Issue opening a file created in version 1.4 in 1.5.2 and an issue training on new videos in 1.5.2 using a model from 1.4

[BobbyLangston]

Hello.

I am trying to use SLEAP to track sperm cells and have had great success and accuracy using the 1.4 version. This software is great but unfortunately I have had issues continuing the training using the 1.5.2 version.

System Info

Software versions:
SLEAP: 1.5.2
Numpy: 2.3.5
Python: 3.13.11
OS: Windows-11-10.0.26200-SP0

Issue One

The first issue is when I try to open a SLEAP file created in 1.4 using 1.5.2, I get the following error:

TypeError: _path_exists: path should be string, bytes, os.PathLike or integer, not list

This is what the full read out is in the terminal:

Happy SLEAPing! :)
Traceback (most recent call last):
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 345, in openProject
    self.execute(OpenProject, filename=filename, first_open=first_open)
    ~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 299, in execute
    command().execute(context=self, params=kwargs)
    ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 195, in execute
    self.do_with_signal(context, params)
    ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 219, in do_with_signal
    cls.do_action(context, params)
    ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 795, in do_action
    context.loadProjectFile(filename)
    ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 330, in loadProjectFile
    self.execute(LoadProjectFile, filename=filename)
    ~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 299, in execute
    command().execute(context=self, params=kwargs)
    ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 193, in execute
    okay = self.ask(context, params)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\gui\commands.py", line 764, in ask
    labels = load_labels_video_search(
        filename, video_search=gui_video_callback
    )
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\sleap_io_adaptors\lf_labels_utils.py", line 898, in load_labels_video_search
    abort = video_search(labels.videos)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\sleap_io_adaptors\lf_labels_utils.py", line 817, in video_callback
    missing = [not os.path.exists(filename) for filename in filenames]

I am not sure what is going on but if anyone has any insight that would be great. Please let me know if more information on this issue would be helpful.

Issue Two

The second issue occurs when I try to continue a training session on a new video using a model from 1.4. The same skeleton is being used. I am using the animal top-down approach and the error only occurs when I get to training the centered instance.

RuntimeError: The size of tensor a (7) must match the size of tensor b (37) at non-singleton dimension 1

The full error is:

Run Path: C:/Users/rlscott4/Desktop/models/022225_model_1_260123_140112.centroid.n=1
Finished training centroid.
Resetting monitor window.
Polling: C:/Users/rlscott4/Desktop/models/022225_model_1_260123_140550.centered_instance.n=1\viz\validation.*.png
Start training centered_instance...
['sleap-nn-train', '--config-name', '260123_140550_config', '--config-dir', 'C:\\Users\\rlscott4\\AppData\\Local\\Temp\\tmpurdmewuh']
Input config:

