ComfyUI Node: Load Frame Pair From Dataset

## LoadFramePairFromDataset:
The `LoadFramePairFromDataset` node is designed to facilitate the loading of a pair of frames from a specified dataset. This node is particularly useful for tasks that require the comparison or sequential processing of frames, such as video frame interpolation, animation, or temporal consistency checks. By leveraging this node, you can easily retrieve the current frame and the previous frame based on a given seed value, ensuring that you have the necessary frame data for your creative or analytical tasks. This node simplifies the process of accessing and managing frame data, making it an essential tool for AI artists working with frame-based datasets.

## LoadFramePairFromDataset Input Parameters:
### frame_dataset
This parameter specifies the dataset from which frames will be loaded. The dataset should be a collection of frame paths that the node can access and retrieve frames from. The dataset is expected to be in the format of `FRAME_DATASET`.

### seed
The `seed` parameter determines the starting point for frame retrieval. It is an integer value that specifies which frame to load from the dataset. The seed value helps in ensuring reproducibility and consistency in frame selection. The default value is 0, with a minimum of 0 and a maximum of 9999999999.

### total_frames
This parameter indicates the total number of frames available in the dataset. It is an integer value that helps in bounding the frame selection process, ensuring that the frame number does not exceed the available frames in the dataset. The default value is 0, with a minimum of 0 and a maximum of 9999999999.

### fit_into
This parameter is used to specify the dimensions into which the frames should be resized or fitted. It ensures that the loaded frames conform to the desired size, which can be crucial for maintaining consistency in subsequent processing steps.

## LoadFramePairFromDataset Output Parameters:
### current_frame
This output parameter provides the current frame loaded from the dataset based on the specified seed value. The frame is returned as an image tensor, which can be used for further processing or analysis.

### previous_frame
This output parameter provides the previous frame relative to the current frame, based on the seed value. It is also returned as an image tensor, allowing for comparative or sequential operations.

### seed
The `seed` value is returned as an integer, indicating the frame number that was used to retrieve the current and previous frames. This helps in tracking and verifying the frame selection process.

## LoadFramePairFromDataset Usage Tips:
- Ensure that your `frame_dataset` is correctly formatted and accessible to avoid any issues with frame loading.
- Use the `seed` parameter to control and reproduce specific frame selections, which can be useful for debugging or consistent processing.
- Adjust the `fit_into` parameter to match the desired output dimensions, ensuring that all frames are uniformly sized for subsequent operations.

## LoadFramePairFromDataset Common Errors and Solutions:
### "Found 0 frames in path <file_path>"
- **Explanation**: This error occurs when the specified dataset path does not contain any frames.
- **Solution**: Verify that the `file_path` parameter points to a valid directory containing frame files. Ensure that the directory is not empty and that the frame files are in a supported format.

### "Frame source not found at <source_path>"
- **Explanation**: This error indicates that the specified source path for the dataset does not exist.
- **Solution**: Check the `source_path` to ensure it is correct and that the directory exists. If the path is incorrect, update it to point to the correct location of your frame dataset.

### "IndexError: list index out of range"
- **Explanation**: This error occurs when the seed value exceeds the number of available frames in the dataset.
- **Solution**: Ensure that the `seed` value is within the range of available frames by checking the `total_frames` parameter. Adjust the seed value accordingly to avoid exceeding the dataset bounds.