[Bug] 'WanTransformerInferAdaCaching' object has no attribute 'infer_modulation'

[endymion-ni]

Description

Briefly describe the bug you encountered.

Meet Error while test AdaCache

'WanTransformerInferAdaCaching' object has no attribute 'infer_modulation'

Steps to Reproduce

run script modified by run_wan_i2v_tea.sh:

python -m lightx2v.infer
--model_cls wan2.1
--task i2v
--model_path $model_path
--config_json ${lightx2v_path}/configs/caching/adacache/wan_i2v_ada.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_result_path ${lightx2v_path}/save_results/output_lightx2v_wan_i2v_ada.mp4

Expected Result

Generate Video with AdaCache speedup.

Actual Result

Get Error report.

Environment Information

Log Information

Please provide relevant error logs or debugging information.

Exception has occurred: AttributeError 'WanTransformerInferAdaCaching' object has no attribute 'infer_modulation' File "/data0/project/LightX2V/lightx2v/models/networks/wan/infer/feature_caching/transformer_infer.py", line 366, in infer_calculating shift_msa, scale_msa, gate_msa, c_shift_msa, c_scale_msa, c_gate_msa = self.infer_modulation(weights.blocks[block_idx].compute_phases[0], embed0) ^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/networks/wan/infer/feature_caching/transformer_infer.py", line 330, in infer x = self.infer_calculating(weights, grid_sizes, embed, x, embed0, seq_lens, freqs, context) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/networks/wan/model.py", line 443, in _infer_cond_uncond x = self.transformer_infer.infer(self.transformer_weights, grid_sizes, embed, x, embed0, seq_lens, freqs, context) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/utils/custom_compiler.py", line 47, in wrapper return state["original_func"](self, *args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/networks/wan/model.py", line 410, in infer noise_pred_cond = self._infer_cond_uncond(inputs, infer_condition=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/runners/default_runner.py", line 155, in run_segment self.model.infer(self.inputs) File "/data0/project/LightX2V/lightx2v/utils/memory_profiler.py", line 18, in wrapper result = func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/runners/default_runner.py", line 291, in run_main latents = self.run_segment(total_steps=total_steps) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/utils/profiler.py", line 77, in sync_wrapper return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/models/runners/default_runner.py", line 366, in run_pipeline gen_video_final = self.run_main() ^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/utils/profiler.py", line 77, in sync_wrapper return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/data0/project/LightX2V/lightx2v/infer.py", line 109, in main runner.run_pipeline(input_info) File "/data0/project/LightX2V/lightx2v/infer.py", line 118, in main() AttributeError: 'WanTransformerInferAdaCaching' object has no attribute 'infer_modulation'

Additional Information

If there is any other information that can help solve the problem, please add it here.

I have debuged some bug before:

1. lightx2v/models/networks/wan/model.py Line 435 you have packed five parameters in pre_infer_out，maybe the right code could be：

        if self.config["feature_caching"] == "Ada":
            embed = pre_infer_out.embed
            grid_sizes = pre_infer_out.grid_sizes
            x = pre_infer_out.x
            embed0 = pre_infer_out.embed0
            seq_lens = pre_infer_out.seq_lens
            freqs = pre_infer_out.freqs
            context = pre_infer_out.context
            x = self.transformer_infer.infer(self.transformer_weights, grid_sizes, embed, x, embed0, seq_lens, freqs, context)
        else:
            x = self.transformer_infer.infer(self.transformer_weights, pre_infer_out)

2. lightx2v/models/networks/wan/infer/feature_caching/transformer_infer.py line 324 An unexcepted variable self.infer_conditional used here. I guess you copy taylorSeer's code but forget to remove this code？