Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / python / google-api-python-client / 860a6378c538d8e2e424c7be150b41ba28120387 / . / docs / dyn / ml_v1beta1.projects.jobs.html
blob: 35a941e878ab9395550895fe5b9bb5fb614e37ce [file] [log] [blame]
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]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_v1beta1.html">Google Cloud Machine Learning</a> . <a href="ml_v1beta1.projects.html">projects</a> . <a href="ml_v1beta1.projects.jobs.html">jobs</a></h1>
76<h2>Instance Methods</h2>
77<p class="toc_element">
78 <code><a href="#cancel">cancel(name=None, body, x__xgafv=None)</a></code></p>
79<p class="firstline">Cancels a running job.</p>
80<p class="toc_element">
81 <code><a href="#create">create(parent=None, body, x__xgafv=None)</a></code></p>
82<p class="firstline">Creates a training or a batch prediction job.</p>
83<p class="toc_element">
84 <code><a href="#get">get(name=None, x__xgafv=None)</a></code></p>
85<p class="firstline">Describes a job.</p>
86<p class="toc_element">
87 <code><a href="#list">list(parent=None, pageSize=None, filter=None, pageToken=None, x__xgafv=None)</a></code></p>
88<p class="firstline">Lists the jobs in the project.</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<h3>Method Details</h3>
93<div class="method">
94 <code class="details" id="cancel">cancel(name=None, body, x__xgafv=None)</code>
95 <pre>Cancels a running job.
96
97Args:
98 name: string, Required. The name of the job to cancel.
99
100Authorization: requires `Editor` role on the parent project. (required)
101 body: object, The request body. (required)
102 The object takes the form of:
103
104{ # Request message for the CancelJob method.
105 }
106
107 x__xgafv: string, V1 error format.
108 Allowed values
109 1 - v1 error format
110 2 - v2 error format
111
112Returns:
113 An object of the form:
114
115 { # A generic empty message that you can re-use to avoid defining duplicated
116 # empty messages in your APIs. A typical example is to use it as the request
117 # or the response type of an API method. For instance:
118 #
119 # service Foo {
120 # rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty);
121 # }
122 #
123 # The JSON representation for `Empty` is empty JSON object `{}`.
124 }</pre>
125</div>
126
127<div class="method">
128 <code class="details" id="create">create(parent=None, body, x__xgafv=None)</code>
129 <pre>Creates a training or a batch prediction job.
130
131Args:
132 parent: string, Required. The project name.
133
134Authorization: requires `Editor` role on the specified project. (required)
135 body: object, The request body. (required)
136 The object takes the form of:
137
138{ # Represents a training or prediction job.
139 "trainingOutput": { # Represents results of a training job. # The current training job result.
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]140 "consumedMLUnits": 3.14, # The amount of ML units consumed by the job.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]141 "trials": [ # Results for individual Hyperparameter trials.
142 { # Represents the result of a single hyperparameter tuning trial from a
143 # training job. The TrainingOutput object that is returned on successful
144 # completion of a training job with hyperparameter tuning includes a list
145 # of HyperparameterOutput objects, one for each successful trial.
146 "hyperparameters": { # The hyperparameters given to this trial.
147 "a_key": "A String",
148 },
149 "trialId": "A String", # The trial id for these results.
150 "allMetrics": [ # All recorded object metrics for this trial.
151 { # An observed value of a metric.
152 "trainingStep": "A String", # The global training step for this metric.
153 "objectiveValue": 3.14, # The objective value at this training step.
154 },
155 ],
156 "finalMetric": { # An observed value of a metric. # The final objective metric seen for this trial.
157 "trainingStep": "A String", # The global training step for this metric.
158 "objectiveValue": 3.14, # The objective value at this training step.
159 },
160 },
161 ],
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]162 "completedTrialCount": "A String", # The number of hyperparameter tuning trials that completed successfully.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]163 },
164 "startTime": "A String", # Output only. When the job processing was started.
165 "errorMessage": "A String", # Output only. The details of a failure or a cancellation.
166 "jobId": "A String", # Required. The user-specified id of the job.
167 "state": "A String", # Output only. The detailed state of a job.
168 "predictionInput": { # Represents input parameters for a prediction job. # Input parameters to create a prediction job.
169 "modelName": "A String", # Use this field if you want to use the default version for the specified
170 # model. The string must use the following format:
171 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]172 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>[YOUR_MODEL]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]173 "inputPaths": [ # Required. The Google Cloud Storage location of the input data files.
174 # May contain wildcards.
175 "A String",
176 ],
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]177 "maxWorkerCount": "A String", # Optional. The maximum number of workers to be used for parallel processing.
178 # Defaults to 10 if not specified.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]179 "outputPath": "A String", # Required. The output Google Cloud Storage location.
180 "dataFormat": "A String", # Required. The format of the input data files.
