reconstruct config and scripts (#81)

Co-authored-by: gushiqiao <gushiqiao@sensetime.com>

scripts/cogvideox/readme.md

+## todo

scripts/run_cogvideox_t2v.sh → scripts/cogvideox/run_cogvideox_t2v.sh

@@ -34,6 +34,6 @@ python -m lightx2v.infer \
 --model_cls cogvideox \
 --task t2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/cogvideox_t2v.json \
+--config_json ${lightx2v_path}/configs/cogvideox/cogvideox_t2v.json \
 --prompt "A little girl smile." \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_cogvideox_t2v.mp4

scripts/deploy/readme.md

+## todo

scripts/hunyuan/readme.md

+## todo

scripts/run_hunyuan_i2v.sh → scripts/hunyuan/run_hunyuan_i2v.sh

@@ -24,15 +24,15 @@ fi
 export TOKENIZERS_PARALLELISM=false
 
 export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
-export DTYPE=BF16
 export ENABLE_PROFILING_DEBUG=true
 export ENABLE_GRAPH_MODE=false
+export DTYPE=BF16
 
 python -m lightx2v.infer \
 --model_cls hunyuan \
 --task i2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/hunyuan_i2v.json \
+--config_json ${lightx2v_path}/configs/hunyuan/hunyuan_i2v.json \
 --prompt "An Asian man with short hair in black tactical uniform and white clothes waves a firework stick." \
 --image_path ${lightx2v_path}/assets/inputs/imgs/img_1.jpg \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_hy_i2v.mp4

scripts/run_hunyuan_t2v.sh → scripts/hunyuan/run_hunyuan_t2v.sh

@@ -32,6 +32,6 @@ python -m lightx2v.infer \
 --model_cls hunyuan \
 --task t2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/hunyuan_t2v.json \
+--config_json ${lightx2v_path}/configs/hunyuan/hunyuan_t2v.json \
 --prompt "A cat walks on the grass, realistic style." \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_hy_t2v.mp4

scripts/run_hunyuan_t2v_dist.sh → scripts/hunyuan/run_hunyuan_t2v_dist.sh

@@ -31,7 +31,7 @@ torchrun --nproc_per_node=4 ${lightx2v_path}/lightx2v/infer.py \
 --model_cls hunyuan \
 --task t2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/dist/hunyuan_t2v_dist_ulysses.json \
+--config_json ${lightx2v_path}/configs/hunyuan/hunyuan_t2v_dist_ulysses.json \
 --prompt "A cat walks on the grass, realistic style." \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_t2v_dist_ulysses.mp4
 
@@ -39,6 +39,6 @@ torchrun --nproc_per_node=4 ${lightx2v_path}/lightx2v/infer.py \
 --model_cls hunyuan \
 --task t2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/dist/hunyuan_t2v_dist_ring.json \
+--config_json ${lightx2v_path}/configs/hunyuan/hunyuan_t2v_dist_ring.json \
 --prompt "A cat walks on the grass, realistic style." \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_hunyuan_t2v_dist_ring.mp4

scripts/readme.md

+## TODO
+
+#### Server
+#### prompt_enhancer
+#### export ENABLE_GRAPH_MODE=false
+#### export DTYPE=BF16

scripts/run_wan_i2v_advanced_ptq.sh

-#!/bin/bash
-
-# set path and first
-lightx2v_path=
-model_path=
-
-# check section
-if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
-    cuda_devices=0
-    echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
-    export CUDA_VISIBLE_DEVICES=${cuda_devices}
-fi
-
-if [ -z "${lightx2v_path}" ]; then
-    echo "Error: lightx2v_path is not set. Please set this variable first."
-    exit 1
-fi
-
-if [ -z "${model_path}" ]; then
-    echo "Error: model_path is not set. Please set this variable first."
-    exit 1
-fi
-
-export TOKENIZERS_PARALLELISM=false
-
-export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
-export DTYPE=BF16
-export ENABLE_PROFILING_DEBUG=true
-export ENABLE_GRAPH_MODE=false
-
-# =========================
-# load quantization weight and inference
-# =========================
-
-export RUNNING_FLAG=infer
-
-python -m lightx2v.infer \
---model_cls wan2.1 \
---task i2v \
---model_path $model_path \
---config_json ${lightx2v_path}/configs/advanced_ptq/wan_i2v.json \
---prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
---negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
---image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
---save_video_path ${lightx2v_path}/save_results/output_lightx2v_wan_i2v.mp4

scripts/run_wan_i2v_quant_auto.sh

-#!/bin/bash
-
-# set path and first
-lightx2v_path=
-model_path=
-
-# check section
-if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
-    cuda_devices=0
-    echo "Warn: CUDA_VISIBLE_DEVICES is not set, using defalt value: ${cuda_devices}, change at shell script or set env variable."
-    export CUDA_VISIBLE_DEVICES=${cuda_devices}
-fi
-
-if [ -z "${lightx2v_path}" ]; then
-    echo "Error: lightx2v_path is not set. Please set this variable first."
-    exit 1
-fi
-
-if [ -z "${model_path}" ]; then
-    echo "Error: model_path is not set. Please set this variable first."
-    exit 1
-fi
-
-export TOKENIZERS_PARALLELISM=false
-
-export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
-export DTYPE=BF16
-export ENABLE_PROFILING_DEBUG=true
-export ENABLE_GRAPH_MODE=false
-
-python -m lightx2v.infer \
---model_cls wan2.1 \
---task i2v \
---model_path $model_path \
---config_json ${lightx2v_path}/configs/quantization/wan_i2v_quant_auto.json \
---prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
---negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
---image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
---save_video_path ${lightx2v_path}/save_results/output_lightx2v_wan_i2v_tea.mp4

