. 198 }, 199 "name": "A String", # Required. The name specified for the model when it was created. 200 # 201 # The model name must be unique within the project it is created in. 202 "onlinePredictionLogging": True or False, # Optional. If true, enables StackDriver Logging for online prediction. 203 # Default is false. 204 "description": "A String", # Optional. The description specified for the model when it was created. 205 } 206 207 x__xgafv: string, V1 error format. 208 Allowed values 209 1 - v1 error format 210 2 - v2 error format 211 212Returns: 213 An object of the form: 214 215 { # Represents a machine learning solution. 216 # 217 # A model can have multiple versions, each of which is a deployed, trained 218 # model ready to receive prediction requests. The model itself is just a 219 # container. 220 "regions": [ # Optional. The list of regions where the model is going to be deployed. 221 # Currently only one region per model is supported. 222 # Defaults to 'us-central1' if nothing is set. 223 # Note: 224 # * No matter where a model is deployed, it can always be accessed by 225 # users from anywhere, both for online and batch prediction. 226 # * The region for a batch prediction job is set by the region field when 227 # submitting the batch prediction job and does not take its value from 228 # this field. 229 "A String", 230 ], 231 "defaultVersion": { # Represents a version of the model. # Output only. The default version of the model. This version will be used to 232 # handle prediction requests that do not specify a version. 233 # 234 # You can change the default version by calling 235 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 236 # 237 # Each version is a trained model deployed in the cloud, ready to handle 238 # prediction requests. A model can have multiple versions. You can get 239 # information about all of the versions of a given model by calling 240 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 241 "description": "A String", # Optional. The description specified for the version when it was created. 242 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 243 # If not set, Google Cloud ML will choose a version. 244 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 245 # model. You should generally use `automatic_scaling` with an appropriate 246 # `min_nodes` instead, but this option is available if you want more 247 # predictable billing. Beware that latency and error rates will increase 248 # if the traffic exceeds that capability of the system to serve it based 249 # on the selected number of nodes. 250 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 251 # starting from the time the model is deployed, so the cost of operating 252 # this model will be proportional to `nodes` * number of hours since 253 # last billing cycle plus the cost for each prediction performed. 254 }, 255 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 256 # create the version. See the 257 # [overview of model 258 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 259 # informaiton. 260 # 261 # When passing Version to 262 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 263 # the model service uses the specified location as the source of the model. 264 # Once deployed, the model version is hosted by the prediction service, so 265 # this location is useful only as a historical record. 266 # The total number of model files can't exceed 1000. 267 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 268 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 269 # response to increases and decreases in traffic. Care should be 270 # taken to ramp up traffic according to the model's ability to scale 271 # or you will start seeing increases in latency and 429 response codes. 272 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 273 # nodes are always up, starting from the time the model is deployed, so the 274 # cost of operating this model will be at least 275 # `rate` * `min_nodes` * number of hours since last billing cycle, 276 # where `rate` is the cost per node-hour as documented in 277 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 278 # even if no predictions are performed. There is additional cost for each 279 # prediction performed. 280 # 281 # Unlike manual scaling, if the load gets too heavy for the nodes 282 # that are up, the service will automatically add nodes to handle the 283 # increased load as well as scale back as traffic drops, always maintaining 284 # at least `min_nodes`. You will be charged for the time in which additional 285 # nodes are used. 286 # 287 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 288 # to a model stops (and after a cool-down period), nodes will be shut down 289 # and no charges will be incurred until traffic to the model resumes. 290 }, 291 "createTime": "A String", # Output only. The time the version was created. 292 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 293 # requests that do not specify a version. 294 # 295 # You can change the default version by calling 296 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 297 "name": "A String", # Required.The name specified for the version when it was created. 298 # 299 # The version name must be unique within the model it is created in. 300 }, 301 "name": "A String", # Required. The name specified for the model when it was created. 302 # 303 # The model name must be unique within the project it is created in. 304 "onlinePredictionLogging": True or False, # Optional. If true, enables StackDriver Logging for online prediction. 305 # Default is false. 306 "description": "A String", # Optional. The description specified for the model when it was created. 307 }</pre> 308</div> 309 310<div class="method"> 311 <code class="details" id="delete">delete(name, x__xgafv=None)</code> 312 <pre>Deletes a model. 313 314You can only delete a model if there are no versions in it. You can delete 315versions by calling 316[projects.models.versions.delete](/ml-engine/reference/rest/v1/projects.models.versions/delete). 317 318Args: 319 name: string, Required. The name of the model. 