Upload scripts/train-sakthai-1.5b-v2.py with huggingface_hub

scripts/train-sakthai-1.5b-v2.py

@@ -14,12 +14,12 @@
 """
 SakThai 1.5B v2 — improved tool-calling fine-tune.
 QLoRA + rsLoRA + all-linear targets + completion-only loss.
-Based on the proven 0.5B v2 recipe, upgraded for Qwen2.5-1.5B-Instruct.
+Trains on v7 + v8 + v8-gap-fill combined (3,440 examples).
 
 Usage on HF Jobs:
   hf jobs uv run --flavor a10g-small --timeout 6h --secrets HF_TOKEN train-sakthai-1.5b-v2.py
 
-Or from URL (after uploading to HF Hub):
+Or from URL:
   hf jobs uv run --flavor a10g-small --timeout 6h --secrets HF_TOKEN
       https://huggingface.co/Nanthasit/sakthai-kaggle-notebooks/resolve/main/train-sakthai-1.5b-v2.py
 """
@@ -32,8 +32,8 @@ from trl import SFTConfig, SFTTrainer
 
 BASE_MODEL   = "Qwen/Qwen2.5-1.5B-Instruct"
 HF_USER      = "Nanthasit"
-ADAPTER_REPO = f"{HF_USER}/sakthai-context-1.5b-tools-v2"
-MERGED_REPO  = f"{HF_USER}/sakthai-context-1.5b-merged-v2"
+ADAPTER_REPO = f"{HF_USER}/sakthai-plus-1.5b-lora"
+MERGED_REPO  = f"{HF_USER}/sakthai-plus-1.5b"
 MAX_SEQ_LEN  = 2048
 HF_TOKEN     = os.environ.get("HF_TOKEN")
 assert HF_TOKEN, "Set HF_TOKEN secret: --secrets HF_TOKEN"
@@ -70,16 +70,20 @@ def to_text(ex):
     )
     return {"text": text}
 
-main = load_dataset(f"{HF_USER}/sakthai-combined-v7", split="train")
+# v10: normalized v7+v8 + argument augmentation (2972 examples)
+main = load_dataset(f"{HF_USER}/sakthai-combined-v10", split="train")
 train_data = main.map(to_text, remove_columns=main.column_names)
+
+# irrelevance supplement
 try:
     supp = load_dataset(f"{HF_USER}/sakthai-irrelevance-supplement", split="train")
     supp_text = supp.map(to_text, remove_columns=supp.column_names)
     train_data = concatenate_datasets([train_data, supp_text])
 except Exception as e:
-    print("irrelevance-supplement unavailable, continuing without it:", e)
+    print("irrelevance-supplement unavailable:", e)
 
-eval_raw = load_dataset(f"{HF_USER}/sakthai-combined-v7", split="test")
+# Use v7 test split for eval (no test split in v10 yet)
+eval_raw = load_dataset(f"{HF_USER}/sakthai-combined-v7", "test", split="train")
 eval_data = eval_raw.map(to_text, remove_columns=eval_raw.column_names)
 print(f"train={len(train_data)}  eval={len(eval_data)}")