Hi , I am using the pre trained model ( horse left to right) to analyze the jumps for the horses, moving from left to right. My workflow was as follows
- Downloaded the pretrained model
- Opened the pretrained model with DLC GUI
- Followed instructions to pretrained
config_pathand used the label tabs to label additional frames and also saved it as .h5 after labelling
- Uploaded the modified pre trained model to google collab . ( with extra labelled frames)
- Merged the data sets from the labeled video with the command
- Created a new training set
- Trained it again
deeplabcut.train_network(path_config_file, shuffle=1, displayiters=1000,saveiters=10000)
- After evaluating the network ,it gave an error of about 4 pixels .
- Analyzing it on similar sequence , we saw that a lot of the body points were missing . ( compared to the video that was just ran on on the pre-trained model)
I was wondering if this is the correct way adding labels to enhance estimation ?