search

catboost / catboost

A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.
https://catboost.ai
Apache License 2.0
8.12k stars 1.19k forks source link

Training continuation for GPU is not yet supported #2736

Open Ethan1X opened 2 months ago

Ethan1X commented 2 months ago

Problem: catboost version: Operating System: CPU: GPU:

Traceback (most recent call last): model.fit(trainx_df, trainy_df, eval_set=valid_pool, use_best_model=True, verbose=False, init_model=model) File "/data/.conda/envs/py3.10/lib/python3.10/site-packages/catboost/core.py", line 5827, in fit return self._fit(X, y, cat_features, text_features, embedding_features, None, sample_weight, None, None, None, None, baseline, File "/data/.conda/envs/py3.10/lib/python3.10/site-packages/catboost/core.py", line 2400, in _fit self._train( File "/data/.conda/envs/py3.10/lib/python3.10/site-packages/catboost/core.py", line 1780, in _train self._object._train(train_pool, test_pool, params, allow_clear_pool, init_model._object if init_model else None) File "_catboost.pyx", line 4833, in _catboost._CatBoost._train File "_catboost.pyx", line 4882, in _catboost._CatBoost._train _catboost.CatBoostError: /src/catboost/catboost/cuda/train_lib/train.cpp:321: Training continuation for GPU is not yet supported

Thank you very much

HamdiBarkous commented 1 month ago

is solving this issue planned in the next releases?

andrey-khropov commented 1 month ago

is solving this issue planned in the next releases?

Not in the next release but it should be eventually supported. But CatBoost is open source so anybody can implement it and send a pull request.