181 "versionName": "A String", # Use this field if you want to specify a version of the model to use. The
182 # string is formatted the same way as `model_version`, with the addition
183 # of the version information:
184 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]185 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>YOUR_MODEL/versions/<var>[YOUR_VERSION]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]186 "region": "A String", # Required. The Google Compute Engine region to run the prediction job in.
187 },
188 "trainingInput": { # Represents input parameters for a training job. # Input parameters to create a training job.
189 "workerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
190 # job's worker nodes.
191 #
192 # The supported values are the same as those described in the entry for
193 # `masterType`.
194 #
195 # This value must be present when `scaleTier` is set to `CUSTOM` and
196 # `workerCount` is greater than zero.
197 "scaleTier": "A String", # Required. Specifies the machine types, the number of replicas for workers
198 # and parameter servers.
199 "masterType": "A String", # Optional. Specifies the type of virtual machine to use for your training
200 # job's master worker.
201 #
202 # The following types are supported:
203 #
204 # <dl>
205 # <dt>standard</dt>
206 # <dd>
207 # A basic machine configuration suitable for training simple models with
208 # small to moderate datasets.
209 # </dd>
210 # <dt>large_model</dt>
211 # <dd>
212 # A machine with a lot of memory, specially suited for parameter servers
213 # when your model is large (having many hidden layers or layers with very
214 # large numbers of nodes).
215 # </dd>
216 # <dt>complex_model_s</dt>
217 # <dd>
218 # A machine suitable for the master and workers of the cluster when your
219 # model requires more computation than the standard machine can handle
220 # satisfactorily.
221 # </dd>
222 # <dt>complex_model_m</dt>
223 # <dd>
224 # A machine with roughly twice the number of cores and roughly double the
225 # memory of <code suppresswarning="true">complex_model_s</code>.
226 # </dd>
227 # <dt>complex_model_l</dt>
228 # <dd>
229 # A machine with roughly twice the number of cores and roughly double the
230 # memory of <code suppresswarning="true">complex_model_m</code>.
231 # </dd>
232 # </dl>
233 #
234 # You must set this value when `scaleTier` is set to `CUSTOM`.
235 "hyperparameters": { # Represents a set of hyperparameters to optimize. # Optional. The set of Hyperparameters to tune.
236 "maxTrials": 42, # Optional. How many training trials should be attempted to optimize
237 # the specified hyperparameters.
238 #
239 # Defaults to one.
240 "params": [ # Required. The set of parameters to tune.
241 { # Represents a single hyperparameter to optimize.
242 "maxValue": 3.14, # Required if typeis `DOUBLE` or `INTEGER`. This field
243 # should be unset if type is `CATEGORICAL`. This value should be integers if
244 # type is `INTEGER`.
245 "minValue": 3.14, # Required if type is `DOUBLE` or `INTEGER`. This field
246 # should be unset if type is `CATEGORICAL`. This value should be integers if
247 # type is INTEGER.
248 "discreteValues": [ # Required if type is `DISCRETE`.
249 # A list of feasible points.
250 # The list should be in strictly increasing order. For instance, this
251 # parameter might have possible settings of 1.5, 2.5, and 4.0. This list
252 # should not contain more than 1,000 values.
253 3.14,
254 ],
255 "parameterName": "A String", # Required. The parameter name must be unique amongst all ParameterConfigs in
256 # a HyperparameterSpec message. E.g., "learning_rate".
257 "categoricalValues": [ # Required if type is `CATEGORICAL`. The list of possible categories.
258 "A String",
259 ],
260 "type": "A String", # Required. The type of the parameter.
261 "scaleType": "A String", # Optional. How the parameter should be scaled to the hypercube.
262 # Leave unset for categorical parameters.
263 # Some kind of scaling is strongly recommended for real or integral
264 # parameters (e.g., `UNIT_LINEAR_SCALE`).
265 },
266 ],
267 "maxParallelTrials": 42, # Optional. The number of training trials to run concurrently.
268 # You can reduce the time it takes to perform hyperparameter tuning by adding
269 # trials in parallel. However, each trail only benefits from the information
270 # gained in completed trials. That means that a trial does not get access to
271 # the results of trials running at the same time, which could reduce the
272 # quality of the overall optimization.
273 #
274 # Each trial will use the same scale tier and machine types.
275 #
276 # Defaults to one.
277 "goal": "A String", # Required. The type of goal to use for tuning. Available types are
278 # `MAXIMIZE` and `MINIMIZE`.
279 #
280 # Defaults to `MAXIMIZE`.
281 },
282 "region": "A String", # Required. The Google Compute Engine region to run the training job in.
283 "args": [ # Optional. Command line arguments to pass to the program.
284 "A String",
285 ],
286 "pythonModule": "A String", # Required. The Python module name to run after installing the packages.
287 "packageUris": [ # Required. The Google Cloud Storage location of the packages with
288 # the training program and any additional dependencies.
289 "A String",
290 ],
291 "workerCount": "A String", # Optional. The number of worker replicas to use for the training job. Each
292 # replica in the cluster will be of the type specified in `worker_type`.
