java.lang.IllegalArgumentException: Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0]

JIRA | Eric Eckstrand | 2 years ago
  1. 0

    -Training dataset: cup98LRN_z.csv -Test dataset: cup98VAL_z.csv BT Response: TARGET_B do_classification: TRUE family: gaussian *** Should return error with message that classification and gaussian are incompatible options or provide a warning that classification will be chosen, not regression. BT *** When a validation_frame is specified, metrics such as AUC, MSE are not provided as part of model output. Instead, only the model training parameters are shown. What is the model output schema when a validation frame is specified? BT *** Lambda search doesn't return multiple models. Is this even desired. *** Lambda search model output includes best_lambda_idx, but doesn't include lambda BT *** L-BFGS only support alpha=0 BT *** when lambda_search=FALSE and lambda exceeds lambda_max, receive an IEA in JAVA: barrier onExCompletion for hex.glm.GLM$1@fe9be90 java.lang.IllegalArgumentException: Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0] at hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:672) at hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:643) at water.H2O$H2OCallback.onCompletion(H2O.java:640) at jsr166y.CountedCompleter.tryComplete(CountedCompleter.java:386) at water.MRTask.compute2(MRTask.java:437) at water.MRTask.compute2(MRTask.java:372) at water.H2O$H2OCountedCompleter.compute(H2O.java:582) at jsr166y.CountedCompleter.exec(CountedCompleter.java:429) at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263) at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974) at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477) at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104) and in R: Error in .h2o.__waitOnJob(conn, job_key) : Job key $0301ac1002baded6ffffffff$_8860d6a5c0376852634f3132f8e74800 failed R should catch this error and return a user-friendly message. Perhaps Java should return a model with all coefficients 0?

    JIRA | 2 years ago | Eric Eckstrand
    java.lang.IllegalArgumentException: Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0]
  2. 0

    -Training dataset: cup98LRN_z.csv -Test dataset: cup98VAL_z.csv BT Response: TARGET_B do_classification: TRUE family: gaussian *** Should return error with message that classification and gaussian are incompatible options or provide a warning that classification will be chosen, not regression. BT *** When a validation_frame is specified, metrics such as AUC, MSE are not provided as part of model output. Instead, only the model training parameters are shown. What is the model output schema when a validation frame is specified? BT *** Lambda search doesn't return multiple models. Is this even desired. *** Lambda search model output includes best_lambda_idx, but doesn't include lambda BT *** L-BFGS only support alpha=0 BT *** when lambda_search=FALSE and lambda exceeds lambda_max, receive an IEA in JAVA: barrier onExCompletion for hex.glm.GLM$1@fe9be90 java.lang.IllegalArgumentException: Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0] at hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:672) at hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:643) at water.H2O$H2OCallback.onCompletion(H2O.java:640) at jsr166y.CountedCompleter.tryComplete(CountedCompleter.java:386) at water.MRTask.compute2(MRTask.java:437) at water.MRTask.compute2(MRTask.java:372) at water.H2O$H2OCountedCompleter.compute(H2O.java:582) at jsr166y.CountedCompleter.exec(CountedCompleter.java:429) at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263) at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974) at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477) at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104) and in R: Error in .h2o.__waitOnJob(conn, job_key) : Job key $0301ac1002baded6ffffffff$_8860d6a5c0376852634f3132f8e74800 failed R should catch this error and return a user-friendly message. Perhaps Java should return a model with all coefficients 0?

    JIRA | 2 years ago | Eric Eckstrand
    java.lang.IllegalArgumentException: Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0]
  3. 0