320 321Authorization: requires `Editor` role on the parent project. (required) 322 x__xgafv: string, V1 error format. 323 Allowed values 324 1 - v1 error format 325 2 - v2 error format 326 327Returns: 328 An object of the form: 329 330 { # This resource represents a long-running operation that is the result of a 331 # network API call. 332 "metadata": { # Service-specific metadata associated with the operation. It typically 333 # contains progress information and common metadata such as create time. 334 # Some services might not provide such metadata. Any method that returns a 335 # long-running operation should document the metadata type, if any. 336 "a_key": "", # Properties of the object. Contains field @type with type URL. 337 }, 338 "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. 339 # programming environments, including REST APIs and RPC APIs. It is used by 340 # [gRPC](https://github.com/grpc). The error model is designed to be: 341 # 342 # - Simple to use and understand for most users 343 # - Flexible enough to meet unexpected needs 344 # 345 # # Overview 346 # 347 # The `Status` message contains three pieces of data: error code, error message, 348 # and error details. The error code should be an enum value of 349 # google.rpc.Code, but it may accept additional error codes if needed. The 350 # error message should be a developer-facing English message that helps 351 # developers *understand* and *resolve* the error. If a localized user-facing 352 # error message is needed, put the localized message in the error details or 353 # localize it in the client. The optional error details may contain arbitrary 354 # information about the error. There is a predefined set of error detail types 355 # in the package `google.rpc` that can be used for common error conditions. 356 # 357 # # Language mapping 358 # 359 # The `Status` message is the logical representation of the error model, but it 360 # is not necessarily the actual wire format. When the `Status` message is 361 # exposed in different client libraries and different wire protocols, it can be 362 # mapped differently. For example, it will likely be mapped to some exceptions 363 # in Java, but more likely mapped to some error codes in C. 364 # 365 # # Other uses 366 # 367 # The error model and the `Status` message can be used in a variety of 368 # environments, either with or without APIs, to provide a 369 # consistent developer experience across different environments. 370 # 371 # Example uses of this error model include: 372 # 373 # - Partial errors. If a service needs to return partial errors to the client, 374 # it may embed the `Status` in the normal response to indicate the partial 375 # errors. 376 # 377 # - Workflow errors. A typical workflow has multiple steps. Each step may 378 # have a `Status` message for error reporting. 379 # 380 # - Batch operations. If a client uses batch request and batch response, the 381 # `Status` message should be used directly inside batch response, one for 382 # each error sub-response. 383 # 384 # - Asynchronous operations. If an API call embeds asynchronous operation 385 # results in its response, the status of those operations should be 386 # represented directly using the `Status` message. 387 # 388 # - Logging. If some API errors are stored in logs, the message `Status` could 389 # be used directly after any stripping needed for security/privacy reasons. 390 "message": "A String", # A developer-facing error message, which should be in English. Any 391 # user-facing error message should be localized and sent in the 392 # google.rpc.Status.details field, or localized by the client. 393 "code": 42, # The status code, which should be an enum value of google.rpc.Code. 394 "details": [ # A list of messages that carry the error details. There will be a 395 # common set of message types for APIs to use. 396 { 397 "a_key": "", # Properties of the object. Contains field @type with type URL. 398 }, 399 ], 400 }, 401 "done": True or False, # If the value is `false`, it means the operation is still in progress. 402 # If true, the operation is completed, and either `error` or `response` is 403 # available. 404 "response": { # The normal response of the operation in case of success. If the original 405 # method returns no data on success, such as `Delete`, the response is 406 # `google.protobuf.Empty`. If the original method is standard 407 # `Get`/`Create`/`Update`, the response should be the resource. For other 408 # methods, the response should have the type `XxxResponse`, where `Xxx` 409 # is the original method name. For example, if the original method name 410 # is `TakeSnapshot()`, the inferred response type is 411 # `TakeSnapshotResponse`. 412 "a_key": "", # Properties of the object. Contains field @type with type URL. 413 }, 414 "name": "A String", # The server-assigned name, which is only unique within the same service that 415 # originally returns it. If you use the default HTTP mapping, the 416 # `name` should have the format of `operations/some/unique/name`. 417 }</pre> 418</div> 419 420<div class="method"> 421 <code class="details" id="get">get(name, x__xgafv=None)</code> 422 <pre>Gets information about a model, including its name, the description (if 423set), and the default version (if at least one version of the model has 424been deployed). 425 426Args: 427 name: string, Required. The name of the model. 428 429Authorization: requires `Viewer` role on the parent project. (required) 430 x__xgafv: string, V1 error format. 431 Allowed values 432 1 - v1 error format 433 2 - v2 error format 434 435Returns: 436 An object of the form: 437 438 { # Represents a machine learning solution. 439 # 440 # A model can have multiple versions, each of which is a deployed, trained 441 # model ready to receive prediction requests. The model itself is just a 442 # container. 443 "regions": [ # Optional. The list of regions where the model is going to be deployed. 444 # Currently only one region per model is supported. 445 # Defaults to 'us-central1' if nothing is set. 446 # Note: 447 # * No matter where a model is deployed, it can always be accessed by 448 # users from anywhere, both for online and batch prediction. 