[AutoScheduler][AutoTVM] Enable xgboost >= 1.7.x new changes (apache#14036)

Enable xgboost >= 1.7.x new changes

Author: cbalint13

docs/install/from_source.rst

@@ -347,7 +347,7 @@ like ``virtualenv``.
 
    .. code:: bash
 
-       pip3 install --user tornado psutil 'xgboost<1.6.0' cloudpickle
+       pip3 install --user tornado psutil 'xgboost>=1.1.0' cloudpickle
 
 Note on M1 macs, you may have trouble installing xgboost / scipy. scipy and xgboost requires some additional dependencies to be installed,
 including openblas and its dependencies. Use the following commands to install scipy and xgboost with the required dependencies and
@@ -363,7 +363,7 @@ configuration. A workaround for this is to do the following commands:
 
         pip install scipy --no-use-pep517
 
-        pip install 'xgboost<1.6.0'
+        pip install 'xgboost>=1.1.0'
 
 Install Contrib Libraries
 -------------------------

python/gen_requirements.py

@@ -276,7 +276,7 @@
     ("torch", None),
     ("torchvision", None),
     ("tornado", None),
-    ("xgboost", ">=1.1.0,<1.6.0"),  # From PR #4953 & Issue #12009
+    ("xgboost", ">=1.1.0"),  # From PR #4953 & Issue #12009
 ]
 
 ################################################################################

python/tvm/auto_scheduler/cost_model/xgb_model.py

@@ -19,6 +19,7 @@
 """Cost model based on xgboost"""
 import multiprocessing
 import logging
+from typing import Dict
 from collections import defaultdict
 
 import numpy as np
@@ -28,6 +29,14 @@
 from ..feature import get_per_store_features_from_measure_pairs, get_per_store_features_from_states
 from ..measure_record import RecordReader
 
+try:
+    from xgboost.callback import TrainingCallback  # type: ignore
+except ImportError:
+
+    class TrainingCallback:  # type: ignore
+        pass
+
+
 xgb = None
 
 logger = logging.getLogger("auto_scheduler")
@@ -198,7 +207,7 @@ def update(self, inputs, results):
             num_boost_round=10000,
             obj=pack_sum_square_error,
             callbacks=[
-                custom_callback(
+                CustomCallback(
                     stopping_rounds=50,
                     metric="tr-p-rmse",
                     fevals=[
@@ -539,125 +548,144 @@ def feval(preds, labels):
     return feval
 
 
-def custom_callback(
-    stopping_rounds,
-    metric,
-    fevals,
-    evals=(),
-    log_file=None,
-    maximize=False,
-    verbose_eval=True,
-    skip_every=2,
-):
-    """Callback function for xgboost to support multiple custom evaluation functions"""
-    # pylint: disable=import-outside-toplevel
-    from xgboost.core import EarlyStopException
-    from xgboost.callback import _fmt_metric
-
-    try:
-        from xgboost.training import aggcv
-    except ImportError:
-        from xgboost.callback import _aggcv as aggcv
-
-    state = {}
-    metric_shortname = metric.split("-")[1]
-
-    def init(env):
-        """internal function"""
-        bst = env.model
-
-        state["maximize_score"] = maximize
-        state["best_iteration"] = 0
-        if maximize:
-            state["best_score"] = float("-inf")
-        else:
-            state["best_score"] = float("inf")
+class XGBoostCallback(TrainingCallback):
+    """Base class for XGBoost callbacks."""
 
-        if bst is not None:
-            if bst.attr("best_score") is not None:
-                state["best_score"] = float(bst.attr("best_score"))
-                state["best_iteration"] = int(bst.attr("best_iteration"))
-                state["best_msg"] = bst.attr("best_msg")
-            else:
-                bst.set_attr(best_iteration=str(state["best_iteration"]))
-                bst.set_attr(best_score=str(state["best_score"]))
-        else:
-            assert env.cvfolds is not None
+    def __call__(self, env: "xgb.core.CallbackEnv"):
+        # Compatibility with xgboost < 1.3
+        return self.after_iteration(env.model, env.iteration, env.evaluation_result_list)
 
-    def callback(env):
-        """internal function"""
-        if not state:
-            init(env)
+    def after_iteration(self, model: "xgb.Booster", epoch: int, evals_log: Dict):
+        raise NotImplementedError
+
+
+class CustomCallback(XGBoostCallback):
+    """
+    Callback function for xgboost.
+    Support custom evaluation function and early-stopping.
+    """
+
+    def __init__(
+        self,
+        stopping_rounds,
+        metric,
+        fevals,
+        evals=(),
+        log_file=None,
+        maximize=False,
+        verbose_eval=True,
+        skip_every=2,
+    ):
+        """Init function"""
+        self.stopping_rounds = stopping_rounds
+        self.metric = metric
+        self.metric_shortname = metric.split("-")[1]
+        self.fevals = fevals
+        self.evals = evals
+        self.log_file = log_file
+        self.maximize = maximize
+        self.verbose_eval = verbose_eval
+        self.skip_every = skip_every
+        self.state = {}
 
