chore(main): release 0.9.8

CHANGELOG.md

@@ -1,12 +1,11 @@
 # Changelog
 
-## [0.9.8](https://github.com/microsoft/qlib/compare/v0.9.7...v0.9.8) (2025-11-13)
+## [0.9.8](https://github.com/microsoft/qlib/compare/v0.9.7...v0.9.8) (2025-11-10)
 
 
 ### Bug Fixes
 
 * download orderbook data error ([#1990](https://github.com/microsoft/qlib/issues/1990)) ([136b2dd](https://github.com/microsoft/qlib/commit/136b2ddf9a16e4106d62b8d1336a56273a8abef0))
-* **gbdt:** correct dtrain assignment in finetune() to use Dataset instead of tuple ([#2049](https://github.com/microsoft/qlib/issues/2049)) ([2b41782](https://github.com/microsoft/qlib/commit/2b41782f0cfb81e8cc065f2915b215758a7838ef))
 * **macd:** remove extra division by close in DEA calculation to ensure dimension consistency ([#2046](https://github.com/microsoft/qlib/issues/2046)) ([66c3622](https://github.com/microsoft/qlib/commit/66c36226aafceabe497e5967f67921e5d3c9d497))
 * replace deprecated pandas fillna(method=) with ffill()/bfill() ([#1987](https://github.com/microsoft/qlib/issues/1987)) ([7095e75](https://github.com/microsoft/qlib/commit/7095e755fa57e011f0483d24b45fc5bd5a4deaf8))
 * spelling errors ([#1996](https://github.com/microsoft/qlib/issues/1996)) ([f26b341](https://github.com/microsoft/qlib/commit/f26b3417363410531dbbb39e425bce6cf05528a1))

qlib/contrib/model/gbdt.py

@@ -51,7 +51,7 @@ class LGBModel(ModelFT, LightGBMFInt):
                     w = reweighter.reweight(df)
                 else:
                     raise ValueError("Unsupported reweighter type.")
-                ds_l.append((lgb.Dataset(x.values, label=y, weight=w, free_raw_data=False), key))
+                ds_l.append((lgb.Dataset(x.values, label=y, weight=w), key))
         return ds_l
 
     def fit(
@@ -109,10 +109,8 @@ class LGBModel(ModelFT, LightGBMFInt):
             verbose level
         """
         # Based on existing model and finetune by train more rounds
-        ds_l = self._prepare_data(dataset, reweighter)
-        dtrain, _ = ds_l[0]
-
-        if dtrain.construct().num_data() == 0:
+        dtrain, _ = self._prepare_data(dataset, reweighter)  # pylint: disable=W0632
+        if dtrain.empty:
             raise ValueError("Empty data from dataset, please check your dataset config.")
         verbose_eval_callback = lgb.log_evaluation(period=verbose_eval)
         self.model = lgb.train(