|
--- |
|
library_name: peft |
|
base_model: shpotes/codegen-350M-mono |
|
datasets: |
|
- flytech/python-codes-25k |
|
pipeline_tag: text-generation |
|
tags: |
|
- code |
|
license: mit |
|
--- |
|
|
|
## How to Get Started with the Model |
|
```python |
|
import torch |
|
from transformers import AutoModelForCausalLM, BitsAndBytesConfig |
|
from peft import PeftModel, PeftConfig |
|
|
|
config = PeftConfig.from_pretrained("yamete4/codegen-350M-mono-QLoRa-flytech") |
|
model = AutoModelForCausalLM.from_pretrained("shpotes/codegen-350M-mono", |
|
quantization_config=BitsAndBytesConfig(config),) |
|
peft_model = PeftModel.from_pretrained(model, "yamete4/codegen-350M-mono-QLoRa-flytech") |
|
|
|
text = "Help me manage my subscriptions!?" |
|
|
|
inputs = tokenizer(text, return_tensors="pt").to(0) |
|
outputs = perf_model.generate(inputs.input_ids, max_new_tokens=250, do_sample=False) |
|
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=False)) |
|
|
|
|
|
``` |
|
|
|
### Framework versions |
|
|
|
- PEFT 0.9.0 |