-        bst = env.model
-        i = env.iteration
-        cvfolds = env.cvfolds
+    def after_iteration(self, model: "xgb.Booster", epoch: int, evals_log: Dict):
+        """Run after each iteration.  Return True when training should stop."""
+        # pylint:disable = import-outside-toplevel
+        try:
+            from xgboost.callback import _fmt_metric  # type: ignore
+        except ImportError:
+            # Compatibility with xgboost >= 1.6
+            def _fmt_metric(value, show_stdv=True):
+                """format metric string"""
+                if len(value) == 2:
+                    return f"{value[0]}:{value[1]:.5f}"
+                if len(value) == 3:
+                    if show_stdv:
+                        return f"{value[0]}:{value[1]:.5f}+{value[2]:.5f}"
+                    return f"{value[0]}:{value[1]:.5f}"
+                raise ValueError("wrong metric value", value)
+
+        ##### init state #####
+        if not self.state:
+            self.state["maximize_score"] = self.maximize
+            self.state["best_iteration"] = 0
+            if self.maximize:
+                self.state["best_score"] = float("-inf")
+            else:
+                self.state["best_score"] = float("inf")
 
+            assert model is not None
+            if model.attr("best_score") is not None:
+                self.state["best_score"] = float(model.attr("best_score"))
+                self.state["best_iteration"] = int(model.attr("best_iteration"))
+                self.state["best_msg"] = model.attr("best_msg")
+            else:
+                model.set_attr(best_iteration=str(self.state["best_iteration"]))
+                model.set_attr(best_score=str(self.state["best_score"]))
         res_dict = {}
 
-        if i % skip_every == 1:
-            return
+        if epoch % self.skip_every == 1:
+            return False
 
         ##### evaluation #####
-        if cvfolds is not None:
-            for feval in fevals:
-                tmp = aggcv([f.eval(i, feval) for f in cvfolds])
-                for k, mean, std in tmp:
-                    res_dict[k] = [mean, std]
-        else:
-            for feval in fevals:
-                bst_eval = bst.eval_set(evals, i, feval)
-                res = [x.split(":") for x in bst_eval.split()]
-                for kv in res[1:]:
-                    res_dict[kv[0]] = [float(kv[1])]
+        for feval in self.fevals:
+            bst_eval = model.eval_set(self.evals, epoch, feval)
+            res = [x.split(":") for x in bst_eval.split()]
+            for kv in res[1:]:
+                res_dict[kv[0]] = [float(kv[1])]
 
         eval_res = []
         keys = list(res_dict.keys())
-        keys.sort(key=lambda x: x if metric_shortname not in x else "a" + x)
+        keys.sort(key=lambda x: x if self.metric_shortname not in x else "a" + x)
         for key in keys:
             v = res_dict[key]
             eval_res.append([key] + v)
 
         ##### print eval result #####
-        if not isinstance(verbose_eval, bool) and verbose_eval and i % verbose_eval == 0:
-            infos = ["XGB iter: %3d" % i]
+        if (
+            not isinstance(self.verbose_eval, bool)
+            and self.verbose_eval
+            and epoch % self.verbose_eval == 0
+        ):
+            infos = ["XGB iter: %3d" % epoch]
             for item in eval_res:
                 if "null" in item[0]:
                     continue
                 infos.append("%s: %.6f" % (item[0], item[1]))
 
             logger.debug("\t".join(infos))
-            if log_file:
-                with open(log_file, "a") as fout:
+            if self.log_file:
+                with open(self.log_file, "a") as fout:
                     fout.write("\t".join(infos) + "\n")
 
         ##### choose score and do early stopping #####
         score = None
         for item in eval_res:
-            if item[0] == metric:
+            if item[0] == self.metric:
                 score = item[1]
                 break
         assert score is not None
 
-        best_score = state["best_score"]
-        best_iteration = state["best_iteration"]
-        maximize_score = state["maximize_score"]
+        best_score = self.state["best_score"]
+        best_iteration = self.state["best_iteration"]
+        maximize_score = self.state["maximize_score"]
+
         if (maximize_score and score > best_score) or (not maximize_score and score < best_score):
-            msg = "[%d] %s" % (env.iteration, "\t".join([_fmt_metric(x) for x in eval_res]))
-            state["best_msg"] = msg
-            state["best_score"] = score
-            state["best_iteration"] = env.iteration
+            msg = "[%d] %s" % (epoch, "\t".join([_fmt_metric(x) for x in eval_res]))
+            self.state["best_msg"] = msg
+            self.state["best_score"] = score
+            self.state["best_iteration"] = epoch
             # save the property to attributes, so they will occur in checkpoint.
-            if env.model is not None:
-                env.model.set_attr(
-                    best_score=str(state["best_score"]),
-                    best_iteration=str(state["best_iteration"]),
-                    best_msg=state["best_msg"],
+            if model is not None:
+                model.set_attr(
+                    best_score=str(self.state["best_score"]),
+                    best_iteration=str(self.state["best_iteration"]),
+                    best_msg=self.state["best_msg"],
                 )
-        elif env.iteration - best_iteration >= stopping_rounds:
-            best_msg = state["best_msg"]
-            if verbose_eval and env.rank == 0:
+        elif epoch - best_iteration >= self.stopping_rounds:
+            best_msg = self.state["best_msg"]
+            if self.verbose_eval:
                 logger.debug("XGB stopped. Best iteration: %s ", best_msg)
-            raise EarlyStopException(best_iteration)
+            return True
 
-    return callback
+        return False