Add Sinhala GPT-2 model for text generation This model was fine-tuned on a custom Sinhala dataset to improve text generation for Sinhala language.

from transformers import Trainer, TrainingArguments

# Define your model
model = GPT2LMHeadModel.from_pretrained('gpt2')

# Prepare your custom dataset here
train_dataset = CustomDataset('your_dataset.txt')

training_args = TrainingArguments(
output_dir='./results',
num_train_epochs=3,
per_device_train_batch_size=8,
save_steps=10_000,
save_total_limit=2,
)

trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset,
)

trainer.train()

README.md

@@ -1,3 +1,26 @@
----
-license: cc-by-nd-4.0
----
+---
+license: cc-by-nd-4.0
+language:
+- si
+- en
+- ta
+metrics:
+- code_eval
+- chrf
+- character
+base_model:
+- openai-community/gpt2
+new_version: openai-community/gpt2
+pipeline_tag: text-classification
+library_name: asteroid
+tags:
+- code
+- climate
+- biology
+- text-generation-inference
+- art
+- legal
+---
+---
+license: cc-by-nd-4.0
+---0