data_config:
  train_labels_path:
  - C:/Users/rlscott4/Desktop/test_soft_substrate.v001.slp
  val_labels_path: null
  validation_fraction: 0.1
  test_file_path: null
  provider: LabelsReader
  user_instances_only: true
  data_pipeline_fw: torch_dataset
  cache_img_path: null
  use_existing_imgs: false
  delete_cache_imgs_after_training: true
  preprocessing:
    ensure_rgb: false
    ensure_grayscale: false
    max_height: 512
    max_width: 512
    scale: 0.4
    crop_size: null
    min_crop_size: 100
  use_augmentations_train: true
  augmentation_config:
    intensity:
      uniform_noise_min: 0.0
      uniform_noise_max: 1.0
      uniform_noise_p: 0.0
      gaussian_noise_mean: 5.0
      gaussian_noise_std: 1.0
      gaussian_noise_p: 0.0
      contrast_min: 0.5
      contrast_max: 2.0
      contrast_p: 0.0
      brightness_min: 0.0
      brightness_max: 2.0
      brightness_p: 0.0
    geometric:
      rotation_min: -180.0
      rotation_max: 180.0
      scale_min: 0.9
      scale_max: 1.1
      translate_width: 0.0
      translate_height: 0.0
      affine_p: 1.0
      erase_scale_min: 0.0001
      erase_scale_max: 0.01
      erase_ratio_min: 1.0
      erase_ratio_max: 1.0
      erase_p: 0.0
      mixup_lambda_min: 0.01
      mixup_lambda_max: 0.05
      mixup_p: 0.0
  skeletons: []
model_config:
  init_weights: default
  pretrained_backbone_weights: E:/reorientation_project/models/022225_model_1250222_153925.centered_instance.n=32/best_model.h5
  pretrained_head_weights: E:/reorientation_project/models/022225_model_1250222_153925.centered_instance.n=32/best_model.h5
  backbone_config:
    unet:
      in_channels: 1
      kernel_size: 3
      filters: 24
      filters_rate: 2.0
      max_stride: 16
      stem_stride: null
      middle_block: true
      up_interpolate: true
      stacks: 1
      convs_per_block: 2
      output_stride: 4
    convnext: null
    swint: null
  head_configs:
    single_instance: null
    centroid: null
    centered_instance:
      confmaps:
        part_names:
        - top
        - center
        - side_left
        - side_right
        - neck
        - neck_2
        - neck_3
        anchor_part: center
        sigma: 0.75
        output_stride: 4
        loss_weight: 1.0
    bottomup: null
    multi_class_bottomup: null
    multi_class_topdown: null
  total_params: null
trainer_config:
  train_data_loader:
    batch_size: 4
    num_workers: 0
    shuffle: true
  val_data_loader:
    batch_size: 4
    num_workers: 0
    shuffle: false
  model_ckpt:
    save_top_k: 1
    save_last: false
  trainer_devices: null
  trainer_device_indices: null
  trainer_accelerator: auto
  profiler: null
  trainer_strategy: auto
  enable_progress_bar: true
  min_train_steps_per_epoch: 200
  train_steps_per_epoch: 200
  visualize_preds_during_training: true
  keep_viz: false
  max_epochs: 200
  seed: null
  use_wandb: false
  save_ckpt: true
  ckpt_dir: C:/Users/rlscott4/Desktop/models
  run_name: 022225_model_1_260123_140550.centered_instance.n=1
  resume_ckpt_path: null
  wandb:
    entity: ''
    project: ''
    name: ''
    save_viz_imgs_wandb: false
    api_key: ''
    wandb_mode: null
    prv_runid: ''
    group: ''
    current_run_id: null
  optimizer_name: Adam
  optimizer:
    lr: 0.0001
    amsgrad: false
  lr_scheduler:
    step_lr: null
    reduce_lr_on_plateau:
      threshold: 1.0e-06
      threshold_mode: abs
      cooldown: 3
      patience: 5
      factor: 0.5
      min_lr: 1.0e-08
  early_stopping:
    min_delta: 1.0e-08
    patience: 10
    stop_training_on_plateau: true
  online_hard_keypoint_mining:
    online_mining: false
    hard_to_easy_ratio: 2.0
    min_hard_keypoints: 2
    max_hard_keypoints: null
    loss_scale: 5.0
  zmq:
    controller_port: 9000
    controller_polling_timeout: 10
    publish_port: 9001
name: ''
description: ''
sleap_nn_version: 0.0.5
filename: ''

