Warning: data is not aligned! This can lead to a speed loss

[jadams777]

When I run the inference example:

python inference.py \
    --data_root "./checkpoints/ditto_trt_Ampere_Plus" \
    --cfg_pkl "./checkpoints/ditto_cfg/v0.4_hubert_cfg_trt.pkl" \
    --audio_path "./example/audio.wav" \
    --source_path "./example/image.png" \
    --output_path "./tmp/result.mp4" 

I get:

Warning: data is not aligned! This can lead to a speed loss

Here is the full result:

In file included from /root/miniconda3/envs/ditto/lib/python3.10/site-packages/numpy/_core/include/numpy/ndarraytypes.h:1909,
                 from /root/miniconda3/envs/ditto/lib/python3.10/site-packages/numpy/_core/include/numpy/ndarrayobject.h:12,
                 from /root/miniconda3/envs/ditto/lib/python3.10/site-packages/numpy/_core/include/numpy/arrayobject.h:5,
                 from /root/.pyxbld/temp.linux-x86_64-cpython-310/workspace/ditto-talkinghead/core/utils/blend/blend.c:1259:
/root/miniconda3/envs/ditto/lib/python3.10/site-packages/numpy/_core/include/numpy/npy_1_7_deprecated_api.h:17:2: warning: #warning "Using deprecated NumPy API, disable it with " "#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION" [-Wcpp]
   17 | #warning "Using deprecated NumPy API, disable it with " \
      |  ^~~~~~~
==================== setup kwargs ====================
max_size <class 'int'> 1920
template_n_frames <class 'int'> -1
crop_scale <class 'float'> 2.3
crop_vx_ratio <class 'int'> 0
crop_vy_ratio <class 'float'> -0.125
crop_flag_do_rot <class 'bool'> True
smo_k_s <class 'int'> 13
emo <class 'numpy.ndarray'> (600, 8)
eye_f0_mode <class 'bool'> False
ch_info <class 'dict'>
overlap_v2 <class 'int'> 10
fix_kp_cond <class 'int'> 1
fix_kp_cond_dim <class 'list'> [0, 202]
sampling_timesteps <class 'int'> 50
online_mode <class 'bool'> False
v_min_max_for_clip <class 'numpy.ndarray'> (4, 265)
smo_k_d <class 'int'> 3
N_d <class 'int'> -1
use_d_keys <class 'NoneType'> None
relative_d <class 'bool'> True
drive_eye <class 'NoneType'> None
delta_eye_arr <class 'numpy.ndarray'> (15, 63)
delta_eye_open_n <class 'int'> 0
fade_type <class 'str'> d0
fade_out_keys <class 'list'> ['exp']
flag_stitching <class 'bool'> True
overall_ctrl_info <class 'dict'> {'delta_pitch': 2}
==================================================
dit: 6it [00:01,  3.99it/s]
writer: 1it [00:13, 13.18s/it][swscaler @ 0x66baa00] Warning: data is not aligned! This can lead to a speed loss
writer: 394it [00:36, 10.72it/s]
ffmpeg -loglevel error -y -i "./tmp/result.mp4.tmp.mp4" -i "./example/audio.wav" -map 0:v -map 1:a -c:v copy -c:a aac "./tmp/result.mp4"
./tmp/result.mp4
(ditto) root@d8b454fef605:/workspace/ditto-talkinghead# cd tmp
(ditto) root@d8b454fef605:/workspace/ditto-talkinghead/tmp# ls
result.mp4  result.mp4.tmp.mp4
(ditto) root@d8b454fef605:/workspace/ditto-talkinghead/tmp# runpodctl send result.mp4
Runpod config file not found, please run `runpodctl config` to create it
7240-theory-common-modem-13
Sending 'result.mp4' (5.0 MB)    
Code is: 7240-theory-common-modem-13
On the other computer run

runpodctl receive 7240-theory-common-modem-13

Sending (->174.65.156.232:57965)
result.mp4 100% |████████████████████| (5.2/5.2 MB, 72 MB/s)

[li-tianqi]

@jadams777
This warning message appears when using imageio to write video files via the FFmpeg backend. This is related to memory alignment during image data processing and is a common warning from FFmpeg's internal scaling component (swscaler).

As discussed in this SuperUser thread, this warning does not affect the correctness of the output and is mainly related to performance optimizations. In most practical scenarios, the impact on speed is negligible.