>>> from transformers import pipeline, set_seed
>>> generator = pipeline('text-generation', model='Someman/gpt2-medium-ne')
>>> set_seed(42)
>>> generator("उच्च अदालतले बिहीबार दिएको आदेशले", max_length=30, num_return_sequences=5)

[{'generated_text': 'उच्च अदालतले बिहीबार दिएको आदेशले महिनात्रि'},
 {'generated_text': 'उच्च अदालतले बिहीबार दिएको आदेशले बिहानैदे'},
 {'generated_text': 'उच्च अदालतले बिहीबार दिएको आदेशले गिरिजाली'},
 {'generated_text': 'उच्च अदालतले बिहीबार दिएको आदेशले गरेको प्रथम त'},
 {'generated_text': 'उच्च अदालतले बिहीबार दिएको आदेशले कुनै साथी'}]

Here is how to use this model to get the features of a given text in PyTorch:

python

from transformers import GPT2Tokenizer, GPT2Model
tokenizer = GPT2Tokenizer.from_pretrained('Someman/gpt2-medium-ne')
model = GPT2Model.from_pretrained('Someman/gpt2-medium-ne')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)

and in TensorFlow:

python

from transformers import GPT2Tokenizer, TFGPT2Model
tokenizer = GPT2Tokenizer.from_pretrained('Someman/gpt2-medium-ne')
model = TFGPT2Model.from_pretrained('Someman/gpt2-medium-ne')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)

More information needed

Training and evaluation data

Training data contains 197k Nepali sentences.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 32
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 1

Training results

Framework versions

Transformers 4.21.1
Pytorch 1.12.0+cu116
Datasets 2.4.0
Tokenizers 0.12.1