yichenzw/coh-mgt-detection

The code implementation of the paper CoCo: Coherence-Enhanced Machine-Generated Text Detection Under Low Resource With Contrastive Learning from XJTU.

coh-mgt-detection's Issues

huggingface dataset load error

I got an error when loading the data set using huggingface, as follows:

datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 2 new columns ({'split', 'index'})

This happened while the json dataset builder was generating data using

hf://datasets/ZachW/MGTDetect_CoCo/gpt3.5-davinci3/gpt3.5-Mixed-davinci3/gpt3.5_mixed_1000_train.jsonl (at revision aa49f92a8667f5a704ff576c728765c236940c6c)

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

My code：

from datasets import load_dataset
dataset = load_dataset("ZachW/MGTDetect_CoCo")

Please add pip format requirements.txt

When I configure the environment, conda create --name coco --file requirements.txt goes wrong. I think a pip format requirements.txt would be helpful.

How to run on multiple GPUs?

I want to use multiple GPUs to speed up the detection process.
In run_detector.py line 1161 args.n_gpu = 1.
How can I run run_detector.py on multiple GPUs, hoping to get your reply.

How to use this model to detect a new sentence?

I have a trained model already. How can i use this model to detect a sentence created by which model?I am looking forward to your reply. Thanks a lot.

Where is the **_Graph.jsonl ?

In the dataset, i don't find the **_Graph.jsonl. But i find the problem in the picture. How can i solve the problem? Thanks for your reply.
微信截图_20240129175640