    pyunit_citi_bike_small.py http://mr-0xa1:8080/job/master_pyunit_small/2367/console exception while polling job http://mr-0xa1:8080/job/master_pyunit_small/2367/artifact/h2o-py/tests/results/java_5_0.out.txt 7-12 12:01:24.964 172.17.2.152:57799 32492 #44883-20 INFO: Method: POST , URI: /3/ModelBuilders/glm, route: /3/ModelBuilders/glm, parms: {lambda=[1e-05], family=poisson, training_frame=py50455748-f2fe-4c07-8616-ae7cf99fda43, validation_frame=pyf0dbc0b9-f35a-4576-9926-c9d86aefcb28, response_column=bikes} 07-12 12:01:24.968 172.17.2.152:57799 32492 FJ-0-109 INFO: Building H2O GLM model with these parameters: 07-12 12:01:24.969 172.17.2.152:57799 32492 FJ-0-109 INFO: {"_model_id":null,"_train":{"name":"py50455748-f2fe-4c07-8616-ae7cf99fda43","type":"Key"},"_valid":{"name":"pyf0dbc0b9-f35a-4576-9926-c9d86aefcb28","type":"Key"},"_ignored_columns":null,"_ignore_const_cols":true,"_weights_column":null,"_offset_column":null,"_score_each_iteration":false,"_response_column":"bikes","_balance_classes":false,"_max_after_balance_size":5.0,"_class_sampling_factors":null,"_max_hit_ratio_k":10,"_max_confusion_matrix_size":20,"_standardize":true,"_family":"poisson","_link":"family_default","_solver":"IRLSM","_tweedie_variance_power":0.0,"_tweedie_link_power":1.0,"_alpha":null,"_lambda":[1.0E-5],"_prior":-1.0,"_lambda_search":false,"_nlambdas":-1,"_non_negative":false,"_exactLambdas":false,"_lambda_min_ratio":-1.0,"_use_all_factor_levels":false,"_max_iterations":-1,"_nfolds":0,"_intercept":true,"_beta_epsilon":1.0E-4,"_objective_epsilon":1.0E-5,"_gradient_epsilon":1.0E-4,"_beta_constraints":null,"_max_active_predictors":-1} 07-12 12:01:25.006 172.17.2.152:57799 32492 FJ-0-109 WARN: _training_frame: Dataset has less than 1/2 of the data after filtering out rows with NAs 07-12 12:01:25.007 172.17.2.152:57799 32492 FJ-0-109 ERRR: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. 07-12 12:01:25.007 172.17.2.152:57799 32492 FJ-0-109 WARN: _lambda: removed 1 lambda values which were greater than lambda_max = 0.0 barrier onExCompletion for hex.glm.GLM$GLMDriver@22fbc1cd java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. at hex.Model$Output.<init>(Model.java:244) at hex.glm.GLMModel$GLMOutput.<init>(GLMModel.java:452) at hex.glm.GLMModel.<init>(GLMModel.java:27) at hex.glm.GLM.init(GLM.java:381) at hex.glm.GLM$GLMDriver.compute2(GLM.java:736) at water.H2O$H2OCountedCompleter.compute(H2O.java:929) at jsr166y.CountedCompleter.exec(CountedCompleter.java:429) at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263) at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974) at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477) at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104) http://mr-0xa1:8080/job/master_pyunit_small/2367/artifact/h2o-py/tests/results/testdir_demos_notebooks_pyunit_citi_bike_small.py.out.txt glm Model Build Progress: [ ] 00% glm Model Build Progress: [ ] 00% Traceback (most recent call last): File "pyunit_citi_bike_small.py", line 13, in <module> h2o.run_test(sys.argv, demo_citibike_small) File "../../../h2o/h2o.py", line 296, in run_test test_to_run(ip, port) File "pyunit_citi_bike_small.py", line 10, in demo_citibike_small h2o.ipy_notebook_exec(h2o.locate("h2o-py/demos/citi_bike_small.ipynb"),save_and_norun=False) File "../../../h2o/h2o.py", line 335, in ipy_notebook_exec exec program in d # safe, but horrible (exec is horrible) File "<string>", line 213, in <module> File "<string>", line 120, in split_fit_predict File "../../../h2o/h2o.py", line 640, in glm return h2o_model_builder.supervised_model_build(x,y,validation_x,validation_y,"glm",kwargs) File "../../../h2o/h2o_model_builder.py", line 25, in supervised_model_build return _model_build(x,y,validation_x,validation_y,algo_url,kwargs) File "../../../h2o/h2o_model_builder.py", line 104, in _model_build job = H2OJob(H2OConnection.post_json("ModelBuilders/"+algo_url, **kwargs), job_type=(algo_url+" Model Build")).poll() File "../../../h2o/job.py", line 48, in poll raise EnvironmentError("Job with key {} failed with an exception: {}".format(self.job_key, self.exception)) EnvironmentError: Job with key $0301ac110298c8e1ffffffff$_848a8390c61dc22aec25507d20b94538 failed with an exception: None

    JIRA | 1 year ago | Kevin Normoyle
    java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute.
  4. Speed up your debug routine!

