From 788e42ea55e80682221e10b7f7cf56daa3c102fe Mon Sep 17 00:00:00 2001 From: Yizhou Chi Date: Mon, 30 Sep 2024 16:06:48 +0800 Subject: [PATCH] update model list --- metagpt/prompts/task_type.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/metagpt/prompts/task_type.py b/metagpt/prompts/task_type.py index 97666874d..74286a28f 100644 --- a/metagpt/prompts/task_type.py +++ b/metagpt/prompts/task_type.py @@ -35,8 +35,8 @@ The current task is about feature engineering. when performing it, please adhere MODEL_TRAIN_PROMPT = """ The current task is about training a model, please ensure high performance: - For tabular datasets - you have access to XGBoost, CatBoost, random forest, extremely randomized trees, k-nearest neighbors, linear regression, etc. -- For image datasets - you have access to ResNet, VGG, Inception, MobileNet, DenseNet, EfficientNet, etc. -- For text datasets - you have access to BERT, GPT-2, RoBERTa, DistilBERT, T5, etc. +- For image datasets - you have access to Swin Transformer, ViT, ResNet, EfficientNet, etc. +- For text datasets - you have access to Electra, DeBERTa, GPT-2, BERT, etc. - Avoid the use of SVM because of its high training time. - Keep in mind that your user prioritizes results and is highly focused on model performance. So, when needed, feel free to use models of any complexity to improve effectiveness, such as XGBoost, CatBoost, etc. - If non-numeric columns exist, perform label encode together with all steps.