apunkt/doc-to-lora
1
0
Fork 0
mirror of https://github.com/SakanaAI/doc-to-lora.git synced 2026-07-23 17:01:04 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Reset model hooks prior to base feature extraction and base model inference passes

Browse source
This commit is contained in:
Sarvesh 2026-06-09 16:44:04 +05:30
parent 70e60aa6aa
commit 073d909468
1 changed files with 6 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

6
demo.ipynb
View file

@ -500,6 +500,9 @@
" \"\"\"\n",
" Extracts visual features layer-by-layer and generates the dynamic LoRA adapter.\n",
" \"\"\"\n",
" # Ensure any patched LoRA hooks are cleared before feature extraction\n",
" model.reset()\n",
"\n",
" internalize_messages = [\n",
" [\n",
" {\n",
@ -571,6 +574,9 @@
" Runs inference comparing base model (using visual tokens) vs Video2LoRA (zero visual tokens).\n",
" \"\"\"\n",
" # === Part A: Base Model Inference (with visual tokens) ===\n",
" # Ensure any patched LoRA hooks are cleared for a clean base VLM forward pass\n",
" model.reset()\n",
"\n",
" base_messages = [\n",
" [\n",
" {\n",