Download upstreams/candi/scripts/gen_text8_cont_embed_sweep.sh from jyliuAI/diffusion_baseline: direct link, hf CLI and curl.
- Browser
- Download file 474 Bytes
-
https://huggingface.co/jyliuAI/diffusion_baseline/resolve/main/upstreams/candi/scripts/gen_text8_cont_embed_sweep.sh
- Command line
-
hf download hf://jyliuAI/diffusion_baseline/upstreams/candi/scripts/gen_text8_cont_embed_sweep.sh
-
curl -L -o gen_text8_cont_embed_sweep.sh https://huggingface.co/jyliuAI/diffusion_baseline/resolve/main/upstreams/candi/scripts/gen_text8_cont_embed_sweep.sh
474 Bytes
| CUDA_VISIBLE_DEVICES=$1 python main_for_sweeps.py --config-name=config_text8 mode=sample_eval algo=candi \ | |
| eval.checkpoint_path=/data/remote_cache/patrick/discrete_diffusion/openwebtext-train/text8-pure-cont-embed-1024/checkpoints/last.ckpt \ | |
| algo.is_embed=True \ | |
| algo.mixed_coeff=0 \ | |
| model=small \ | |
| model.length=1024 \ | |
| algo.min_percentile=.05 \ | |
| algo.max_percentile=.4\ | |
| algo.pure_continuous=true; | |