Instructions to use bigcode/astraios-1b-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use bigcode/astraios-1b-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("bigcode/starcoderbase-1b") model = PeftModel.from_pretrained(base_model, "bigcode/astraios-1b-lora") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -99,11 +99,8 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
|
|
| 99 |
|
| 100 |
peft_checkpoint = "bigcode/astraios-1b-lora"
|
| 101 |
checkpoint = "bigcode/starcoderbase-1b"
|
| 102 |
-
model =
|
| 103 |
-
|
| 104 |
-
model.merge_and_unload()
|
| 105 |
-
except:
|
| 106 |
-
pass
|
| 107 |
device = "cuda" # for GPU usage or "cpu" for CPU usage
|
| 108 |
|
| 109 |
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
|
|
|
| 99 |
|
| 100 |
peft_checkpoint = "bigcode/astraios-1b-lora"
|
| 101 |
checkpoint = "bigcode/starcoderbase-1b"
|
| 102 |
+
model = AutoModelForCausalLM.from_pretrained(checkpoint)
|
| 103 |
+
model = PeftModel.from_pretrained(model, peft_checkpoint)
|
|
|
|
|
|
|
|
|
|
| 104 |
device = "cuda" # for GPU usage or "cpu" for CPU usage
|
| 105 |
|
| 106 |
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|