293 #
294 # This value can only be used when `scale_tier` is set to `CUSTOM`. If you
295 # set this value, you must also set `worker_type`.
296 "parameterServerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
297 # job's parameter server.
298 #
299 # The supported values are the same as those described in the entry for
300 # `master_type`.
301 #
302 # This value must be present when `scaleTier` is set to `CUSTOM` and
303 # `parameter_server_count` is greater than zero.
304 "parameterServerCount": "A String", # Optional. The number of parameter server replicas to use for the training
305 # job. Each replica in the cluster will be of the type specified in
306 # `parameter_server_type`.
307 #
308 # This value can only be used when `scale_tier` is set to `CUSTOM`.If you
309 # set this value, you must also set `parameter_server_type`.
310 },
311 "endTime": "A String", # Output only. When the job processing was completed.
312 "predictionOutput": { # Represents results of a prediction job. # The current prediction job result.
313 "outputPath": "A String", # The output Google Cloud Storage location provided at the job creation time.
314 "predictionCount": "A String", # The number of generated predictions.
315 "errorCount": "A String", # The number of data instances which resulted in errors.
316 },
317 "createTime": "A String", # Output only. When the job was created.
318 }
319
320 x__xgafv: string, V1 error format.
321 Allowed values
322 1 - v1 error format
323 2 - v2 error format
324
325Returns:
326 An object of the form:
327
328 { # Represents a training or prediction job.
329 "trainingOutput": { # Represents results of a training job. # The current training job result.
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]330 "consumedMLUnits": 3.14, # The amount of ML units consumed by the job.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]331 "trials": [ # Results for individual Hyperparameter trials.
332 { # Represents the result of a single hyperparameter tuning trial from a
333 # training job. The TrainingOutput object that is returned on successful
334 # completion of a training job with hyperparameter tuning includes a list
335 # of HyperparameterOutput objects, one for each successful trial.
336 "hyperparameters": { # The hyperparameters given to this trial.
337 "a_key": "A String",
338 },
339 "trialId": "A String", # The trial id for these results.
340 "allMetrics": [ # All recorded object metrics for this trial.
341 { # An observed value of a metric.
342 "trainingStep": "A String", # The global training step for this metric.
343 "objectiveValue": 3.14, # The objective value at this training step.
344 },
345 ],
346 "finalMetric": { # An observed value of a metric. # The final objective metric seen for this trial.
347 "trainingStep": "A String", # The global training step for this metric.
348 "objectiveValue": 3.14, # The objective value at this training step.
349 },
350 },
351 ],
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]352 "completedTrialCount": "A String", # The number of hyperparameter tuning trials that completed successfully.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]353 },
354 "startTime": "A String", # Output only. When the job processing was started.
355 "errorMessage": "A String", # Output only. The details of a failure or a cancellation.
356 "jobId": "A String", # Required. The user-specified id of the job.
357 "state": "A String", # Output only. The detailed state of a job.
358 "predictionInput": { # Represents input parameters for a prediction job. # Input parameters to create a prediction job.
359 "modelName": "A String", # Use this field if you want to use the default version for the specified
360 # model. The string must use the following format:
361 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]362 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>[YOUR_MODEL]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]363 "inputPaths": [ # Required. The Google Cloud Storage location of the input data files.
364 # May contain wildcards.
365 "A String",
366 ],
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]367 "maxWorkerCount": "A String", # Optional. The maximum number of workers to be used for parallel processing.
368 # Defaults to 10 if not specified.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]369 "outputPath": "A String", # Required. The output Google Cloud Storage location.
370 "dataFormat": "A String", # Required. The format of the input data files.
371 "versionName": "A String", # Use this field if you want to specify a version of the model to use. The
372 # string is formatted the same way as `model_version`, with the addition
373 # of the version information:
374 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]375 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>YOUR_MODEL/versions/<var>[YOUR_VERSION]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]376 "region": "A String", # Required. The Google Compute Engine region to run the prediction job in.
377 },
378 "trainingInput": { # Represents input parameters for a training job. # Input parameters to create a training job.
379 "workerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
380 # job's worker nodes.
381 #
382 # The supported values are the same as those described in the entry for
383 # `masterType`.
384 #
385 # This value must be present when `scaleTier` is set to `CUSTOM` and
386 # `workerCount` is greater than zero.
387 "scaleTier": "A String", # Required. Specifies the machine types, the number of replicas for workers
388 # and parameter servers.
389 "masterType": "A String", # Optional. Specifies the type of virtual machine to use for your training
390 # job's master worker.
391 #
392 # The following types are supported:
393 #
394 # <dl>
395 # <dt>standard</dt>
396 # <dd>
397 # A basic machine configuration suitable for training simple models with
398 # small to moderate datasets.
399 # </dd>
400 # <dt>large_model</dt>
401 # <dd>
402 # A machine with a lot of memory, specially suited for parameter servers
403 # when your model is large (having many hidden layers or layers with very
404 # large numbers of nodes).