449 # * The region for a batch prediction job is set by the region field when 450 # submitting the batch prediction job and does not take its value from 451 # this field. 452 "A String", 453 ], 454 "defaultVersion": { # Represents a version of the model. # Output only. The default version of the model. This version will be used to 455 # handle prediction requests that do not specify a version. 456 # 457 # You can change the default version by calling 458 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 459 # 460 # Each version is a trained model deployed in the cloud, ready to handle 461 # prediction requests. A model can have multiple versions. You can get 462 # information about all of the versions of a given model by calling 463 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 464 "description": "A String", # Optional. The description specified for the version when it was created. 465 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 466 # If not set, Google Cloud ML will choose a version. 467 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 468 # model. You should generally use `automatic_scaling` with an appropriate 469 # `min_nodes` instead, but this option is available if you want more 470 # predictable billing. Beware that latency and error rates will increase 471 # if the traffic exceeds that capability of the system to serve it based 472 # on the selected number of nodes. 473 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 474 # starting from the time the model is deployed, so the cost of operating 475 # this model will be proportional to `nodes` * number of hours since 476 # last billing cycle plus the cost for each prediction performed. 477 }, 478 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 479 # create the version. See the 480 # [overview of model 481 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 482 # informaiton. 483 # 484 # When passing Version to 485 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 486 # the model service uses the specified location as the source of the model. 487 # Once deployed, the model version is hosted by the prediction service, so 488 # this location is useful only as a historical record. 489 # The total number of model files can't exceed 1000. 490 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 491 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 492 # response to increases and decreases in traffic. Care should be 493 # taken to ramp up traffic according to the model's ability to scale 494 # or you will start seeing increases in latency and 429 response codes. 495 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 496 # nodes are always up, starting from the time the model is deployed, so the 497 # cost of operating this model will be at least 498 # `rate` * `min_nodes` * number of hours since last billing cycle, 499 # where `rate` is the cost per node-hour as documented in 500 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 501 # even if no predictions are performed. There is additional cost for each 502 # prediction performed. 503 # 504 # Unlike manual scaling, if the load gets too heavy for the nodes 505 # that are up, the service will automatically add nodes to handle the 506 # increased load as well as scale back as traffic drops, always maintaining 507 # at least `min_nodes`. You will be charged for the time in which additional 508 # nodes are used. 509 # 510 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 511 # to a model stops (and after a cool-down period), nodes will be shut down 512 # and no charges will be incurred until traffic to the model resumes. 513 }, 514 "createTime": "A String", # Output only. The time the version was created. 515 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 516 # requests that do not specify a version. 517 # 518 # You can change the default version by calling 519 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 520 "name": "A String", # Required.The name specified for the version when it was created. 521 # 522 # The version name must be unique within the model it is created in. 523 }, 524 "name": "A String", # Required. The name specified for the model when it was created. 525 # 526 # The model name must be unique within the project it is created in. 527 "onlinePredictionLogging": True or False, # Optional. If true, enables StackDriver Logging for online prediction. 528 # Default is false. 529 "description": "A String", # Optional. The description specified for the model when it was created. 530 }</pre> 531</div> 532 533<div class="method"> 534 <code class="details" id="list">list(parent, pageToken=None, x__xgafv=None, pageSize=None)</code> 535 <pre>Lists the models in a project. 536 537Each project can contain multiple models, and each model can have multiple 538versions. 539 540Args: 541 parent: string, Required. The name of the project whose models are to be listed. 542 543Authorization: requires `Viewer` role on the specified project. (required) 544 pageToken: string, Optional. A page token to request the next page of results. 545 546You get the token from the `next_page_token` field of the response from 547the previous call. 548 x__xgafv: string, V1 error format. 549 Allowed values 550 1 - v1 error format 551 2 - v2 error format 552 pageSize: integer, Optional. The number of models to retrieve per "page" of results. If there 553are more remaining results than this number, the response message will 554contain a valid value in the `next_page_token` field. 555 556The default value is 20, and the maximum page size is 100. 557 558Returns: 559 An object of the form: 560 561 { # Response message for the ListModels method. 562 "models": [ # The list of models. 563 { # Represents a machine learning solution. 564 # 565 # A model can have multiple versions, each of which is a deployed, trained 566 # model ready to receive prediction requests. The model itself is just a 567 # container. 568 "regions": [ # Optional. The list of regions where the model is going to be deployed. 