    Automated exception search integrated into your IDE

  5. 0

    pyunit_citi_bike_small.py http://mr-0xa1:8080/job/master_pyunit_small/2367/console exception while polling job http://mr-0xa1:8080/job/master_pyunit_small/2367/artifact/h2o-py/tests/results/java_5_0.out.txt 7-12 12:01:24.964 172.17.2.152:57799 32492 #44883-20 INFO: Method: POST , URI: /3/ModelBuilders/glm, route: /3/ModelBuilders/glm, parms: {lambda=[1e-05], family=poisson, training_frame=py50455748-f2fe-4c07-8616-ae7cf99fda43, validation_frame=pyf0dbc0b9-f35a-4576-9926-c9d86aefcb28, response_column=bikes} 07-12 12:01:24.968 172.17.2.152:57799 32492 FJ-0-109 INFO: Building H2O GLM model with these parameters: 07-12 12:01:24.969 172.17.2.152:57799 32492 FJ-0-109 INFO: {"_model_id":null,"_train":{"name":"py50455748-f2fe-4c07-8616-ae7cf99fda43","type":"Key"},"_valid":{"name":"pyf0dbc0b9-f35a-4576-9926-c9d86aefcb28","type":"Key"},"_ignored_columns":null,"_ignore_const_cols":true,"_weights_column":null,"_offset_column":null,"_score_each_iteration":false,"_response_column":"bikes","_balance_classes":false,"_max_after_balance_size":5.0,"_class_sampling_factors":null,"_max_hit_ratio_k":10,"_max_confusion_matrix_size":20,"_standardize":true,"_family":"poisson","_link":"family_default","_solver":"IRLSM","_tweedie_variance_power":0.0,"_tweedie_link_power":1.0,"_alpha":null,"_lambda":[1.0E-5],"_prior":-1.0,"_lambda_search":false,"_nlambdas":-1,"_non_negative":false,"_exactLambdas":false,"_lambda_min_ratio":-1.0,"_use_all_factor_levels":false,"_max_iterations":-1,"_nfolds":0,"_intercept":true,"_beta_epsilon":1.0E-4,"_objective_epsilon":1.0E-5,"_gradient_epsilon":1.0E-4,"_beta_constraints":null,"_max_active_predictors":-1} 07-12 12:01:25.006 172.17.2.152:57799 32492 FJ-0-109 WARN: _training_frame: Dataset has less than 1/2 of the data after filtering out rows with NAs 07-12 12:01:25.007 172.17.2.152:57799 32492 FJ-0-109 ERRR: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. 07-12 12:01:25.007 172.17.2.152:57799 32492 FJ-0-109 WARN: _lambda: removed 1 lambda values which were greater than lambda_max = 0.0 barrier onExCompletion for hex.glm.GLM$GLMDriver@22fbc1cd java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. at hex.Model$Output.<init>(Model.java:244) at hex.glm.GLMModel$GLMOutput.<init>(GLMModel.java:452) at hex.glm.GLMModel.<init>(GLMModel.java:27) at hex.glm.GLM.init(GLM.java:381) at hex.glm.GLM$GLMDriver.compute2(GLM.java:736) at water.H2O$H2OCountedCompleter.compute(H2O.java:929) at jsr166y.CountedCompleter.exec(CountedCompleter.java:429) at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263) at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974) at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477) at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104) http://mr-0xa1:8080/job/master_pyunit_small/2367/artifact/h2o-py/tests/results/testdir_demos_notebooks_pyunit_citi_bike_small.py.out.txt glm Model Build Progress: [ ] 00% glm Model Build Progress: [ ] 00% Traceback (most recent call last): File "pyunit_citi_bike_small.py", line 13, in <module> h2o.run_test(sys.argv, demo_citibike_small) File "../../../h2o/h2o.py", line 296, in run_test test_to_run(ip, port) File "pyunit_citi_bike_small.py", line 10, in demo_citibike_small h2o.ipy_notebook_exec(h2o.locate("h2o-py/demos/citi_bike_small.ipynb"),save_and_norun=False) File "../../../h2o/h2o.py", line 335, in ipy_notebook_exec exec program in d # safe, but horrible (exec is horrible) File "<string>", line 213, in <module> File "<string>", line 120, in split_fit_predict File "../../../h2o/h2o.py", line 640, in glm return h2o_model_builder.supervised_model_build(x,y,validation_x,validation_y,"glm",kwargs) File "../../../h2o/h2o_model_builder.py", line 25, in supervised_model_build return _model_build(x,y,validation_x,validation_y,algo_url,kwargs) File "../../../h2o/h2o_model_builder.py", line 104, in _model_build job = H2OJob(H2OConnection.post_json("ModelBuilders/"+algo_url, **kwargs), job_type=(algo_url+" Model Build")).poll() File "../../../h2o/job.py", line 48, in poll raise EnvironmentError("Job with key {} failed with an exception: {}".format(self.job_key, self.exception)) EnvironmentError: Job with key $0301ac110298c8e1ffffffff$_848a8390c61dc22aec25507d20b94538 failed with an exception: None