2026-01-23 14:05:59 | INFO | sleap_nn.train:run_training:24 | Started training at: 2026-01-23 14:05:59.258573
2026-01-23 14:05:59 | INFO | sleap_nn.training.model_trainer:_setup_train_val_labels:216 | Creating train-val split...
2026-01-23 14:05:59 | INFO | sleap_nn.training.model_trainer:_setup_train_val_labels:261 | # Train Labeled frames: 1
2026-01-23 14:05:59 | INFO | sleap_nn.training.model_trainer:_setup_train_val_labels:262 | # Val Labeled frames: 1
2026-01-23 14:05:59 | INFO | sleap_nn.training.model_trainer:setup_config:510 | Setting up config...
2026-01-23 14:06:00 | INFO | sleap_nn.legacy_models:get_keras_first_layer_channels:57 | Found first layer 'model_weights/stack0_enc0_conv0/stack0_enc0_conv0/kernel:0' with 1 input channels
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:849 | Setting up for training...
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:_setup_model_ckpt_dir:573 | Setting up model ckpt dir: `C:/Users/rlscott4/Desktop/models/022225_model_1_260123_140550.centered_instance.n=1`...
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:864 | Setting up visualization train and val datasets...
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:868 | Setting up Trainer...
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:_setup_loggers_callbacks:645 | Setting up callbacks and loggers...
2026-01-23 14:06:00 | INFO | sleap_nn.training.callbacks:__init__:228 | Training controller subscribed to: tcp://127.0.0.1:9000 (topic: )
2026-01-23 14:06:00 | INFO | sleap_nn.training.callbacks:__init__:287 | ProgressReporterZMQ publishing to tcp://127.0.0.1:9001 for ''
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:897 | Trainer devices: auto
INFO: GPU available: True (cuda), used: True
INFO:lightning.pytorch.utilities.rank_zero:GPU available: True (cuda), used: True
INFO: TPU available: False, using: 0 TPU cores
INFO:lightning.pytorch.utilities.rank_zero:TPU available: False, using: 0 TPU cores
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:950 | Training on 1 device(s)
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:951 | Training on cuda:0 accelerator
2026-01-23 14:06:00 | INFO | sleap_nn.training.model_trainer:train:955 | Setting up lightning module for centered_instance model...
2026-01-23 14:06:00 | INFO | sleap_nn.training.lightning_modules:__init__:197 | Loading backbone weights from `E:/reorientation_project/models/022225_model_1250222_153925.centered_instance.n=32/best_model.h5` ...
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/CenteredInstanceConfmapsHead/CenteredInstanceConfmapsHead/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/CenteredInstanceConfmapsHead/CenteredInstanceConfmapsHead/kernel:0
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:272 | Successfully mapped 28/30 legacy weights to PyTorch parameters
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:365 | Successfully loaded 28/28 weights from legacy model
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:376 | Verifying weight assignments...
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:439 | ✓ All weight assignments verified successfully
2026-01-23 14:06:01 | INFO | sleap_nn.training.lightning_modules:__init__:226 | Loading head weights from `E:/reorientation_project/models/022225_model_1250222_153925.centered_instance.n=32/best_model.h5` ...
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec0_s16_to_s8_refine_conv0/stack0_dec0_s16_to_s8_refine_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec0_s16_to_s8_refine_conv0/stack0_dec0_s16_to_s8_refine_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec0_s16_to_s8_refine_conv1/stack0_dec0_s16_to_s8_refine_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec0_s16_to_s8_refine_conv1/stack0_dec0_s16_to_s8_refine_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec1_s8_to_s4_refine_conv0/stack0_dec1_s8_to_s4_refine_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec1_s8_to_s4_refine_conv0/stack0_dec1_s8_to_s4_refine_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec1_s8_to_s4_refine_conv1/stack0_dec1_s8_to_s4_refine_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_dec1_s8_to_s4_refine_conv1/stack0_dec1_s8_to_s4_refine_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc0_conv0/stack0_enc0_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc0_conv0/stack0_enc0_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc0_conv1/stack0_enc0_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc0_conv1/stack0_enc0_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc1_conv0/stack0_enc1_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc1_conv0/stack0_enc1_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc1_conv1/stack0_enc1_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc1_conv1/stack0_enc1_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc2_conv0/stack0_enc2_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc2_conv0/stack0_enc2_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc2_conv1/stack0_enc2_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc2_conv1/stack0_enc2_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc3_conv0/stack0_enc3_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc3_conv0/stack0_enc3_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc3_conv1/stack0_enc3_conv1/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc3_conv1/stack0_enc3_conv1/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc5_middle_expand_conv0/stack0_enc5_middle_expand_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc5_middle_expand_conv0/stack0_enc5_middle_expand_conv0/kernel:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc6_middle_contract_conv0/stack0_enc6_middle_contract_conv0/bias:0
2026-01-23 14:06:01 | WARNING | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:263 | No matching PyTorch parameter found for model_weights/stack0_enc6_middle_contract_conv0/stack0_enc6_middle_contract_conv0/kernel:0
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:map_legacy_to_pytorch_layers:272 | Successfully mapped 2/30 legacy weights to PyTorch parameters
