All Questions
9 questions
0
votes
0
answers
178
views
TypeError: stat: path should be string, bytes, os.PathLike or integer, not NoneType. Huggingface model locally
I'm trying to run a LM locally I don't really have much knowledge of huggingface.
What I did is creating account there then creating a token that can read/write.
I created a project did pip install ...
- 338
0
votes
1
answer
161
views
batch is NoneType from trl dataloader for PPOTrainer, huggingface python
My code is essentially just a copy and paste of the following huggingface article at https://huggingface.co/docs/trl/en/ppo_trainer. I have my own reward model that I have trained, and now I'm trying ...
- 51
0
votes
2
answers
744
views
isinstance() arg 2 must be a type, a tuple of types, or a union
I'm getting an error message when trying to train my model, but some reason its giving me the same message every time I alter it.
Here is the code:
# Define training arguments
training_args = ...
- 11
5
votes
0
answers
2k
views
Google Colab: cuBLAS, cuFFT, and cuDNN factory registration warnings
I was playing around with Wizard-Vicuna in colab. I tried the 13B model and the other day it was working fine. I can change the prompts and it would generate alright. However, yesterday when I tried ...
- 51
1
vote
0
answers
388
views
Hugging Face Evaluator Module - Evaluation Speed seems very slow?
I've been learning how to fine-train HuggingFace text classification models using their tutorials and some other information online.
So far I've managed to train a model on a custom dataset - import ...
- 61
2
votes
1
answer
1k
views
Low score and wrong answer for Flan-T5-XXL "question-answering" task
I'm trying to run Flan-T5-XXL model for a "question-answering" task.
Here's how I loaded and executed the model:
model_id = "~/Downloads/test_LLM/flan-t5-xxl"
tokenizer = ...
- 169
0
votes
1
answer
2k
views
Cannot get DataCollator to prepare tf dataset
I’m trying to follow this tutorial to fine-tune bert for a NER task using my own dataset. https://www.philschmid.de/huggingface-transformers-keras-tf. Below is my shortened code, and the error due to ...
- 1
1
vote
0
answers
249
views
"ValueError: initial_value must be specified." error when compiling bert for text classification
I'm following the hugging face tutorial for the sequence classification and while trying to fine-tune 'distilbert-base-multilingual-cased' I'm having the following error when running the model.compile(...
- 1,743
0
votes
0
answers
592
views
ValueError: `Checkpoint` was expecting model to be a trackable object (an object derived from `Trackable`), got RobertaForSequenceClassification
I am training a text classification model using RoBERTa. (https://huggingface.co/siebert/sentiment-roberta-large-english)
I use google colab for running the code.
I am facing a valueError i have trid ...
- 21