Unable to Create Labeled Video after Analyzation #2796
Description
Is there an existing issue for this?
- I have searched the existing issues
Bug description
After Following steps from the Single Animal User Guide, I was able to Extract & Label Frames, train a network, and finally analyze a video. When trying to create a labeled video though, I get and error - "No unfiltered data file found in C:\dir for video and scorer".
This happens right after a successful output from the analyze function.
Operating System
Win11
DeepLabCut version
3.0.0rc6
DeepLabCut mode
single animal
Device type
gpu
NVIDIA-SMI 556.12 Driver Version: 556.12 CUDA Version: 12.5
Steps To Reproduce
Followed Steps from Single animal User Guide in Project documentation,
Created a new project
Extracted Frames
Labeled Frames
Created Training Dataset
Trained Network
Analyzed Videos
Create videos
Relevant log output
Analyzing videos with C:\conda_envs\Cat-004-Midas-2024-11-26\dlc-models-pytorch\iteration-0\Cat-004Nov26-trainset95shuffle1\train\snapshot-best-020.pt
C:\conda_envs\dlcdlc\lib\site-packages\deeplabcut\pose_estimation_pytorch\runners\base.py:80: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
snapshot = torch.load(snapshot_path, map_location=device)
Starting to analyze C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-2-vid1_30fps.mp4
Video metadata:
Overall # of frames: 614
Duration of video [s]: 20.47
fps: 30.0
resolution: w=1920, h=1080
Running pose prediction with batch size 8
100%|████████████████████████████████████████████████████████████████████████████████| 614/614 [02:48<00:00, 3.63it/s]
The videos are analyzed. Now your research can truly start!
You can create labeled videos with 'create_labeled_video'.
If the tracking is not satisfactory for some videos, consider expanding the training set. You can use the function 'extract_outlier_frames' to extract a few representative outlier frames.
Out[3]: 'DLC_Resnet50_Cat-004Nov26shuffle1_snapshot_020'
In [5]: deeplabcut.create_labeled_video(^M
...: "C:\\conda_envs\\Cat-004-Midas-2024-11-26\\config.yaml",^M
...: ["C:\\conda_envs\\Cat-004-Midas-2024-11-26\\videos\\camera-2-vid1_30fps.mp4"],^M
...: videotype=".mp4",^M
...: trailpoints=2,^M
...: draw_skeleton=True,^M
...: save_frames=True^M
...: )
Starting to process video: C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-2-vid1_30fps.mp4
Loading C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-2-vid1_30fps.mp4 and data.
No unfiltered data file found in C:\conda_envs\Cat-004-Midas-2024-11-26\videos for video camera-2-vid1_30fps and scorer DLC_Resnet50_Cat-004Nov26shuffle1_snapshot_020.Anything else?
Config file as follows
# Project definitions (do not edit)
Task: Cat-004
scorer: Midas
date: Nov26
multianimalproject: false
identity:
# Project path (change when moving around)
project_path: C:\conda_envs\Cat-004-Midas-2024-11-26
# Default DeepLabCut engine to use for shuffle creation (either pytorch or tensorflow)
engine: pytorch
# Annotation data set configuration (and individual video cropping parameters)
video_sets:
C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-1-vid1.mp4:
crop: 746, 1094, 506, 754
C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-1-vid1_30fps.mp4:
crop: 812, 1071, 526, 754
C:\conda_envs\Cat-004-Midas-2024-11-26\videos\camera-2-vid1.mp4:
crop: 614, 974, 375, 673
bodyparts:
- front_paw_left
- back_paw_left
- ankle_left
- knee_left
- shoulder_left
- front_paw_right
- back_paw_right
- ankle_right
- knee_right
- shoulder_right
- front_paw_left_back
- back_paw_left_back
- ankle_left_back
- knee_left_back
- hip_bottom_left
- hip_top_left
- front_paw_right_back
- back_paw_right_back
- ankle_right_back
- knee_right_back
- hip_bottom_right
- hip_top_right
- head
- ear_left
- ear_right
- nape
- spine_top
- spine_mid
- spine_bottom
- tail_base
- tail_mid
- tail_tip
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
# Fraction of video to start/stop when extracting frames for labeling/refinement
start: 0
stop: 1
numframes2pick: 20
# Plotting configuration
skeleton:
- - front_paw_left
- back_paw_left
- - ankle_left
- knee_left
- shoulder_left
- - front_paw_right
- back_paw_right
- - ankle_right
- knee_right
- shoulder_right
- - front_paw_left_back
- back_paw_left_back
- - ankle_left_back
- knee_left_back
- hip_bottom_left
- hip_top_left
- - front_paw_right_back
- back_paw_right_back
- - ankle_right_back
- knee_right_back
- hip_bottom_right
- hip_top_right
- - head
- ear_left
- ear_right
- - nape
- spine_top
- spine_mid
- spine_bottom
- - tail_base
- tail_mid
- tail_tip
skeleton_color: black
pcutoff: 0.6
dotsize: 12
alphavalue: 0.7
colormap: rainbow
# Training,Evaluation and Analysis configuration
TrainingFraction:
- 0.95
iteration: 0
default_net_type: resnet_50
default_augmenter: default
snapshotindex: -1
detector_snapshotindex: -1
batch_size: 8
detector_batch_size: 1
# Cropping Parameters (for analysis and outlier frame detection)
cropping: false
#if cropping is true for analysis, then set the values here:
x1: 0
x2: 640
y1: 277
y2: 624
# Refinement configuration (parameters from annotation dataset configuration also relevant in this stage)
corner2move2:
- 50
- 50
move2corner: true
# Conversion tables to fine-tune SuperAnimal weights
SuperAnimalConversionTables:Screenshot of videos folder withing project
I tried with two different videos to make sure one wasn't corrupted or something. I tried changing the names of both and recreating, but was not working. Any help would be greatly appreciated.
Code of Conduct
- I agree to follow this project's Code of Conduct