@@ -51,7 +51,7 @@ def _serialize(self) -> dict:
51
51
52
52
class PyMLTrainAndPredictWorkflow (WorkflowProperty ):
53
53
"""
54
- Next generation of ExabyteML. We expect workflows to have a format as follows:
54
+ We expect workflows to have a format as follows:
55
55
56
56
Workflow_Head_Subworkflow - Contains various units which prepare an ML job. For example, we may have the following
57
57
units present.
@@ -203,7 +203,7 @@ def is_using_dataset(self):
203
203
@property
204
204
def workflow_specific_config (self ) -> dict :
205
205
"""
206
- Generates the specific config for the new implementation of ExabyteML . The remainder of the config is
206
+ Generates the specific config for the workflow . The remainder of the config is
207
207
generated inside of the parent Workflow class.
208
208
209
209
Returns:
@@ -224,168 +224,3 @@ def workflow_specific_config(self) -> dict:
224
224
}
225
225
226
226
return specific_config
227
-
228
-
229
- class ExabyteMLPredictWorkflow (WorkflowProperty ):
230
- """
231
- Legacy implementation of Exabyte ML's predict Workflow property class.
232
- """
233
-
234
- def __init__ (self , name , parser , * args , ** kwargs ):
235
- super ().__init__ (name , parser , * args , ** kwargs )
236
-
237
- self .model = self .parser .model
238
- self .targets = self .parser .targets
239
- self .features = self .parser .features
240
- self .scaling_params_per_feature = self .parser .scaling_params_per_feature
241
-
242
- @property
243
- def workflow_specific_config (self ) -> dict :
244
- """
245
- Generates the specific config for a legacy ExabyteML workflow. The remainder of the config is generated
246
- inside of the parent Worfklow class.
247
-
248
- Returns:
249
- dict
250
- """
251
- specific_config = {
252
- "units" : [
253
- {
254
- "_id" : "LCthJ6E2QabYCZqf4" ,
255
- "name" : "ml_predict_subworkflow" ,
256
- "type" : "subworkflow" ,
257
- "flowchartId" : "subworkflow" ,
258
- "head" : True ,
259
- }
260
- ],
261
- "subworkflows" : [
262
- {
263
- "name" : "ml_predict_subworkflow" ,
264
- "isDraft" : True ,
265
- "application" : {
266
- "version" : "0.2.0" ,
267
- "summary" : "Exabyte Machine Learning Engine" ,
268
- "name" : "exabyteml" ,
269
- "shortName" : "ml" ,
270
- "build" : "Default" ,
271
- },
272
- "units" : [
273
- {
274
- "status" : "idle" ,
275
- "statusTrack" : [],
276
- "head" : True ,
277
- "flowchartId" : "io" ,
278
- "name" : "input" ,
279
- "application" : {
280
- "version" : "0.2.0" ,
281
- "summary" : "Exabyte Machine Learning Engine" ,
282
- "name" : "exabyteml" ,
283
- "shortName" : "ml" ,
284
- "build" : "Default" ,
285
- },
286
- "results" : [],
287
- "next" : "data_transformation_manipulation" ,
288
- "source" : "api" ,
289
- "postProcessors" : [],
290
- "preProcessors" : [],
291
- "subtype" : "dataFrame" ,
292
- "input" : [
293
- {
294
- "endpoint" : "dataframe" ,
295
- "endpoint_options" : {
296
- "headers" : {},
297
- "data" : {"features" : self .features , "ids" : [], "targets" : self .targets },
298
- "method" : "POST" ,
299
- "params" : {},
300
- "jobId" : "" ,
301
- },
302
- }
303
- ],
304
- "type" : "io" ,
305
- "monitors" : [],
306
- },
307
- {
308
- "status" : "idle" ,
309
- "statusTrack" : [],
310
- "head" : False ,
311
- "flowchartId" : "data_transformation_manipulation" ,
312
- "name" : "clean data" ,
313
- "monitors" : [],
314
- "results" : [],
315
- "next" : "data_transformation_scale_and_reduce" ,
316
- "application" : {
317
- "version" : "0.2.0" ,
318
- "summary" : "Exabyte Machine Learning Engine" ,
319
- "name" : "exabyteml" ,
320
- "shortName" : "ml" ,
321
- "build" : "Default" ,
322
- },
323
- "postProcessors" : [],
324
- "preProcessors" : [],
325
- "operationType" : "manipulation" ,
326
- "operation" : "data_transformation" ,
327
- "type" : "processing" ,
328
- "inputData" : {
329
- "cleanMissingData" : True ,
330
- "replaceNoneValuesWith" : 0 ,
331
- "removeDuplicateRows" : True ,
332
- },
333
- },
334
- {
335
- "status" : "idle" ,
336
- "statusTrack" : [],
337
- "head" : False ,
338
- "flowchartId" : "data_transformation_scale_and_reduce" ,
339
- "name" : "scale and reduce" ,
340
- "monitors" : [],
341
- "results" : [],
342
- "next" : "score" ,
343
- "application" : {
344
- "version" : "0.2.0" ,
345
- "build" : "Default" ,
346
- "name" : "exabyteml" ,
347
- "shortName" : "ml" ,
348
- "summary" : "Exabyte Machine Learning Engine" ,
349
- },
350
- "postProcessors" : [],
351
- "preProcessors" : [],
352
- "operationType" : "scale_and_reduce" ,
353
- "operation" : "data_transformation" ,
354
- "type" : "processing" ,
355
- "inputData" : {
356
- "scaler" : "standard_scaler" ,
357
- "perFeature" : self .scaling_params_per_feature ,
358
- },
359
- },
360
- {
361
- "status" : "idle" ,
362
- "statusTrack" : [],
363
- "executable" : {"name" : "score" },
364
- "flowchartId" : "score" ,
365
- "name" : "score" ,
366
- "head" : False ,
367
- "results" : [{"name" : "predicted_properties" }],
368
- "application" : {
369
- "version" : "0.2.0" ,
370
- "build" : "Default" ,
371
- "name" : "exabyteml" ,
372
- "shortName" : "ml" ,
373
- "summary" : "Exabyte Machine Learning Engine" ,
374
- },
375
- "postProcessors" : [],
376
- "preProcessors" : [],
377
- "context" : {},
378
- "input" : [],
379
- "flavor" : {"name" : "score" },
380
- "type" : "execution" ,
381
- "monitors" : [{"name" : "standard_output" }],
382
- },
383
- ],
384
- "model" : self .model ,
385
- "_id" : "LCthJ6E2QabYCZqf4" ,
386
- "properties" : self .targets ,
387
- }
388
- ],
389
- "properties" : self .targets ,
390
- }
391
- return specific_config
0 commit comments