LightX2V icon indicating copy to clipboard operation
LightX2V copied to clipboard

[Bug] 使用SkyReels-V2出现错误

Open xululi0223 opened this issue 4 months ago • 2 comments

Description

我在运行scripts/wan/run_wan_skyreels_v2_df.sh的t2v时,出现错误

Steps to Reproduce

  1. 我首先修改了scripts/wan/run_wan_skyreels_v2_df.sh中T2V命令的--model_cls参数为wan2.1_skyreels_v2_df
  2. 运行该脚本,首先出现了AttributeError: 'EasyDict' object has no attribute 'target_shape'错误,我通过将wan_skyreels_v2_df_runner.py中的run_input_encoder()换为其他函数名可以解决。
  3. 随后再运行该脚本时,报错为:TypeError: WanRunner.run_text_encoder() takes 3 positional arguments but 5 were given

Log Information

Traceback (most recent call last): File "", line 198, in _run_module_as_main File "", line 88, in _run_code File "Lightx2v-main/LightX2V/lightx2v/infer.py", line 82, in main() File "Lightx2v-main/LightX2V/lightx2v/infer.py", line 73, in main runner.run_pipeline() File "Lightx2v-main/LightX2V/lightx2v/models/runners/wan/wan_skyreels_v2_df_runner.py", line 129, in run_pipeline self.run_input_encoder_1() File "Lightx2v-main/LightX2V/lightx2v/models/runners/wan/wan_skyreels_v2_df_runner.py", line 61, in run_input_encoder_1 text_encoder_output = self.run_text_encoder(self.config["prompt"], self.text_encoders, self.config, image_encoder_output) TypeError: WanRunner.run_text_encoder() takes 3 positional arguments but 5 were given

xululi0223 avatar Aug 07 '25 12:08 xululi0223