# Token classification metric **URL:** \ **Category:** 🤗Transformers\ **Created:** [October 3, 2024, 9:59pm UTC](https://discuss.huggingface.co/t/token-classification-metric/110041 "2024-10-03T21:59:29Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![AbdulrahmanAhmed](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/abdulrahmanahmed/32/32380_2.png) [@AbdulrahmanAhmed](https://discuss.huggingface.co/u/AbdulrahmanAhmed)\ **Post date:** [October 3, 2024, 9:59pm UTC](https://discuss.huggingface.co/t/token-classification-metric/110041/1 "2024-10-03T21:59:29Z")
i tried to make a token classification task by fine-tuning distilbert and training with tensorflow. and I used accuracy as a metric in the compile method. i found the loss was decreasing but the accuracy didn’t increase so I realise that accuracy metric is not appropriate for NER tasks I searched and I found that f1-score is a good choose but I got an error ValueError Traceback (most recent call last) in \() ----\> 1 history = model.fit(train\_dataset.take(1) , validation\_data = val\_dataset , epochs = training\_args[“epochs”], callbacks= metric\_callback) 5 frames /usr/local/lib/python3.10/dist-packages/evaluate/module.py in add\_batch(self, predictions, references, \*\*kwargs) 544 f"Input references: {summarize\_if\_long\_list(references)}" 545 ) → 546 raise ValueError(error\_msg) from None 547 548 def add(self, \*, prediction=None, reference=None, \*\*kwargs): ValueError: Predictions and/or references don’t match the expected format. Expected format: {‘predictions’: Sequence(feature=Value(dtype=‘string’, id=‘label’), length=-1, id=‘sequence’), ‘references’: Sequence(feature=Value(dtype=‘string’, id=‘label’), length=-1, id=‘sequence’)}, so I need an metric for NER task