[Task 1/95] Current/Best: 125.21/ 165.29 GFLOPS | Progress: (48/1000) | 93.94 sTraceback (most recent call last):
File "ssd_tvm_zjj.py", line 151, in <module>
autotvm.callback.log_to_file(tuning_option["tuning_records"]),
File "/home/dl3/ZJJ/TVM_source/tvm/python/tvm/autotvm/tuner/xgboost_tuner.py", line 105, in tune
super(XGBTuner, self).tune(*args, **kwargs)
File "/home/dl3/ZJJ/TVM_source/tvm/python/tvm/autotvm/tuner/tuner.py", line 169, in tune
self.update(inputs, results)
File "/home/dl3/ZJJ/TVM_source/tvm/python/tvm/autotvm/tuner/model_based_tuner.py", line 281, in update
self.cost_model.fit(self.xs, self.ys, self.plan_size)
File "/home/dl3/ZJJ/TVM_source/tvm/python/tvm/autotvm/tuner/xgboost_cost_model.py", line 210, in fit
verbose_eval=self.log_interval,
File "/home/dl3/ZJJ/TVM_source/tvm/python/tvm/autotvm/tuner/xgboost_cost_model.py", line 462, in custom_callback
from xgboost.core import EarlyStopException
ImportError: cannot import name 'EarlyStopException' from 'xgboost.core' (/home/dl3/anaconda3/envs/WH/lib/python3.7/site-packages/xgboost/core.py)
Done.
解决办法:pip install xgboost==1.5.0
参考网址: Auto Scheduler (Ansor) XGBoost cost model import error - Troubleshooting - Apache TVM Discuss