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:365 | Successfully loaded 2/2 weights from legacy model
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:376 | Verifying weight assignments...
2026-01-23 14:06:01 | INFO | sleap_nn.legacy_models:load_legacy_model_weights:439 | ✓ All weight assignments verified successfully
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:959 | Backbone model: UNet(
  (encoders): ModuleList(
    (0): Encoder(
      (encoder_stack): ModuleList(
        (0): SimpleConvBlock(
          (blocks): Sequential(
            (stack0_enc0_conv0): Conv2d(1, 24, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc0_act0_relu): ReLU()
            (stack0_enc0_conv1): Conv2d(24, 24, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc0_act1_relu): ReLU()
          )
        )
        (1): SimpleConvBlock(
          (blocks): Sequential(
            (stack0_enc1_pool): MaxPool2dWithSamePadding(kernel_size=2, stride=2, padding=same, dilation=1, ceil_mode=False)
            (stack0_enc1_conv0): Conv2d(24, 48, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc1_act0_relu): ReLU()
            (stack0_enc1_conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc1_act1_relu): ReLU()
          )
        )
        (2): SimpleConvBlock(
          (blocks): Sequential(
            (stack0_enc2_pool): MaxPool2dWithSamePadding(kernel_size=2, stride=2, padding=same, dilation=1, ceil_mode=False)
            (stack0_enc2_conv0): Conv2d(48, 96, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc2_act0_relu): ReLU()
            (stack0_enc2_conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc2_act1_relu): ReLU()
          )
        )
        (3): SimpleConvBlock(
          (blocks): Sequential(
            (stack0_enc3_pool): MaxPool2dWithSamePadding(kernel_size=2, stride=2, padding=same, dilation=1, ceil_mode=False)
            (stack0_enc3_conv0): Conv2d(96, 192, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc3_act0_relu): ReLU()
            (stack0_enc3_conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_enc3_act1_relu): ReLU()
          )
        )
        (4): Sequential(
          (stack0_enc4_last_pool): MaxPool2dWithSamePadding(kernel_size=2, stride=2, padding=same, dilation=1, ceil_mode=False)
        )
      )
    )
  )
  (decoders): ModuleList(
    (0): Decoder(
      (decoder_stack): ModuleList(
        (0): SimpleUpsamplingBlock(
          (blocks): Sequential(
            (stack0_dec0_s16_to_s8_interp_bilinear): Upsample(scale_factor=2.0, mode='bilinear')
            (stack0_dec0_s16_to_s8_refine_conv0): Conv2d(576, 192, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_dec0_s16_to_s8_refine_conv0_act_relu): ReLU()
            (stack0_dec0_s16_to_s8_refine_conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_dec0_s16_to_s8_refine_conv1_act_relu): ReLU()
          )
        )
        (1): SimpleUpsamplingBlock(
          (blocks): Sequential(
            (stack0_dec1_s8_to_s4_interp_bilinear): Upsample(scale_factor=2.0, mode='bilinear')
            (stack0_dec1_s8_to_s4_refine_conv0): Conv2d(288, 96, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_dec1_s8_to_s4_refine_conv0_act_relu): ReLU()
            (stack0_dec1_s8_to_s4_refine_conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=same)
            (stack0_dec1_s8_to_s4_refine_conv1_act_relu): ReLU()
          )
        )
      )
    )
  )
  (middle_blocks): ModuleList(
    (0): SimpleConvBlock(
      (blocks): Sequential(
        (stack0_enc5_middle_expand_conv0): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=same)
        (stack0_enc5_middle_expand_act0_relu): ReLU()
      )
    )
    (1): SimpleConvBlock(
      (blocks): Sequential(
        (stack0_enc6_middle_contract_conv0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=same)
        (stack0_enc6_middle_contract_act0_relu): ReLU()
      )
    )
  )
)
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:960 | Head model: ModuleList(
  (0): Sequential(
    (CenteredInstanceConfmapsHead): Sequential(
      (0): Conv2d(96, 7, kernel_size=(1, 1), stride=(1, 1), padding=same)
      (1): Identity()
    )
  )
)
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:962 | Total model parameters: 4310863
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:967 | Input image shape: torch.Size([1, 1, 112, 112])
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:1021 | Finished trainer set up. [1.0s]
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:1024 | Starting training loop...
C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\callbacks\model_checkpoint.py:881: Checkpoint directory C:\Users\rlscott4\Desktop\models\022225_model_1_260123_140550.centered_instance.n=1 exists and is not empty.
INFO: LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
INFO:lightning.pytorch.accelerators.cuda:LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┓
┃   ┃ Name                     ┃ Type              ┃ Params ┃ Mode  ┃ FLOPs ┃
┡━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━┩
│ 0 │ model                    │ Model             │  4.3 M │ train │     0 │
│ 1 │ instance_peaks_inf_layer │ FindInstancePeaks │      0 │ train │     0 │
└───┴──────────────────────────┴───────────────────┴────────┴───────┴───────┘
Trainable params: 4.3 M
Non-trainable params: 0
Total params: 4.3 M
Total estimated model params size (MB): 17
Modules in train mode: 66
Modules in eval mode: 0
Total FLOPs: 0
C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\connectors\data_connector.p
y:434: The 'val_dataloader' does not have many workers which may be a bottleneck. Consider increasing the value of the
`num_workers` argument` to `num_workers=15` in the `DataLoader` to improve performance.
C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\loss.py:616: UserWarning: Using a
target size (torch.Size([1, 37, 28, 28])) that is different to the input size (torch.Size([1, 7, 28, 28])). This will
likely lead to incorrect results due to broadcasting. Please ensure they have the same size.
  return F.mse_loss(input, target, reduction=self.reduction)