    JIRA | 1 year ago | Kevin Normoyle
    java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute.
  6. 0

    the last block of this notebook calls glm: glm0 = h2o.glm(x =train.drop("bikes"), y =train ["bikes"], validation_x=test .drop("bikes"), validation_y=test ["bikes"], Lambda=[1e-5], family="poisson") which intermittently fails with: INFO: Building H2O GLM model with these parameters: 08-05 19:02:04.409 172.16.2.18:54321 54601 FJ-0-7 INFO: {"_model_id":null,"_train":{"name":"py48db56af-03c8-4b75-87cc-ca6a50ee42b1","type":"Key"},"_valid":{"name":"pyc774a1d8-02bc-4726-be60-f44faa823093","type":"Key"},"_nfolds":0,"_keep_cross_validation_predictions":false,"_fold_assignment":"AUTO","_distribution":"AUTO","_tweedie_power":1.5,"_ignored_columns":null,"_ignore_const_cols":true,"_weights_column":null,"_offset_column":null,"_fold_column":null,"_score_each_iteration":false,"_response_column":"bikes","_balance_classes":false,"_max_after_balance_size":5.0,"_class_sampling_factors":null,"_max_hit_ratio_k":10,"_max_confusion_matrix_size":20,"_checkpoint":null,"_standardize":true,"_family":"poisson","_link":"family_default","_solver":"IRLSM","_tweedie_variance_power":0.0,"_tweedie_link_power":1.0,"_alpha":null,"_lambda":[1.0E-5],"_prior":-1.0,"_lambda_search":false,"_nlambdas":-1,"_non_negative":false,"_exactLambdas":false,"_lambda_min_ratio":-1.0,"_use_all_factor_levels":false,"_max_iterations":-1,"_intercept":true,"_beta_epsilon":1.0E-4,"_objective_epsilon":1.0E-5,"_gradient_epsilon":1.0E-4,"_beta_constraints":null,"_max_active_predictors":-1} 08-05 19:02:04.411 172.16.2.18:54321 54601 FJ-0-7 WARN: _training_frame: Dataset has less than 1/2 of the data after filtering out rows with NAs 08-05 19:02:04.411 172.16.2.18:54321 54601 FJ-0-7 ERRR: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. 08-05 19:02:04.412 172.16.2.18:54321 54601 FJ-0-7 WARN: _lambda: removed 1 lambda values which were greater than lambda_max = 0.0 barrier onExCompletion for hex.glm.GLM$GLMDriver@752c388a java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute. at hex.Model$Output.<init>(Model.java:306) at hex.glm.GLMModel$GLMOutput.<init>(GLMModel.java:457) at hex.glm.GLMModel.<init>(GLMModel.java:27) at hex.glm.GLM.init(GLM.java:380) at hex.glm.GLM$GLMDriver.compute2(GLM.java:741) at water.H2O$H2OCountedCompleter.compute(H2O.java:964) at jsr166y.CountedCompleter.exec(CountedCompleter.java:429) at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263) at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974) at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477) at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104) I'll try to repro the exact datasets and encode in junit.

    JIRA | 1 year ago | Eric Eckstrand
    java.lang.IllegalArgumentException: ERROR on field: _lambda: All passed lambda values are > lambda_max = 0.0, nothing to compute.

    Not finding the right solution?
    Take a tour to get the most out of Samebug.

    Tired of useless tips?

    Automated exception search integrated into your IDE

    Root Cause Analysis

    1. java.lang.IllegalArgumentException

      Given lambda(s) are all > lambda_max = 0.00849253869600811, have nothing to run with. lambda = [10.0]

      at hex.glm.GLM$GLMDriver$1$1.callback()
    2. hex.glm
      GLM$GLMDriver$1$1.callback
      1. hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:672)
      2. hex.glm.GLM$GLMDriver$1$1.callback(GLM.java:643)
      2 frames
    3. water
      H2O$H2OCallback.onCompletion
      1. water.H2O$H2OCallback.onCompletion(H2O.java:640)
      1 frame
    4. jsr166y
      CountedCompleter.tryComplete
      1. jsr166y.CountedCompleter.tryComplete(CountedCompleter.java:386)
      1 frame
    5. water
      H2O$H2OCountedCompleter.compute
      1. water.MRTask.compute2(MRTask.java:437)
      2. water.MRTask.compute2(MRTask.java:372)
      3. water.H2O$H2OCountedCompleter.compute(H2O.java:582)
      3 frames
    6. jsr166y
      ForkJoinWorkerThread.run
      1. jsr166y.CountedCompleter.exec(CountedCompleter.java:429)
      2. jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263)
      3. jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974)
      4. jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477)
      5. jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104)
      5 frames