405 # </dd>
406 # <dt>complex_model_s</dt>
407 # <dd>
408 # A machine suitable for the master and workers of the cluster when your
409 # model requires more computation than the standard machine can handle
410 # satisfactorily.
411 # </dd>
412 # <dt>complex_model_m</dt>
413 # <dd>
414 # A machine with roughly twice the number of cores and roughly double the
415 # memory of <code suppresswarning="true">complex_model_s</code>.
416 # </dd>
417 # <dt>complex_model_l</dt>
418 # <dd>
419 # A machine with roughly twice the number of cores and roughly double the
420 # memory of <code suppresswarning="true">complex_model_m</code>.
421 # </dd>
422 # </dl>
423 #
424 # You must set this value when `scaleTier` is set to `CUSTOM`.
425 "hyperparameters": { # Represents a set of hyperparameters to optimize. # Optional. The set of Hyperparameters to tune.
426 "maxTrials": 42, # Optional. How many training trials should be attempted to optimize
427 # the specified hyperparameters.
428 #
429 # Defaults to one.
430 "params": [ # Required. The set of parameters to tune.
431 { # Represents a single hyperparameter to optimize.
432 "maxValue": 3.14, # Required if typeis `DOUBLE` or `INTEGER`. This field
433 # should be unset if type is `CATEGORICAL`. This value should be integers if
434 # type is `INTEGER`.
435 "minValue": 3.14, # Required if type is `DOUBLE` or `INTEGER`. This field
436 # should be unset if type is `CATEGORICAL`. This value should be integers if
437 # type is INTEGER.
438 "discreteValues": [ # Required if type is `DISCRETE`.
439 # A list of feasible points.
440 # The list should be in strictly increasing order. For instance, this
441 # parameter might have possible settings of 1.5, 2.5, and 4.0. This list
442 # should not contain more than 1,000 values.
443 3.14,
444 ],
445 "parameterName": "A String", # Required. The parameter name must be unique amongst all ParameterConfigs in
446 # a HyperparameterSpec message. E.g., "learning_rate".
447 "categoricalValues": [ # Required if type is `CATEGORICAL`. The list of possible categories.
448 "A String",
449 ],
450 "type": "A String", # Required. The type of the parameter.
451 "scaleType": "A String", # Optional. How the parameter should be scaled to the hypercube.
452 # Leave unset for categorical parameters.
453 # Some kind of scaling is strongly recommended for real or integral
454 # parameters (e.g., `UNIT_LINEAR_SCALE`).
455 },
456 ],
457 "maxParallelTrials": 42, # Optional. The number of training trials to run concurrently.
458 # You can reduce the time it takes to perform hyperparameter tuning by adding
459 # trials in parallel. However, each trail only benefits from the information
460 # gained in completed trials. That means that a trial does not get access to
461 # the results of trials running at the same time, which could reduce the
462 # quality of the overall optimization.
463 #
464 # Each trial will use the same scale tier and machine types.
465 #
466 # Defaults to one.
467 "goal": "A String", # Required. The type of goal to use for tuning. Available types are
468 # `MAXIMIZE` and `MINIMIZE`.
469 #
470 # Defaults to `MAXIMIZE`.
471 },
472 "region": "A String", # Required. The Google Compute Engine region to run the training job in.
473 "args": [ # Optional. Command line arguments to pass to the program.
474 "A String",
475 ],
476 "pythonModule": "A String", # Required. The Python module name to run after installing the packages.
477 "packageUris": [ # Required. The Google Cloud Storage location of the packages with
478 # the training program and any additional dependencies.
479 "A String",
480 ],
481 "workerCount": "A String", # Optional. The number of worker replicas to use for the training job. Each
482 # replica in the cluster will be of the type specified in `worker_type`.
483 #
484 # This value can only be used when `scale_tier` is set to `CUSTOM`. If you
485 # set this value, you must also set `worker_type`.
486 "parameterServerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
487 # job's parameter server.
488 #
489 # The supported values are the same as those described in the entry for
490 # `master_type`.
491 #
492 # This value must be present when `scaleTier` is set to `CUSTOM` and
493 # `parameter_server_count` is greater than zero.
494 "parameterServerCount": "A String", # Optional. The number of parameter server replicas to use for the training
495 # job. Each replica in the cluster will be of the type specified in
496 # `parameter_server_type`.
497 #
498 # This value can only be used when `scale_tier` is set to `CUSTOM`.If you
499 # set this value, you must also set `parameter_server_type`.
500 },
501 "endTime": "A String", # Output only. When the job processing was completed.
502 "predictionOutput": { # Represents results of a prediction job. # The current prediction job result.
503 "outputPath": "A String", # The output Google Cloud Storage location provided at the job creation time.
504 "predictionCount": "A String", # The number of generated predictions.
505 "errorCount": "A String", # The number of data instances which resulted in errors.
