# Cannot initialize deberta-v3-base tokenizer **URL:** \ **Category:** 🤗Tokenizers\ **Created:** [March 3, 2022, 7:14pm UTC](https://discuss.huggingface.co/t/cannot-initialize-deberta-v3-base-tokenizer/15322 "2022-03-03T19:14:04Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![MagicFlute](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/magicflute/32/4164_2.png) [@MagicFlute](https://discuss.huggingface.co/u/MagicFlute)\ **Post date:** [March 3, 2022, 7:14pm UTC](https://discuss.huggingface.co/t/cannot-initialize-deberta-v3-base-tokenizer/15322/1 "2022-03-03T19:14:04Z")
when calling `tokenizer = AutoTokenizer.from_pretrained("microsoft/deberta-v3-base")` I get a `ValueError: This tokenizer cannot be instantiated. Please make sure you have `sentencepiece` installed in order to use this tokenizer.` Tried conda` install -c conda-forge transformers[sentencepiece]` as well as `conda install -c conda-forge sentencepiece` to no result ---
**Author:** ![simonschoe](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/simonschoe/32/20951_2.png) [@simonschoe](https://discuss.huggingface.co/u/simonschoe)\ **Post date:** [September 21, 2022, 12:08pm UTC](https://discuss.huggingface.co/t/cannot-initialize-deberta-v3-base-tokenizer/15322/2 "2022-09-21T12:08:23Z")
Running `pip install sentencepiece` and restarting the kernel should do the trick. 😉 ---
**Author:** ![carlosaguayo](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/carlosaguayo/32/31818_2.png) [@carlosaguayo](https://discuss.huggingface.co/u/carlosaguayo)\ **Post date:** [October 9, 2022, 4:05pm UTC](https://discuss.huggingface.co/t/cannot-initialize-deberta-v3-base-tokenizer/15322/3 "2022-10-09T16:05:52Z")
> [@simonschoe](#): > > `pip install sentencepiece` Thank you!