Instructions to use albertkingdom/deepseek-coder-7b-text2sql-magicoder-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use albertkingdom/deepseek-coder-7b-text2sql-magicoder-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-coder-7b-instruct-v1.5") model = PeftModel.from_pretrained(base_model, "albertkingdom/deepseek-coder-7b-text2sql-magicoder-lora") - Notebooks
- Google Colab
- Kaggle
Commit History
-
docs: drop underpowered n=20 in-distribution table, rely on Spider for SQL capability
Verified signatureb0293c3
-
docs: fix in-distribution results table, correct comparison claim, add training details
Verified signaturec0c29e8
-
Update quick-start snippet to reference renamed repo
Verified signaturedc4e2b0 -
Fix license metadata (DeepSeek Model License, not apache-2.0) and add dataset credits
Verified signature25d97e7 -
Rewrite model card: correct repo references, mixed-training results, intended use & limitations
Verified signature60d594a -
Replace with SQL+Magicoder mixed-training adapter (better SQL generalization + retained code capability, see experiment_log.md)
Verified signatureb738d48
-
Update adapter: mix Magicoder code data (30% SQL / 70% code) to mitigate code-capability degradation
Verified signature4ae5e5b
-
Add LoRA adapter
f15478b -
initial commit
Verified signaturebfc87ac