506 },
507 "createTime": "A String", # Output only. When the job was created.
508 }</pre>
509</div>
510
511<div class="method">
512 <code class="details" id="get">get(name=None, x__xgafv=None)</code>
513 <pre>Describes a job.
514
515Args:
516 name: string, Required. The name of the job to get the description of.
517
518Authorization: requires `Viewer` role on the parent project. (required)
519 x__xgafv: string, V1 error format.
520 Allowed values
521 1 - v1 error format
522 2 - v2 error format
523
524Returns:
525 An object of the form:
526
527 { # Represents a training or prediction job.
528 "trainingOutput": { # Represents results of a training job. # The current training job result.
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]529 "consumedMLUnits": 3.14, # The amount of ML units consumed by the job.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]530 "trials": [ # Results for individual Hyperparameter trials.
531 { # Represents the result of a single hyperparameter tuning trial from a
532 # training job. The TrainingOutput object that is returned on successful
533 # completion of a training job with hyperparameter tuning includes a list
534 # of HyperparameterOutput objects, one for each successful trial.
535 "hyperparameters": { # The hyperparameters given to this trial.
536 "a_key": "A String",
537 },
538 "trialId": "A String", # The trial id for these results.
539 "allMetrics": [ # All recorded object metrics for this trial.
540 { # An observed value of a metric.
541 "trainingStep": "A String", # The global training step for this metric.
542 "objectiveValue": 3.14, # The objective value at this training step.
543 },
544 ],
545 "finalMetric": { # An observed value of a metric. # The final objective metric seen for this trial.
546 "trainingStep": "A String", # The global training step for this metric.
547 "objectiveValue": 3.14, # The objective value at this training step.
548 },
549 },
550 ],
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]551 "completedTrialCount": "A String", # The number of hyperparameter tuning trials that completed successfully.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]552 },
553 "startTime": "A String", # Output only. When the job processing was started.
554 "errorMessage": "A String", # Output only. The details of a failure or a cancellation.
555 "jobId": "A String", # Required. The user-specified id of the job.
556 "state": "A String", # Output only. The detailed state of a job.
557 "predictionInput": { # Represents input parameters for a prediction job. # Input parameters to create a prediction job.
558 "modelName": "A String", # Use this field if you want to use the default version for the specified
559 # model. The string must use the following format:
560 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]561 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>[YOUR_MODEL]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]562 "inputPaths": [ # Required. The Google Cloud Storage location of the input data files.
563 # May contain wildcards.
564 "A String",
565 ],
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]566 "maxWorkerCount": "A String", # Optional. The maximum number of workers to be used for parallel processing.
567 # Defaults to 10 if not specified.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]568 "outputPath": "A String", # Required. The output Google Cloud Storage location.
569 "dataFormat": "A String", # Required. The format of the input data files.
570 "versionName": "A String", # Use this field if you want to specify a version of the model to use. The
571 # string is formatted the same way as `model_version`, with the addition
572 # of the version information:
573 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]574 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>YOUR_MODEL/versions/<var>[YOUR_VERSION]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]575 "region": "A String", # Required. The Google Compute Engine region to run the prediction job in.
576 },
577 "trainingInput": { # Represents input parameters for a training job. # Input parameters to create a training job.
578 "workerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
579 # job's worker nodes.
580 #
581 # The supported values are the same as those described in the entry for
582 # `masterType`.
583 #
584 # This value must be present when `scaleTier` is set to `CUSTOM` and
585 # `workerCount` is greater than zero.
586 "scaleTier": "A String", # Required. Specifies the machine types, the number of replicas for workers
587 # and parameter servers.
588 "masterType": "A String", # Optional. Specifies the type of virtual machine to use for your training
589 # job's master worker.
590 #
591 # The following types are supported:
592 #
593 # <dl>
594 # <dt>standard</dt>
595 # <dd>
596 # A basic machine configuration suitable for training simple models with
597 # small to moderate datasets.
598 # </dd>
599 # <dt>large_model</dt>
600 # <dd>
601 # A machine with a lot of memory, specially suited for parameter servers
602 # when your model is large (having many hidden layers or layers with very
603 # large numbers of nodes).
604 # </dd>
605 # <dt>complex_model_s</dt>
606 # <dd>
607 # A machine suitable for the master and workers of the cluster when your
608 # model requires more computation than the standard machine can handle
609 # satisfactorily.
610 # </dd>
611 # <dt>complex_model_m</dt>
612 # <dd>
613 # A machine with roughly twice the number of cores and roughly double the
614 # memory of <code suppresswarning="true">complex_model_s</code>.
615 # </dd>
616 # <dt>complex_model_l</dt>
617 # <dd>
618 # A machine with roughly twice the number of cores and roughly double the
619 # memory of <code suppresswarning="true">complex_model_m</code>.
620 # </dd>
621 # </dl>
622 #
623 # You must set this value when `scaleTier` is set to `CUSTOM`.
