Upload blackbox yahpo-iaml_xgboost

yahpo/iaml_xgboost/best_params_resnet.json

@@ -0,0 +1 @@
+{"d": 384, "d_hidden_factor": 2.7930060842224904, "hidden_dropout": 0.09343650341288146, "lr": 0.0007388150356328408, "mixup": false, "n_layers": 6, "opt_tfms_alpha": true, "opt_tfms_auc": true, "opt_tfms_colsample_bylevel": true, "opt_tfms_colsample_bytree": true, "opt_tfms_eta": false, "opt_tfms_f1": false, "opt_tfms_gamma": false, "opt_tfms_ias": false, "opt_tfms_lambda": true, "opt_tfms_logloss": false, "opt_tfms_max_depth": false, "opt_tfms_mec": false, "opt_tfms_min_child_weight": true, "opt_tfms_mmce": true, "opt_tfms_nrounds": true, "opt_tfms_rammodel": false, "opt_tfms_rampredict": true, "opt_tfms_ramtrain": true, "opt_tfms_rate_drop": false, "opt_tfms_skip_drop": true, "opt_tfms_subsample": true, "opt_tfms_timepredict": false, "opt_tfms_timetrain": false, "opt_tfms_trainsize": false, "tfms_alpha": "tlog", "tfms_auc": "tnexp", "tfms_colsample_bylevel": "tnexp", "tfms_colsample_bytree": "tnexp", "tfms_lambda": "tlog", "tfms_min_child_weight": "tnexp", "tfms_mmce": "tlog", "tfms_nrounds": "tlog", "tfms_rampredict": "tnexp", "tfms_ramtrain": "tnexp", "tfms_skip_drop": "tnexp", "tfms_subsample": "tlog", "use_residual_dropout": false}

yahpo/iaml_xgboost/config_space.json

@@ -0,0 +1,202 @@
+{
+  "hyperparameters": [
+    {
+      "name": "alpha",
+      "type": "uniform_float",
+      "log": true,
+      "lower": 0.00010000000000000009,
+      "upper": 999.9999999999998,
+      "default": 0.316227766
+    },
+    {
+      "name": "booster",
+      "type": "categorical",
+      "choices": [
+        "gblinear",
+        "gbtree",
+        "dart"
+      ],
+      "default": "gblinear",
+      "probabilities": null
+    },
+    {
+      "name": "lambda",
+      "type": "uniform_float",
+      "log": true,
+      "lower": 0.00010000000000000009,
+      "upper": 999.9999999999998,
+      "default": 0.316227766
+    },
+    {
+      "name": "nrounds",
+      "type": "uniform_int",
+      "log": true,
+      "lower": 3,
+      "upper": 2000,
+      "default": 77
+    },
+    {
+      "name": "subsample",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.1,
+      "upper": 1.0,
+      "default": 0.55
+    },
+    {
+      "name": "task_id",
+      "type": "categorical",
+      "choices": [
+        "40981",
+        "41146",
+        "1489",
+        "1067"
+      ],
+      "default": "40981",
+      "probabilities": null
+    },
+    {
+      "name": "trainsize",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.03,
+      "upper": 1.0,
+      "default": 0.525
+    },
+    {
+      "name": "colsample_bylevel",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.01,
+      "upper": 1.0,
+      "default": 0.505
+    },
+    {
+      "name": "colsample_bytree",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.01,
+      "upper": 1.0,
+      "default": 0.505
+    },
+    {
+      "name": "eta",
+      "type": "uniform_float",
+      "log": true,
+      "lower": 0.00010000000000000009,
+      "upper": 1.0,
+      "default": 0.01
+    },
+    {
+      "name": "gamma",
+      "type": "uniform_float",
+      "log": true,
+      "lower": 0.00010000000000000009,
+      "upper": 6.999999999999999,
+      "default": 0.0264575131
+    },
+    {
+      "name": "max_depth",
+      "type": "uniform_int",
+      "log": false,
+      "lower": 1,
+      "upper": 15,
+      "default": 8
+    },
+    {
+      "name": "min_child_weight",
+      "type": "uniform_float",
+      "log": true,
+      "lower": 2.718281828459045,
+      "upper": 149.99999999999997,
+      "default": 20.1926292064
+    },
+    {
+      "name": "rate_drop",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.0,
+      "upper": 1.0,
+      "default": 0.5
+    },
+    {
+      "name": "skip_drop",
+      "type": "uniform_float",
+      "log": false,
+      "lower": 0.0,
+      "upper": 1.0,
+      "default": 0.5
+    }
+  ],
+  "conditions": [
+    {
+      "child": "colsample_bylevel",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "colsample_bytree",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "eta",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "gamma",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "max_depth",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "min_child_weight",
+      "parent": "booster",
+      "type": "IN",
+      "values": [
+        "dart",
+        "gbtree"
+      ]
+    },
+    {
+      "child": "rate_drop",
+      "parent": "booster",
+      "type": "EQ",
+      "value": "dart"
+    },
+    {
+      "child": "skip_drop",
+      "parent": "booster",
+      "type": "EQ",
+      "value": "dart"
+    }
+  ],
+  "forbiddens": [],
+  "python_module_version": "0.4.19",
+  "json_format_version": 0.2
+}

