Try to run the basic sample

[allanchamon]

I am trying to run basic sample and the error has been reported: (.venv) PS C:\Users\allan\Documents\promptbench> & c:/Users/allan/Documents/promptbench/.venv/Scripts/python.exe c:/Users/allan/Documents/promptbench/examples/basic.py Traceback (most recent call last): File "c:\Users\allan\Documents\promptbench\examples\basic.py", line 5, in import promptbench as pb File "C:\Users\allan\Documents\promptbench\promptbench_init_.py", line 6, in from .dataload import DatasetLoader, SUPPORTED_DATASETS, SUPPORTED_DATASETS_VLM File "C:\Users\allan\Documents\promptbench\promptbench\dataload_init_.py", line 4, in from .dataset import * File "C:\Users\allan\Documents\promptbench\promptbench\dataload\dataset.py", line 13, in from datasets import load_dataset File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets_init_.py", line 18, in from .arrow_dataset import Dataset File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets\arrow_dataset.py", line 76, in from .arrow_reader import ArrowReader File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets\arrow_reader.py", line 32, in from .download.download_config import DownloadConfig File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets\download_init_.py", line 9, in from .download_manager import DownloadManager, DownloadMode File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets\download\download_manager.py", line 35, in from ..utils.file_utils import ( File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\datasets\utils\file_utils.py", line 37, in from aiohttp.client_exceptions import ClientError File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\aiohttp_init_.py", line 6, in from .client import ( File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\aiohttp\client.py", line 85, in from .connector import ( File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\aiohttp\connector.py", line 34, in import aiohappyeyeballs File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\aiohappyeyeballs_init_.py", line 3, in from .impl import start_connection File "C:\Users\allan\Documents\promptbench.venv\lib\site-packages\aiohappyeyeballs\impl.py", line 22, in socket_factory: Optional[SocketFactoryType] = None, File "C:\Python39\lib\typing.py", line 243, in inner return func(*args, **kwds) File "C:\Python39\lib\typing.py", line 316, in getitem return self._getitem(self, parameters) File "C:\Python39\lib\typing.py", line 433, in Optional return Union[arg, type(None)] File "C:\Python39\lib\typing.py", line 243, in inner return func(*args, **kwds) File "C:\Python39\lib\typing.py", line 316, in getitem return self._getitem(self, parameters) File "C:\Python39\lib\typing.py", line 421, in Union parameters = _remove_dups_flatten(parameters) File "C:\Python39\lib\typing.py", line 215, in _remove_dups_flatten all_params = set(params) TypeError: unhashable type: 'list'

Any help about that? Thanks in advance.

[kaijiezhu11]

Hi, can you share the code you are running with me?

I tested the following code, it worked well.

import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2")
print(dataset)

BTW, can you please check if the datasets package has been successfully installed?

[allanchamon]

The code is a python source copied from the notebook "basic.ipynb" and is following attached. Thanks.

Em seg., 17 de mar. de 2025 às 17:03, Kaijie Zhu @.***> escreveu:

Hi, can you share the code you are running with me?

I tested the following code, it worked well.

import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2") print(dataset)

BTW, can you please check if the datasets package has been successfully installed?

— Reply to this email directly, view it on GitHub https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7KQ4JWU33OHEPJJOW6RDVL2U4TCBAVCNFSM6AAAAABYSVCNZOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZQG4YDOMRRHE . You are receiving this because you authored the thread.Message ID: @.***> [image: Immortalise]Immortalise left a comment (microsoft/promptbench#94) https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219

Hi, can you share the code you are running with me?

I tested the following code, it worked well.

import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2") print(dataset)

BTW, can you please check if the datasets package has been successfully installed?

— Reply to this email directly, view it on GitHub https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7KQ4JWU33OHEPJJOW6RDVL2U4TCBAVCNFSM6AAAAABYSVCNZOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZQG4YDOMRRHE . You are receiving this because you authored the thread.Message ID: @.***>

[allanchamon]

May I ask for more help?

I am trying to load a pre-trained model locally, it is a derivation of llama 3.1, can I load this model locally and make some attacks against it?

BTW!

Em qui., 20 de mar. de 2025 às 12:41, Allan Chamon Figueiredo < @.***> escreveu:

The code is a python source copied from the notebook "basic.ipynb" and is following attached. Thanks.

Em seg., 17 de mar. de 2025 às 17:03, Kaijie Zhu @.***> escreveu:

Hi, can you share the code you are running with me?

I tested the following code, it worked well.

import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2") print(dataset)

BTW, can you please check if the datasets package has been successfully installed?

— Reply to this email directly, view it on GitHub https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7KQ4JWU33OHEPJJOW6RDVL2U4TCBAVCNFSM6AAAAABYSVCNZOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZQG4YDOMRRHE . You are receiving this because you authored the thread.Message ID: @.***> [image: Immortalise]Immortalise left a comment (microsoft/promptbench#94) https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219

Hi, can you share the code you are running with me?

I tested the following code, it worked well.

import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2") print(dataset)

BTW, can you please check if the datasets package has been successfully installed?

— Reply to this email directly, view it on GitHub https://github.com/microsoft/promptbench/issues/94#issuecomment-2730707219, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7KQ4JWU33OHEPJJOW6RDVL2U4TCBAVCNFSM6AAAAABYSVCNZOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZQG4YDOMRRHE . You are receiving this because you authored the thread.Message ID: @.***>

[allanchamon]

I am using datasets package version 2.15.0, I´ve tried to use the 2.19.2 as well, the same erros happens when I try to execute the code snippet you suggested:

`import promptbench as pb

dataset = pb.DatasetLoader.load_dataset("sst2") print(dataset)`

TypeError: unhashable type: 'list'

[allanchamon]

Solved! I have changed the version of packages below in order to create the conda enviroment correctly: sentencepiece==0.1.99->0.2.0 tokenizers==0.15.0->0.15.1

I have the packages below so that I could use CUDA support: conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia

[github-actions[bot]]

Stale issue message