624 "hyperparameters": { # Represents a set of hyperparameters to optimize. # Optional. The set of Hyperparameters to tune.
625 "maxTrials": 42, # Optional. How many training trials should be attempted to optimize
626 # the specified hyperparameters.
627 #
628 # Defaults to one.
629 "params": [ # Required. The set of parameters to tune.
630 { # Represents a single hyperparameter to optimize.
631 "maxValue": 3.14, # Required if typeis `DOUBLE` or `INTEGER`. This field
632 # should be unset if type is `CATEGORICAL`. This value should be integers if
633 # type is `INTEGER`.
634 "minValue": 3.14, # Required if type is `DOUBLE` or `INTEGER`. This field
635 # should be unset if type is `CATEGORICAL`. This value should be integers if
636 # type is INTEGER.
637 "discreteValues": [ # Required if type is `DISCRETE`.
638 # A list of feasible points.
639 # The list should be in strictly increasing order. For instance, this
640 # parameter might have possible settings of 1.5, 2.5, and 4.0. This list
641 # should not contain more than 1,000 values.
642 3.14,
643 ],
644 "parameterName": "A String", # Required. The parameter name must be unique amongst all ParameterConfigs in
645 # a HyperparameterSpec message. E.g., "learning_rate".
646 "categoricalValues": [ # Required if type is `CATEGORICAL`. The list of possible categories.
647 "A String",
648 ],
649 "type": "A String", # Required. The type of the parameter.
650 "scaleType": "A String", # Optional. How the parameter should be scaled to the hypercube.
651 # Leave unset for categorical parameters.
652 # Some kind of scaling is strongly recommended for real or integral
653 # parameters (e.g., `UNIT_LINEAR_SCALE`).
654 },
655 ],
656 "maxParallelTrials": 42, # Optional. The number of training trials to run concurrently.
657 # You can reduce the time it takes to perform hyperparameter tuning by adding
658 # trials in parallel. However, each trail only benefits from the information
659 # gained in completed trials. That means that a trial does not get access to
660 # the results of trials running at the same time, which could reduce the
661 # quality of the overall optimization.
662 #
663 # Each trial will use the same scale tier and machine types.
664 #
665 # Defaults to one.
666 "goal": "A String", # Required. The type of goal to use for tuning. Available types are
667 # `MAXIMIZE` and `MINIMIZE`.
668 #
669 # Defaults to `MAXIMIZE`.
670 },
671 "region": "A String", # Required. The Google Compute Engine region to run the training job in.
672 "args": [ # Optional. Command line arguments to pass to the program.
673 "A String",
674 ],
675 "pythonModule": "A String", # Required. The Python module name to run after installing the packages.
676 "packageUris": [ # Required. The Google Cloud Storage location of the packages with
677 # the training program and any additional dependencies.
678 "A String",
679 ],
680 "workerCount": "A String", # Optional. The number of worker replicas to use for the training job. Each
681 # replica in the cluster will be of the type specified in `worker_type`.
682 #
683 # This value can only be used when `scale_tier` is set to `CUSTOM`. If you
684 # set this value, you must also set `worker_type`.
685 "parameterServerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
686 # job's parameter server.
687 #
688 # The supported values are the same as those described in the entry for
689 # `master_type`.
690 #
691 # This value must be present when `scaleTier` is set to `CUSTOM` and
692 # `parameter_server_count` is greater than zero.
693 "parameterServerCount": "A String", # Optional. The number of parameter server replicas to use for the training
694 # job. Each replica in the cluster will be of the type specified in
695 # `parameter_server_type`.
696 #
697 # This value can only be used when `scale_tier` is set to `CUSTOM`.If you
698 # set this value, you must also set `parameter_server_type`.
699 },
700 "endTime": "A String", # Output only. When the job processing was completed.
701 "predictionOutput": { # Represents results of a prediction job. # The current prediction job result.
702 "outputPath": "A String", # The output Google Cloud Storage location provided at the job creation time.
703 "predictionCount": "A String", # The number of generated predictions.
704 "errorCount": "A String", # The number of data instances which resulted in errors.
705 },
706 "createTime": "A String", # Output only. When the job was created.
707 }</pre>
708</div>
709
710<div class="method">
711 <code class="details" id="list">list(parent=None, pageSize=None, filter=None, pageToken=None, x__xgafv=None)</code>
712 <pre>Lists the jobs in the project.
713
714Args:
715 parent: string, Required. The name of the project for which to list jobs.
716
717Authorization: requires `Viewer` role on the specified project. (required)
718 pageSize: integer, Optional. The number of jobs to retrieve per "page" of results. If there
719are more remaining results than this number, the response message will
720contain a valid value in the `next_page_token` field.
721
722The default value is 20, and the maximum page size is 100.
723 filter: string, Optional. Specifies the subset of jobs to retrieve.
724 pageToken: string, Optional. A page token to request the next page of results.
725
726You get the token from the `next_page_token` field of the response from
727the previous call.