569 # Currently only one region per model is supported. 570 # Defaults to 'us-central1' if nothing is set. 571 # Note: 572 # * No matter where a model is deployed, it can always be accessed by 573 # users from anywhere, both for online and batch prediction. 574 # * The region for a batch prediction job is set by the region field when 575 # submitting the batch prediction job and does not take its value from 576 # this field. 577 "A String", 578 ], 579 "defaultVersion": { # Represents a version of the model. # Output only. The default version of the model. This version will be used to 580 # handle prediction requests that do not specify a version. 581 # 582 # You can change the default version by calling 583 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 584 # 585 # Each version is a trained model deployed in the cloud, ready to handle 586 # prediction requests. A model can have multiple versions. You can get 587 # information about all of the versions of a given model by calling 588 # [projects.models.versions.list](/ml-engine/reference/rest/v1/projects.models.versions/list). 589 "description": "A String", # Optional. The description specified for the version when it was created. 590 "runtimeVersion": "A String", # Optional. The Google Cloud ML runtime version to use for this deployment. 591 # If not set, Google Cloud ML will choose a version. 592 "manualScaling": { # Options for manually scaling a model. # Manually select the number of nodes to use for serving the 593 # model. You should generally use `automatic_scaling` with an appropriate 594 # `min_nodes` instead, but this option is available if you want more 595 # predictable billing. Beware that latency and error rates will increase 596 # if the traffic exceeds that capability of the system to serve it based 597 # on the selected number of nodes. 598 "nodes": 42, # The number of nodes to allocate for this model. These nodes are always up, 599 # starting from the time the model is deployed, so the cost of operating 600 # this model will be proportional to `nodes` * number of hours since 601 # last billing cycle plus the cost for each prediction performed. 602 }, 603 "deploymentUri": "A String", # Required. The Google Cloud Storage location of the trained model used to 604 # create the version. See the 605 # [overview of model 606 # deployment](/ml-engine/docs/concepts/deployment-overview) for more 607 # informaiton. 608 # 609 # When passing Version to 610 # [projects.models.versions.create](/ml-engine/reference/rest/v1/projects.models.versions/create) 611 # the model service uses the specified location as the source of the model. 612 # Once deployed, the model version is hosted by the prediction service, so 613 # this location is useful only as a historical record. 614 # The total number of model files can't exceed 1000. 615 "lastUseTime": "A String", # Output only. The time the version was last used for prediction. 616 "automaticScaling": { # Options for automatically scaling a model. # Automatically scale the number of nodes used to serve the model in 617 # response to increases and decreases in traffic. Care should be 618 # taken to ramp up traffic according to the model's ability to scale 619 # or you will start seeing increases in latency and 429 response codes. 620 "minNodes": 42, # Optional. The minimum number of nodes to allocate for this model. These 621 # nodes are always up, starting from the time the model is deployed, so the 622 # cost of operating this model will be at least 623 # `rate` * `min_nodes` * number of hours since last billing cycle, 624 # where `rate` is the cost per node-hour as documented in 625 # [pricing](https://cloud.google.com/ml-engine/pricing#prediction_pricing), 626 # even if no predictions are performed. There is additional cost for each 627 # prediction performed. 628 # 629 # Unlike manual scaling, if the load gets too heavy for the nodes 630 # that are up, the service will automatically add nodes to handle the 631 # increased load as well as scale back as traffic drops, always maintaining 632 # at least `min_nodes`. You will be charged for the time in which additional 633 # nodes are used. 634 # 635 # If not specified, `min_nodes` defaults to 0, in which case, when traffic 636 # to a model stops (and after a cool-down period), nodes will be shut down 637 # and no charges will be incurred until traffic to the model resumes. 638 }, 639 "createTime": "A String", # Output only. The time the version was created. 640 "isDefault": True or False, # Output only. If true, this version will be used to handle prediction 641 # requests that do not specify a version. 642 # 643 # You can change the default version by calling 644 # [projects.methods.versions.setDefault](/ml-engine/reference/rest/v1/projects.models.versions/setDefault). 645 "name": "A String", # Required.The name specified for the version when it was created. 646 # 647 # The version name must be unique within the model it is created in. 648 }, 649 "name": "A String", # Required. The name specified for the model when it was created. 650 # 651 # The model name must be unique within the project it is created in. 652 "onlinePredictionLogging": True or False, # Optional. If true, enables StackDriver Logging for online prediction. 653 # Default is false. 654 "description": "A String", # Optional. The description specified for the model when it was created. 655 }, 656 ], 657 "nextPageToken": "A String", # Optional. Pass this token as the `page_token` field of the request for a 658 # subsequent call. 659 }</pre> 660</div> 661 662<div class="method"> 663 <code class="details" id="list_next">list_next(previous_request, previous_response)</code> 664 <pre>Retrieves the next page of results. 665 666Args: 667 previous_request: The request for the previous page. (required) 668 previous_response: The response from the request for the previous page. (required) 669 670Returns: 671 A request object that you can call 'execute()' on to request the next 672 page. Returns None if there are no more items in the collection. 673 </pre> 674</div> 675 676</body></html>