ComfyUI Node: Sampler & Scheduler Crawler (CRT)

Sampler & Scheduler Crawler (CRT):

The SamplerSchedulerCrawler node is designed to facilitate the selection and management of samplers and schedulers within a creative AI environment. Its primary function is to dynamically choose a sampler and a scheduler based on provided seeds, allowing for a systematic exploration of different combinations. This node is particularly beneficial for users who wish to experiment with various sampling and scheduling techniques to achieve optimal results in their AI-generated art. By leveraging the seeds, it ensures that the selection process is both repeatable and customizable, providing a structured approach to testing and refining AI models. The node also supports a "sweep" mode, which enables users to automatically iterate through all possible combinations of samplers and schedulers, thus enhancing the creative exploration process.

Sampler & Scheduler Crawler (CRT) Input Parameters:

sampler_seed

The sampler_seed is an integer input that determines the selection of the sampler from the available list. It acts as an index, ensuring that the choice of sampler is consistent and repeatable across different runs. The seed value is modulated by the number of available samplers to select the current sampler. The minimum value is 0, the maximum is 18,446,744,073,709,551,615 (0xFFFFFFFFFFFFFFFF), and the default is 0. This parameter is crucial for users who want to maintain consistency in their sampling process or explore different samplers systematically.

scheduler_seed

The scheduler_seed is an integer input that influences the selection of the scheduler. Similar to the sampler_seed, it is used to index into the list of available schedulers, ensuring a consistent and repeatable selection process. The seed is adjusted based on the number of full cycles through the sampler list if the sweep mode is enabled. The minimum value is 0, the maximum is 18,446,744,073,709,551,615 (0xFFFFFFFFFFFFFFFF), and the default is 0. This parameter is essential for users aiming to explore different scheduling strategies or maintain consistency in their scheduling process.

sweep_schedulers

The sweep_schedulers is a boolean parameter that, when enabled, allows the scheduler index to increment only after the sampler list has completed a full cycle. This feature is particularly useful for users who want to test every possible combination of samplers and schedulers sequentially by simply incrementing the sampler_seed. The default value is False, with options labeled as "Sweep ON" and "Sweep OFF". This parameter provides a powerful tool for comprehensive exploration and testing of AI model configurations.

Sampler & Scheduler Crawler (CRT) Output Parameters:

sampler_name

The sampler_name output provides the name of the currently selected sampler. This output is crucial for users to understand which sampler is being used in the current configuration, allowing for better tracking and analysis of results.

scheduler

The scheduler output gives the name of the currently selected scheduler. This information is vital for users to identify the scheduling strategy applied in the current setup, aiding in the evaluation and comparison of different scheduling techniques.

sampler_str

The sampler_str output is a string representation of the selected sampler, providing a human-readable format that can be used for logging or display purposes.

scheduler_str

The scheduler_str output is a string representation of the selected scheduler, offering a clear and concise way to communicate the chosen scheduling method.

selection_label

The selection_label output combines the names of the selected sampler and scheduler into a single string, formatted as "sampler

• scheduler". This label is useful for quick reference and documentation of the current configuration.

list_info

The list_info output is a detailed string that includes information about the sweep mode status and lists all available samplers and schedulers, highlighting the currently selected ones. This output is invaluable for users who need a comprehensive overview of the available options and the current selections.

Sampler & Scheduler Crawler (CRT) Usage Tips:

• Utilize the sweep_schedulers option to systematically explore all combinations of samplers and schedulers, which can be particularly useful for discovering optimal configurations.
• Keep track of the sampler_seed and scheduler_seed values to ensure repeatability in your experiments, allowing you to reproduce successful configurations easily.

Sampler & Scheduler Crawler (CRT) Common Errors and Solutions:

No available samplers

• Explanation: This error occurs when the list of available samplers is empty.
• Solution: Ensure that the sampler list is correctly populated. If not, the node defaults to using "euler" as a fallback sampler.

No available schedulers

• Explanation: This error arises when the list of available schedulers is empty.
• Solution: Verify that the scheduler list is correctly populated. If not, the node defaults to using "normal" as a fallback scheduler.

Invalid seed value

• Explanation: This error can occur if the seed values exceed the maximum allowed range.
• Solution: Ensure that the sampler_seed and scheduler_seed values are within the specified range of 0 to 18,446,744,073,709,551,615.