yahpo/iaml_xgboost/encoding.json

@@ -0,0 +1 @@
+{"booster": {"#na#": 0, "dart": 1, "gblinear": 2, "gbtree": 3}, "task_id": {"#na#": 0, "1067": 1, "1489": 2, "40981": 3, "41146": 4}}

yahpo/iaml_xgboost/metadata.json

@@ -0,0 +1 @@
+{"metric_elapsed_time": "time", "metric_default": "val_accuracy", "resource_attr": "st_worker_iter"}

yahpo/iaml_xgboost/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd51307d61b7cd746f5cee82b174aeecf05f9459da07e36ca8bf267d8d2d4d94
+size 29827930

yahpo/iaml_xgboost/param_set.R

@@ -0,0 +1,50 @@
+search_space = ps(
+  booster = p_fct(levels = c("gblinear", "gbtree", "dart")),
+  nrounds = p_dbl(lower = 1, upper = log(2000), tags = c("int", "log"), trafo = function(x) as.integer(round(exp(x)))),
+  eta = p_dbl(lower = log(1e-4), upper = log(1), tags = "log", trafo = function(x) exp(x), depends = booster %in% c("dart", "gbtree")),
+  gamma = p_dbl(lower = log(1e-4), upper = log(7), tags = "log", trafo = function(x) exp(x), depends = booster %in% c("dart", "gbtree")),
+  lambda = p_dbl(lower = log(1e-4), upper = log(1000), tags = "log", trafo = function(x) exp(x)),
+  alpha = p_dbl(lower = log(1e-4), upper = log(1000), tags = "log", trafo = function(x) exp(x)),
+  subsample = p_dbl(lower = 0.1, upper = 1),
+  max_depth = p_int(lower = 1L, upper = 15L, depends = booster %in% c("dart", "gbtree")),
+  min_child_weight = p_dbl(lower = 1, upper = log(150), tags = "log", trafo = function(x) exp(x), depends = booster %in% c("dart", "gbtree")),
+  colsample_bytree = p_dbl(lower = 0.01, upper = 1, depends = booster %in% c("dart", "gbtree")),
+  colsample_bylevel = p_dbl(lower = 0.01, upper = 1, depends = booster %in% c("dart", "gbtree")),
+  rate_drop = p_dbl(lower = 0, upper = 1, depends = booster == "dart"),
+  skip_drop = p_dbl(lower = 0, upper = 1, depends = booster == "dart"),
+  trainsize = p_dbl(lower = 0.03, upper = 1, tags = "budget"),
+  task_id = p_fct(levels = c("40981", "41146", "1489", "1067"), tags = "task_id")
+)
+
+domain = ps(
+  booster = p_fct(levels = c("gblinear", "gbtree", "dart")),
+  nrounds = p_int(lower = 3, upper = 2000),
+  eta = p_dbl(lower = 1e-4, upper = 1, depends = booster %in% c("dart", "gbtree")),
+  gamma = p_dbl(lower = 1e-4, upper = 7, depends = booster %in% c("dart", "gbtree")),
+  lambda = p_dbl(lower = 1e-4, upper = 1000),
+  alpha = p_dbl(lower = 1e-4, upper = 1000),
+  subsample = p_dbl(lower = 0.1, upper = 1),
+  max_depth = p_int(lower = 1L, upper = 15L, depends = booster %in% c("dart", "gbtree")),
+  min_child_weight = p_dbl(lower = exp(1), upper = 150, depends = booster %in% c("dart", "gbtree")),
+  colsample_bytree = p_dbl(lower = 0.01, upper = 1, depends = booster %in% c("dart", "gbtree")),
+  colsample_bylevel = p_dbl(lower = 0.01, upper = 1, depends = booster %in% c("dart", "gbtree")),
+  rate_drop = p_dbl(lower = 0, upper = 1, depends = booster == "dart"),
+  skip_drop = p_dbl(lower = 0, upper = 1, depends = booster == "dart"),
+  trainsize = p_dbl(lower = 0.03, upper = 1, tags = "budget"),
+  task_id = p_fct(levels = c("40981", "41146", "1489", "1067"), tags = "task_id")
+)
+
+codomain = ps(
+  mmce = p_dbl(lower = 0, upper = 1, tags = "minimize"),
+  f1 = p_dbl(lower = 0, upper = 1, tags = "maximize"),
+  auc = p_dbl(lower = 0, upper = 1, tags = "maximize"),
+  logloss = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  ramtrain = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  rammodel = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  rampredict = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  timetrain = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  timepredict = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  mec = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  ias = p_dbl(lower = 0, upper = Inf, tags = "minimize"),
+  nf = p_dbl(lower = 0, upper = Inf, tags = "minimize")
+)