hendrikstrobelt / sentimenter_minimal_hai

Minimal starting point for rapid prototyping interactive Human-AI tools

A minimal starting point for rapid prototyping interactive Human-AI tools

Just a quick connection of the huggingface pipeline for sentiment analysis (see here) with a small visual interface made with D3js.

by Hendrik Strobelt (MIT-IBM Watson AI Lab)

Perequisits

Create new Conda environment:

conda create -n hugging python=3.6
conda activate hugging

install flask:

conda install -c anaconda flask

install huggingface (and pytorch):

conda install pytorch -c pytorch
pip install transformers

Run

conda activate hugging
python server.py

visit http://localhost:8888

sentimenter_minimal_hai's Issues

Readme instructions may needs updation

Running python server.py command throws following error:
Neither PyTorch nor TensorFlow >= 2.0 have been found.Models won't be available and only tokenizers, configurationand file/data utilities can be used.

Please update readme for tensorflow/ pytorch dependency.

Error on running server.py

Installed tensorflow, but Getting error on running the server
Traceback (most recent call last):
File "/Users/rohitdhamija/opt/miniconda3/envs/hugging/lib/python3.6/site-packages/transformers/configuration_utils.py", line 242, in get_config_dict
raise EnvironmentError
OSError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "server.py", line 17, in
nlp = pipeline('sentiment-analysis')
File "/Users/rohitdhamija/opt/miniconda3/envs/hugging/lib/python3.6/site-packages/transformers/pipelines.py", line 1858, in pipeline
tokenizer = AutoTokenizer.from_pretrained(tokenizer)
File "/Users/rohitdhamija/opt/miniconda3/envs/hugging/lib/python3.6/site-packages/transformers/tokenization_auto.py", line 206, in from_pretrained
config = AutoConfig.from_pretrained(pretrained_model_name_or_path, **kwargs)
File "/Users/rohitdhamija/opt/miniconda3/envs/hugging/lib/python3.6/site-packages/transformers/configuration_auto.py", line 203, in from_pretrained
config_dict, _ = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs)
File "/Users/rohitdhamija/opt/miniconda3/envs/hugging/lib/python3.6/site-packages/transformers/configuration_utils.py", line 251, in get_config_dict
raise EnvironmentError(msg)
OSError: Can't load config for 'distilbert-base-uncased-finetuned-sst-2-english'. Make sure that:

• 'distilbert-base-uncased-finetuned-sst-2-english' is a correct model identifier listed on 'https://huggingface.co/models'

• or 'distilbert-base-uncased-finetuned-sst-2-english' is the correct path to a directory containing a config.json file