scripts/run_wan_i2v_quant_offline.sh

-#!/bin/bash
-
-# set path and first
-lightx2v_path=
-model_path=
-
-# check section
-if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
-    cuda_devices=0
-    echo "Warn: CUDA_VISIBLE_DEVICES is not set, using defalt value: ${cuda_devices}, change at shell script or set env variable."
-    export CUDA_VISIBLE_DEVICES=${cuda_devices}
-fi
-
-if [ -z "${lightx2v_path}" ]; then
-    echo "Error: lightx2v_path is not set. Please set this variable first."
-    exit 1
-fi
-
-if [ -z "${model_path}" ]; then
-    echo "Error: model_path is not set. Please set this variable first."
-    exit 1
-fi
-
-export TOKENIZERS_PARALLELISM=false
-
-export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
-export DTYPE=BF16
-export ENABLE_PROFILING_DEBUG=true
-export ENABLE_GRAPH_MODE=false
-
-python -m lightx2v.infer \
---model_cls wan2.1 \
---task i2v \
---model_path $model_path \
---config_json ${lightx2v_path}/configs/quantization/wan_i2v_quant_offline.json \
---prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
---negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
---image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
---save_video_path ${lightx2v_path}/save_results/output_lightx2v_wan_i2v_tea.mp4

scripts/run_wan_t2v_sparge.sh

-#!/bin/bash
-
-# set path and first
-#! test with 1.3B
-lightx2v_path=
-model_path=
-
-# check section
-if [ -z "${CUDA_VISIBLE_DEVICES}" ]; then
-    cuda_devices=0
-    echo "Warn: CUDA_VISIBLE_DEVICES is not set, using default value: ${cuda_devices}, change at shell script or set env variable."
-    export CUDA_VISIBLE_DEVICES=${cuda_devices}
-fi
-
-if [ -z "${lightx2v_path}" ]; then
-    echo "Error: lightx2v_path is not set. Please set this variable first."
-    exit 1
-fi
-
-if [ -z "${model_path}" ]; then
-    echo "Error: model_path is not set. Please set this variable first."
-    exit 1
-fi
-
-export TOKENIZERS_PARALLELISM=false
-
-export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
-export DTYPE=BF16
-export ENABLE_PROFILING_DEBUG=true
-export ENABLE_GRAPH_MODE=false
-
-python -m lightx2v.infer \
---model_cls wan2.1 \
---task t2v \
---model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_t2v_sparge.json \
---prompt "Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage." \
---negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
---save_video_path ${lightx2v_path}/save_results/output_lightx2v_wan_t2v.mp4

scripts/server/readme.md

+## todo

scripts/start_api_with_dist_inference.sh → scripts/server/start_api_with_dist_inference.sh

@@ -40,7 +40,7 @@ python -m lightx2v.api_server_dist \
 --model_cls wan2.1 \
 --task i2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_i2v_dist.json \
+--config_json ${lightx2v_path}/configs/wan/wan_i2v_dist.json \
 --port 8000 \
 --start_inference \
 --nproc_per_node 2

scripts/start_multi_servers.sh → scripts/server/start_multi_servers.sh

@@ -40,4 +40,4 @@ python -m lightx2v.api_multi_servers \
     --model_cls wan2.1 \
     --task t2v \
     --model_path $model_path \
-    --config_json ${lightx2v_path}/configs/wan_t2v.json
+    --config_json ${lightx2v_path}/configs/wan/wan_t2v.json

scripts/start_server.sh → scripts/server/start_server.sh

@@ -33,5 +33,5 @@ python -m lightx2v.api_server \
 --model_cls wan2.1 \
 --task t2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_t2v.json \
+--config_json ${lightx2v_path}/configs/wan/wan_t2v.json \
 --port 8000

scripts/wan/readme.md

+## todo

scripts/run_wan_i2v.sh → scripts/wan/run_wan_i2v.sh

@@ -27,13 +27,13 @@ export PYTHONPATH=${lightx2v_path}:$PYTHONPATH
 
 export ENABLE_PROFILING_DEBUG=true
 export ENABLE_GRAPH_MODE=false
-export DTYPE=BF16
+export DTYPE=BF16  # remove this can get high quality video
 
 python -m lightx2v.infer \
 --model_cls wan2.1 \
 --task i2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_i2v.json \
+--config_json ${lightx2v_path}/configs/wan/wan_i2v.json \
 --prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
 --negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
 --image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \

scripts/run_wan_i2v_causvid.sh → scripts/wan/run_wan_i2v_causvid.sh

@@ -32,7 +32,7 @@ python -m lightx2v.infer \
 --task i2v \
 --image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_i2v_causvid.json \
+--config_json ${lightx2v_path}/configs/causvid/wan_i2v_causvid.json \
 --prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
 --negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_wan_i2v_causvid.mp4

scripts/run_wan_i2v_dist.sh → scripts/wan/run_wan_i2v_dist.sh

@@ -32,7 +32,7 @@ torchrun --nproc_per_node=4 -m lightx2v.infer \
 --model_cls wan2.1 \
 --task i2v \
 --model_path $model_path \
---config_json ${lightx2v_path}/configs/wan_i2v_dist.json \
+--config_json ${lightx2v_path}/configs/wan/wan_i2v_dist.json \
 --prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
 --negative_prompt "镜头晃动，色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走" \
 --image_path ${lightx2v_path}/assets/inputs/imgs/img_0.jpg \