Upload kaggle/notebook_director_lora.py with huggingface_hub

kaggle/notebook_director_lora.py

@@ -0,0 +1,76 @@
+"""Kaggle notebook body: train the director LoRA.
+
+The HF bundle holds only two files needed here, so fetch them by URL instead
+of snapshot_download: this avoids importing huggingface_hub before the
+transformers 5.x upgrade (a stale already-imported hub breaks transformers'
+`from huggingface_hub.utils import httpx`).
+
+Output lands in /kaggle/working/director_lora_out: adapter/, model_fp16/,
+sample.txt, train_meta.json. The ONNX export is a separate step.
+"""
+import os
+import shutil
+import subprocess
+import sys
+import time
+import urllib.request
+from pathlib import Path
+
+ROOT = Path("/kaggle/working")
+REPO = os.environ.get("BUNDLE_REPO", "idefinity/webfm-eval")
+BASE = f"https://huggingface.co/datasets/{REPO}/resolve/main"
+SFT = ROOT / "director_lora" / "sft.jsonl"          # new protocol, not the old radio set
+TRAIN = ROOT / "kaggle" / "director_lora_train.py"
+OUT = ROOT / "director_lora_out"
+
+FILES = {"director_lora/sft.jsonl": SFT, "kaggle/director_lora_train.py": TRAIN}
+
+
+def log(msg: str) -> None:
+    print(f"[{time.strftime('%H:%M:%S')}] {msg}", flush=True)
+
+
+def fetch(url: str, dst: Path) -> None:
+    dst.parent.mkdir(parents=True, exist_ok=True)
+    for attempt in range(3):
+        try:
+            with urllib.request.urlopen(url, timeout=180) as r, open(dst, "wb") as f:
+                shutil.copyfileobj(r, f)
+            size = dst.stat().st_size
+            if size < 1000:
+                raise RuntimeError(f"only {size} bytes")
+            log(f"fetched {url} -> {dst} ({size/1e6:.2f} MB)")
+            return
+        except Exception as exc:
+            log(f"fetch attempt {attempt + 1} failed: {exc}")
+            time.sleep(4)
+    raise SystemExit(f"cannot fetch {url}")
+
+
+def main() -> None:
+    for rel, dst in FILES.items():
+        fetch(f"{BASE}/{rel}", dst)
+
+    cmd = ("pip install -q --upgrade 'transformers>=5' "
+           "'huggingface_hub>=1.31,<2' peft accelerate")
+    log(f"$ {cmd}")
+    if subprocess.run(cmd, shell=True).returncode != 0:
+        raise SystemExit("dependency install failed")
+
+    env = dict(os.environ, WORK_ROOT=str(ROOT), DATA=str(SFT), LORA_OUT=str(OUT),
+               EPOCHS=os.environ.get("EPOCHS", "2"),
+               LR=os.environ.get("LR", "1e-4"),
+               R=os.environ.get("R", "16"),
+               ALPHA=os.environ.get("ALPHA", "32"),
+               MAX_LEN=os.environ.get("MAX_LEN", "2048"),
+               ACCUM=os.environ.get("ACCUM", "4"),
+               GRAD_CKPT=os.environ.get("GRAD_CKPT", "0"))
+    log(f"training on {SFT}")
+    rc = subprocess.run([sys.executable, str(TRAIN)], env=env).returncode
+    log(f"train rc={rc}; outputs in {OUT}")
+    if rc != 0:
+        raise SystemExit("director LoRA training failed")
+
+
+if __name__ == "__main__":
+    main()