728 x__xgafv: string, V1 error format.
729 Allowed values
730 1 - v1 error format
731 2 - v2 error format
732
733Returns:
734 An object of the form:
735
736 { # Response message for the ListJobs method.
737 "nextPageToken": "A String", # Optional. Pass this token as the `page_token` field of the request for a
738 # subsequent call.
739 "jobs": [ # The list of jobs.
740 { # Represents a training or prediction job.
741 "trainingOutput": { # Represents results of a training job. # The current training job result.
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]742 "consumedMLUnits": 3.14, # The amount of ML units consumed by the job.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]743 "trials": [ # Results for individual Hyperparameter trials.
744 { # Represents the result of a single hyperparameter tuning trial from a
745 # training job. The TrainingOutput object that is returned on successful
746 # completion of a training job with hyperparameter tuning includes a list
747 # of HyperparameterOutput objects, one for each successful trial.
748 "hyperparameters": { # The hyperparameters given to this trial.
749 "a_key": "A String",
750 },
751 "trialId": "A String", # The trial id for these results.
752 "allMetrics": [ # All recorded object metrics for this trial.
753 { # An observed value of a metric.
754 "trainingStep": "A String", # The global training step for this metric.
755 "objectiveValue": 3.14, # The objective value at this training step.
756 },
757 ],
758 "finalMetric": { # An observed value of a metric. # The final objective metric seen for this trial.
759 "trainingStep": "A String", # The global training step for this metric.
760 "objectiveValue": 3.14, # The objective value at this training step.
761 },
762 },
763 ],
Jon Wayne Parrott692617a2017-01-06 09:58:29 -0800[diff] [blame]764 "completedTrialCount": "A String", # The number of hyperparameter tuning trials that completed successfully.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]765 },
766 "startTime": "A String", # Output only. When the job processing was started.
767 "errorMessage": "A String", # Output only. The details of a failure or a cancellation.
768 "jobId": "A String", # Required. The user-specified id of the job.
769 "state": "A String", # Output only. The detailed state of a job.
770 "predictionInput": { # Represents input parameters for a prediction job. # Input parameters to create a prediction job.
771 "modelName": "A String", # Use this field if you want to use the default version for the specified
772 # model. The string must use the following format:
773 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]774 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>[YOUR_MODEL]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]775 "inputPaths": [ # Required. The Google Cloud Storage location of the input data files.
776 # May contain wildcards.
777 "A String",
778 ],
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]779 "maxWorkerCount": "A String", # Optional. The maximum number of workers to be used for parallel processing.
780 # Defaults to 10 if not specified.
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]781 "outputPath": "A String", # Required. The output Google Cloud Storage location.
782 "dataFormat": "A String", # Required. The format of the input data files.
783 "versionName": "A String", # Use this field if you want to specify a version of the model to use. The
784 # string is formatted the same way as `model_version`, with the addition
785 # of the version information:
786 #
Jon Wayne Parrottdc6c1ef2016-10-14 11:04:30 -0700[diff] [blame]787 # `"projects/<var>[YOUR_PROJECT]</var>/models/<var>YOUR_MODEL/versions/<var>[YOUR_VERSION]</var>"`
Sai Cheemalapatiea3a5e12016-10-12 14:05:53 -0700[diff] [blame]788 "region": "A String", # Required. The Google Compute Engine region to run the prediction job in.
789 },
790 "trainingInput": { # Represents input parameters for a training job. # Input parameters to create a training job.
791 "workerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
792 # job's worker nodes.
793 #
794 # The supported values are the same as those described in the entry for
795 # `masterType`.
796 #
797 # This value must be present when `scaleTier` is set to `CUSTOM` and
798 # `workerCount` is greater than zero.
799 "scaleTier": "A String", # Required. Specifies the machine types, the number of replicas for workers
800 # and parameter servers.
801 "masterType": "A String", # Optional. Specifies the type of virtual machine to use for your training
802 # job's master worker.
803 #
804 # The following types are supported:
805 #
806 # <dl>
807 # <dt>standard</dt>
808 # <dd>
809 # A basic machine configuration suitable for training simple models with
810 # small to moderate datasets.
811 # </dd>
812 # <dt>large_model</dt>
813 # <dd>
814 # A machine with a lot of memory, specially suited for parameter servers
815 # when your model is large (having many hidden layers or layers with very
816 # large numbers of nodes).
817 # </dd>
818 # <dt>complex_model_s</dt>
819 # <dd>
820 # A machine suitable for the master and workers of the cluster when your
821 # model requires more computation than the standard machine can handle
822 # satisfactorily.
823 # </dd>
824 # <dt>complex_model_m</dt>
825 # <dd>
826 # A machine with roughly twice the number of cores and roughly double the
827 # memory of <code suppresswarning="true">complex_model_s</code>.
828 # </dd>
829 # <dt>complex_model_l</dt>
830 # <dd>
831 # A machine with roughly twice the number of cores and roughly double the
832 # memory of <code suppresswarning="true">complex_model_m</code>.
