Hugging Face's logo
Hugging Face
Search models, datasets, users...
Models
Datasets
Spaces
Docs
Solutions
Pricing




ClassCat
/
gpt2-greek Copied
private
Text Generation
PyTorch
Transformers

cc100
Greek
gpt2
License:
cc-by-sa-4.0
Model card
Files and versions
Community
Settings
gpt2-greek
/
README.md
ClassCat's picture
ClassCat
Create README.md
0765806
less than a minute ago
raw
history
blame
edit
delete
965 Bytes
---
language: el
license: cc-by-sa-4.0
datasets:
- cc100
widget:
- text: "Αυτό είναι ένα"
- text: "Ανοιξα την"
- text: "Ευχαριστώ για το"
- text: "Έχει πολύ καιρό που δεν έχουμε"
---

## Greek GPT2 model (Uncased)

### Prerequisites

transformers==4.19.2

### Model architecture

This model uses approximately half the size of GPT2 base model parameters.

### Tokenizer

Using BPE tokenizer with vocabulary size 50,000.

### Training Data 

* Subset of [CC-100/el](https://data.statmt.org/cc-100/) : Monolingual Datasets from Web Crawl Data
* Subset of [oscar](https://huggingface.co/datasets/oscar)
* [wiki40b/el](https://www.tensorflow.org/datasets/catalog/wiki40b#wiki40bel) (Greek Wikipedia)

### Usage

```python
from transformers import pipeline
generator = pipeline('text-generation', model='ClassCat/gpt2-greek')
generator("Αυτό είναι ένα", max_length=50, num_return_sequences=5)
```