From 8993b9aff731d67dc4cc9a4e6671694196e71817 Mon Sep 17 00:00:00 2001 From: Leonard Date: Wed, 3 May 2023 19:44:45 +0200 Subject: [PATCH] Corrected some errors in the README as well as wrong default train_num for the HF connector --- README.md | 2 +- refinery/adapter/transformers.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 3a26947..d42fa1c 100644 --- a/README.md +++ b/README.md @@ -158,7 +158,7 @@ Transformers are great, but often times, you want to finetune them for your down ```python from refinery.adapter import transformers -dataset, mapping = transformers.build_dataset(client, "headline", "__clickbait") +dataset, mapping, index = transformers.build_classification_dataset(client, "headline", "__clickbait") ``` From here, you can follow the [finetuning example](https://huggingface.co/docs/transformers/training) provided in the official Hugging Face documentation. A next step could look as follows: diff --git a/refinery/adapter/transformers.py b/refinery/adapter/transformers.py index 3269a23..31b61b3 100644 --- a/refinery/adapter/transformers.py +++ b/refinery/adapter/transformers.py @@ -9,7 +9,7 @@ def build_classification_dataset( client: Client, sentence_input: str, classification_label: str, - num_train: Optional[int] = 100, + num_train: Optional[int] = None, ): """Build a classification dataset from a refinery client and a config string useable for HuggingFace finetuning.