833 # </dd>
834 # </dl>
835 #
836 # You must set this value when `scaleTier` is set to `CUSTOM`.
837 "hyperparameters": { # Represents a set of hyperparameters to optimize. # Optional. The set of Hyperparameters to tune.
838 "maxTrials": 42, # Optional. How many training trials should be attempted to optimize
839 # the specified hyperparameters.
840 #
841 # Defaults to one.
842 "params": [ # Required. The set of parameters to tune.
843 { # Represents a single hyperparameter to optimize.
844 "maxValue": 3.14, # Required if typeis `DOUBLE` or `INTEGER`. This field
845 # should be unset if type is `CATEGORICAL`. This value should be integers if
846 # type is `INTEGER`.
847 "minValue": 3.14, # Required if type is `DOUBLE` or `INTEGER`. This field
848 # should be unset if type is `CATEGORICAL`. This value should be integers if
849 # type is INTEGER.
850 "discreteValues": [ # Required if type is `DISCRETE`.
851 # A list of feasible points.
852 # The list should be in strictly increasing order. For instance, this
853 # parameter might have possible settings of 1.5, 2.5, and 4.0. This list
854 # should not contain more than 1,000 values.
855 3.14,
856 ],
857 "parameterName": "A String", # Required. The parameter name must be unique amongst all ParameterConfigs in
858 # a HyperparameterSpec message. E.g., "learning_rate".
859 "categoricalValues": [ # Required if type is `CATEGORICAL`. The list of possible categories.
860 "A String",
861 ],
862 "type": "A String", # Required. The type of the parameter.
863 "scaleType": "A String", # Optional. How the parameter should be scaled to the hypercube.
864 # Leave unset for categorical parameters.
865 # Some kind of scaling is strongly recommended for real or integral
866 # parameters (e.g., `UNIT_LINEAR_SCALE`).
867 },
868 ],
869 "maxParallelTrials": 42, # Optional. The number of training trials to run concurrently.
870 # You can reduce the time it takes to perform hyperparameter tuning by adding
871 # trials in parallel. However, each trail only benefits from the information
872 # gained in completed trials. That means that a trial does not get access to
873 # the results of trials running at the same time, which could reduce the
874 # quality of the overall optimization.
875 #
876 # Each trial will use the same scale tier and machine types.
877 #
878 # Defaults to one.
879 "goal": "A String", # Required. The type of goal to use for tuning. Available types are
880 # `MAXIMIZE` and `MINIMIZE`.
881 #
882 # Defaults to `MAXIMIZE`.
883 },
884 "region": "A String", # Required. The Google Compute Engine region to run the training job in.
885 "args": [ # Optional. Command line arguments to pass to the program.
886 "A String",
887 ],
888 "pythonModule": "A String", # Required. The Python module name to run after installing the packages.
889 "packageUris": [ # Required. The Google Cloud Storage location of the packages with
890 # the training program and any additional dependencies.
891 "A String",
892 ],
893 "workerCount": "A String", # Optional. The number of worker replicas to use for the training job. Each
894 # replica in the cluster will be of the type specified in `worker_type`.
895 #
896 # This value can only be used when `scale_tier` is set to `CUSTOM`. If you
897 # set this value, you must also set `worker_type`.
898 "parameterServerType": "A String", # Optional. Specifies the type of virtual machine to use for your training
899 # job's parameter server.
900 #
901 # The supported values are the same as those described in the entry for
902 # `master_type`.
903 #
904 # This value must be present when `scaleTier` is set to `CUSTOM` and
905 # `parameter_server_count` is greater than zero.
906 "parameterServerCount": "A String", # Optional. The number of parameter server replicas to use for the training
907 # job. Each replica in the cluster will be of the type specified in
908 # `parameter_server_type`.
909 #
910 # This value can only be used when `scale_tier` is set to `CUSTOM`.If you
911 # set this value, you must also set `parameter_server_type`.
912 },
913 "endTime": "A String", # Output only. When the job processing was completed.
914 "predictionOutput": { # Represents results of a prediction job. # The current prediction job result.
915 "outputPath": "A String", # The output Google Cloud Storage location provided at the job creation time.
916 "predictionCount": "A String", # The number of generated predictions.
917 "errorCount": "A String", # The number of data instances which resulted in errors.
918 },
919 "createTime": "A String", # Output only. When the job was created.
920 },
921 ],
922 }</pre>
923</div>
924
925<div class="method">
926 <code class="details" id="list_next">list_next(previous_request, previous_response)</code>
927 <pre>Retrieves the next page of results.
928
929Args:
930 previous_request: The request for the previous page. (required)
931 previous_response: The response from the request for the previous page. (required)
932
933Returns:
934 A request object that you can call 'execute()' on to request the next
935 page. Returns None if there are no more items in the collection.
936 </pre>
937</div>
938
939</body></html>