2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:1037 | Finished training loop. [0.0 min]
2026-01-23 14:06:01 | INFO | sleap_nn.training.model_trainer:train:1064 | Deleting viz folder at C:\Users\rlscott4\Desktop\models\022225_model_1_260123_140550.centered_instance.n=1\viz...
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "C:\Users\rlscott4\.local\bin\sleap-nn-train.exe\__main__.py", line 10, in <module>
    sys.exit(train())
             ~~~~~^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1485, in __call__
    return self.main(*args, **kwargs)
           ~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1406, in main
    rv = self.invoke(ctx)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 1269, in invoke
    return ctx.invoke(self.callback, **ctx.params)
           ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\click\core.py", line 824, in invoke
    return callback(*args, **kwargs)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap\cli.py", line 87, in train
    run_training(cfg)
    ~~~~~~~~~~~~^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\train.py", line 27, in run_training
    trainer.train()
    ~~~~~~~~~~~~~^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\training\model_trainer.py", line 1026, in train
    self.trainer.fit(
    ~~~~~~~~~~~~~~~~^
        self.lightning_model,
        ^^^^^^^^^^^^^^^^^^^^^
    ...<2 lines>...
        ckpt_path=self.config.trainer_config.resume_ckpt_path,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 584, in fit
    call._call_and_handle_interrupt(
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
        self,
        ^^^^^
    ...<6 lines>...
        weights_only,
        ^^^^^^^^^^^^^
    )
    ^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\call.py", line 49, in _call_and_handle_interrupt
    return trainer_fn(*args, **kwargs)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 630, in _fit_impl
    self._run(model, ckpt_path=ckpt_path, weights_only=weights_only)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1079, in _run
    results = self._run_stage()
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1121, in _run_stage
    self._run_sanity_check()
    ~~~~~~~~~~~~~~~~~~~~~~^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\trainer.py", line 1150, in _run_sanity_check
    val_loop.run()
    ~~~~~~~~~~~~^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\utilities.py", line 179, in _decorator
    return loop_run(self, *args, **kwargs)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\evaluation_loop.py", line 146, in run
    self._evaluation_step(batch, batch_idx, dataloader_idx, dataloader_iter)
    ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\loops\evaluation_loop.py", line 441, in _evaluation_step
    output = call._call_strategy_hook(trainer, hook_name, *step_args)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\trainer\call.py", line 329, in _call_strategy_hook
    output = fn(*args, **kwargs)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\lightning\pytorch\strategies\strategy.py", line 412, in validation_step
    return self.lightning_module.validation_step(*args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\sleap_nn\training\lightning_modules.py", line 797, in validation_step
    val_loss = nn.MSELoss()(y_preds, y)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1773, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\module.py", line 1784, in _call_impl
    return forward_call(*args, **kwargs)
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\modules\loss.py", line 616, in forward
    return F.mse_loss(input, target, reduction=self.reduction)
           ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\nn\functional.py", line 3868, in mse_loss
    expanded_input, expanded_target = torch.broadcast_tensors(input, target)
                                      ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
  File "C:\Users\rlscott4\AppData\Roaming\uv\tools\sleap\Lib\site-packages\torch\functional.py", line 77, in broadcast_tensors
    return _VF.broadcast_tensors(tensors)  # type: ignore[attr-defined]
           ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
RuntimeError: The size of tensor a (7) must match the size of tensor b (37) at non-singleton dimension 1

