GPT2 trained to generate ЗНО (Ukrainian exam SAT type of thing) essays
Generated texts are not very cohesive yet but I'm working on it.
The Hosted inference API outputs (on the right) are too short for some reason. Trying to fix it.
Use the code from the example below. The model takes "ZNOTITLE: your essay title" inputs.
Example of usage:
from transformers import AlbertTokenizer, GPT2LMHeadModel
tokenizer = AlbertTokenizer.from_pretrained("kyryl0s/gpt2-uk-zno-edition")
model = GPT2LMHeadModel.from_pretrained("kyryl0s/gpt2-uk-zno-edition")
input_ids = tokenizer.encode("ZNOTITLE: За яку працю треба більше поважати людину - за фізичну чи інтелектуальну?", add_special_tokens=False, return_tensors='pt')
outputs = model.generate(
input_ids,
do_sample=True,
num_return_sequences=1,
max_length=250
)
for i, out in enumerate(outputs):
print("{}: {}".format(i, tokenizer.decode(out)))
- Downloads last month
- 28
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.