GPT2 model from Arabic Wikipedia dataset based on gpt2-small (using Fastai2).
Intended uses & limitations
How to use
An example is provided in this
colab notebook
.
Both text and poetry (fine-tuned model) generation are included.
Limitations and bias
GPT2-small-arabic (trained on Arabic Wikipedia) has several limitations in terms of coverage (Arabic Wikipeedia quality, no diacritics) and training performance.
Use as demonstration or proof of concepts but not as production code.
Training data
This pretrained model used the Arabic Wikipedia dump (around 900 MB).
Training procedure
Training was done using
Fastai2
library on Kaggle, using free GPU.
Eval results
Final perplexity reached was 72.19, loss: 4.28, accuracy: 0.307
BibTeX entry and citation info
@inproceedings{Abed Khooli,
year={2020}
}
Runs of akhooli gpt2-small-arabic on huggingface.co
390
Total runs
0
24-hour runs
16
3-day runs
-2
7-day runs
-308
30-day runs
More Information About gpt2-small-arabic huggingface.co Model
gpt2-small-arabic huggingface.co
gpt2-small-arabic huggingface.co is an AI model on huggingface.co that provides gpt2-small-arabic's model effect (), which can be used instantly with this akhooli gpt2-small-arabic model. huggingface.co supports a free trial of the gpt2-small-arabic model, and also provides paid use of the gpt2-small-arabic. Support call gpt2-small-arabic model through api, including Node.js, Python, http.
gpt2-small-arabic huggingface.co is an online trial and call api platform, which integrates gpt2-small-arabic's modeling effects, including api services, and provides a free online trial of gpt2-small-arabic, you can try gpt2-small-arabic online for free by clicking the link below.
akhooli gpt2-small-arabic online free url in huggingface.co:
gpt2-small-arabic is an open source model from GitHub that offers a free installation service, and any user can find gpt2-small-arabic on GitHub to install. At the same time, huggingface.co provides the effect of gpt2-small-arabic install, users can directly use gpt2-small-arabic installed effect in huggingface.co for debugging and trial. It also supports api for free installation.