Any information on this would be helpful. Ideally, I would be able to continue using the old model to track new videos in 1.5.

Let me know if any additional information would be helpful.
Thanks, Bobby.

[gitttt-1234]

Hi @BobbyLangston ,

Thanks for reporting!

Issue One: File Loading Error

This seems to be a bug in SLEAP 1.5.2 that occurs when loading projects containing image sequences (multiple image files as frames rather than a video file). The bug should be fixed in the latest pre-release.

Solution: Upgrade to the pre-release version which contains the fix:

uv tool install sleap --prerelease=allow

This will install SLEAP 1.6.0a3 (or later) which includes the fix.

Issue Two: Transfer Learning Error

The model architecture looks correct - the weights transferred successfully from TensorFlow to PyTorch, and the head is outputting 7 channels as expected. However, the error suggests your training data (.slp file) has a skeleton with 37 nodes, which is causing the mismatch. Could you run the following command and share the output so we can verify the skeleton in your labels file?

sio show path/to/slp/file.slp --skeleton

Note: If your file has multiple skeletons, you can fix this with the sleap-io CLI (as shown below). This keeps the most frequently used skeleton and removes the others. See the docs for more options.

# Preview what will be fixed 
sio fix your_file.slp --consolidate-skeletons --dry-run -v
                                                                                                                                                  
# Apply the fix                                                                                                                                 
sio fix your_file.slp --consolidate-skeletons -o fixed.slp  

Let us know if you have any questions!

Thanks,

Divya

[BobbyLangston]

Hello Divya,

Thanks for the response.

Regarding Issue One

Your fix worked for this bug. When I first tried it, however, the version wasn't updated when I would open the GUI. This was fixed by uninstalling and then reinstalling and allowing the prerelease version. The UI in the GUI is a lot better than previous versions. Thank you.

Regarding Issue Two

I ran sio show path/to/slp/file.slp --skeleton, and it returned:

PS C:\Users\rlscott4> sio show "C:\Users\rlscott4\Desktop\test_soft_substrate.v001.slp" --skeleton
╭─ sleap-io ───────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ test_soft_substrate.v001.slp                                                                                         │
│ C:\Users\rlscott4\Desktop                                                                                            │
│                                                                                                                      │
│ Type:     Labels                                                                                                     │
│ Size:     14.7 KB                                                                                                    │
│                                                                                                                      │
│ 2 videos | 1 frame | 1 user instances                                                                                │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
  Full: C:\Users\rlscott4\Desktop\test_soft_substrate.v001.slp

Skeleton Details

Skeleton 0: Skeleton-1 (37 nodes, 40 edges)

Python code:
  nodes = ["top", "center", "neck", "neck_2", "neck_3", "tail_8", "side_2", "tail_1", "tail_4", "side_1", "tail_3", "tail_2", "tail_6", "tail_5", "tail_7", "tail_9", "tail_10", "tail_11", "tail_12", "tail_13", "tail_14", "tail_15", "tail_16", "tail_17", "tail_18", "tail_19", "tail_20", "tail_21", "tail_22", "tail_23", "tail_24", "tail_25", "tail_26", "tail_27", "tail_28", "tail_29", "tail_30"]
  edge_inds = [(0, 1), (0, 9), (0, 6), (1, 9), (1, 6), (1, 2), (2, 3), (3, 4), (4, 7), (5, 15), (6, 2), (7, 11), (8, 13), (9, 2), (10, 8), (11, 10), (12, 14), (13, 12), (14, 5), (15, 16), (16, 17), (17, 18), (18, 19), (19, 20), (20, 21), (21, 22), (22, 23), (23, 24), (24, 25), (25, 26), (26, 27), (27, 28), (28, 29), (29, 30), (30, 31), (31, 32), (32, 33), (33, 34), (34, 35), (35, 36)]

Nodes:

    #   Name
 ───────────────
    0   top
    1   center
    2   neck
    3   neck_2
    4   neck_3
    5   tail_8
    6   side_2
    7   tail_1
    8   tail_4
    9   side_1
   10   tail_3
   11   tail_2
   12   tail_6
   13   tail_5
   14   tail_7
   15   tail_9
   16   tail_10
   17   tail_11
   18   tail_12
   19   tail_13
   20   tail_14
   21   tail_15
   22   tail_16
   23   tail_17
   24   tail_18
   25   tail_19
   26   tail_20
   27   tail_21
   28   tail_22
   29   tail_23
   30   tail_24
   31   tail_25
   32   tail_26
   33   tail_27
   34   tail_28
   35   tail_29
   36   tail_30


