Files
hao-ai-lab-FastVideo/scripts/text_condition/sample_video_257.sh
T
2024-05-25 02:54:40 +00:00

15 lines
547 B
Bash

CUDA_VISIBLE_DEVICES=2 python opensora/sample/sample_t2v.py \
--model_path 257f_512_10node_bs1_lr2e-5_8img/checkpoint-5000/model \
--version 65x512x512 \
--image_size 512 \
--cache_dir "./cache_dir" \
--text_encoder_name DeepFloyd/t5-v1_1-xxl \
--text_prompt examples/prompt_list_0.txt \
--ae CausalVAEModel_4x8x8 \
--ae_path "/dxyl_data02/CausalVAEModel_4x8x8/" \
--save_img_path "./sample_videos_257f_5k_5.5_100s" \
--fps 24 \
--guidance_scale 5.5 \
--num_sampling_steps 100 \
--enable_tiling