Downloads · 30 days
22
39% of all-time downloads
jahnaviym/crate-clap-lora
crate-clap-lora is a feature extraction model from jahnaviym. Use it when you need embeddings to search or compare text. It is set up for peft. The card lists the license as mit.
LoRA adapter over laion/clap-htsat-unfused fine-tuned so producer terms base CLAP barely knows — boom-bap, tape-saturated, rimshot, reese bass — pull the right audio. Part of Crate, sound-native search for music produ…
Downloads · 30 days
22
39% of all-time downloads
All-time downloads
56
Public
Repo size
16.4 MB
Likes
0
Public
Click a slice to open those files.
.safetensors16.1 MB · 100%
From the Hugging Face model README
LoRA adapter over laion/clap-htsat-unfused fine-tuned so producer terms base CLAP
barely knows — boom-bap, tape-saturated, rimshot, reese bass — pull the right
audio. Part of Crate, sound-native
search for music producers (hum / drop a track / describe it → one embedding space).
| metric | base CLAP | fine-tuned |
|---|---|---|
| recall@1 | 0.210 | 0.405 |
| recall@10 | 0.746 | 0.951 |
| recall@5 | 0.580 | 0.868 |
from transformers import ClapModel, ClapProcessor
from peft import PeftModel
base = "laion/clap-htsat-unfused"
model = PeftModel.from_pretrained(ClapModel.from_pretrained(base), "jahnaviym/crate-clap-lora").merge_and_unload()
proc = ClapProcessor.from_pretrained(base)
# proc(text=[...]) / proc(audios=[...], sampling_rate=48000) → get_text/audio_features
See the repo for the full pipeline and eval.