Edges:

    #   Source         Destination   Indices
 ─────────────────────────────────────────────
    0   top       ->   center        (0, 1)
    1   top       ->   side_1        (0, 9)
    2   top       ->   side_2        (0, 6)
    3   center    ->   side_1        (1, 9)
    4   center    ->   side_2        (1, 6)
    5   center    ->   neck          (1, 2)
    6   neck      ->   neck_2        (2, 3)
    7   neck_2    ->   neck_3        (3, 4)
    8   neck_3    ->   tail_1        (4, 7)
    9   tail_8    ->   tail_9        (5, 15)
   10   side_2    ->   neck          (6, 2)
   11   tail_1    ->   tail_2        (7, 11)
   12   tail_4    ->   tail_5        (8, 13)
   13   side_1    ->   neck          (9, 2)
   14   tail_3    ->   tail_4        (10, 8)
   15   tail_2    ->   tail_3        (11, 10)
   16   tail_6    ->   tail_7        (12, 14)
   17   tail_5    ->   tail_6        (13, 12)
   18   tail_7    ->   tail_8        (14, 5)
   19   tail_9    ->   tail_10       (15, 16)
   20   tail_10   ->   tail_11       (16, 17)
   21   tail_11   ->   tail_12       (17, 18)
   22   tail_12   ->   tail_13       (18, 19)
   23   tail_13   ->   tail_14       (19, 20)
   24   tail_14   ->   tail_15       (20, 21)
   25   tail_15   ->   tail_16       (21, 22)
   26   tail_16   ->   tail_17       (22, 23)
   27   tail_17   ->   tail_18       (23, 24)
   28   tail_18   ->   tail_19       (24, 25)
   29   tail_19   ->   tail_20       (25, 26)
   30   tail_20   ->   tail_21       (26, 27)
   31   tail_21   ->   tail_22       (27, 28)
   32   tail_22   ->   tail_23       (28, 29)
   33   tail_23   ->   tail_24       (29, 30)
   34   tail_24   ->   tail_25       (30, 31)
   35   tail_25   ->   tail_26       (31, 32)
   36   tail_26   ->   tail_27       (32, 33)
   37   tail_27   ->   tail_28       (33, 34)
   38   tail_28   ->   tail_29       (34, 35)
   39   tail_29   ->   tail_30       (35, 36)


Provenance

  filename: C:\Users\rlscott4\Desktop\test_soft_substrate.v001.slp

Am I right in only seeing one skeleton? I ran the same command on the file I trained the model on and the outputs are identical. I also tried to run the training on the newest version and was met with the same runtime error as before. Running sio fix your_file.slp --consolidate-skeletons --dry-run -v shows:

PS C:\Users\rlscott4> sio fix "C:\Users\rlscott4\Desktop\test_soft_substrate.v001.slp" --consolidate-skeletons --dry-run -v
Loading: C:\Users\rlscott4\Desktop\test_soft_substrate.v001.slp
  2 videos, 1 frames, 1 skeletons, 0 tracks

Analyzing...

✓ Videos: No duplicates found
✓ Skeletons: 1 skeleton(s), all in use
✓ Video Color: All videos consistent
  Sample_Trenton012.avi: setting=grayscale, shape=grayscale
  SpermOnHydrogel.avi: setting=grayscale, shape=grayscale
✓ Predictions: None

Actions:
  → Remove unused tracks
  → Remove empty frames

[DRY RUN - no changes made]

This again makes me think I have only one skeleton.

Please let me know what you think and what I should do next. Thanks for everything so far.

[gitttt-1234]

Hi @BobbyLangston !

Thank you for sharing the logs!

There's only one skeleton, but I think the issue is your data now has 37 nodes, but your old model (that you passed in as pretrained weights) was trained with a data that had only 7 nodes. If you want to train a new model with the 37-node dataset, then you could set pretrained weights paths to null and also set part_names to null -> This would be auto-generated from the slp file.

Let me know if you have any questions!

Thanks,

Divya