ComfyUI Node
NNT Train Model
A ComfyUI node in NNT Neural Network Toolkit/Models with 27 inputs and 3 outputs.
NNT Train Model
- MODEL
- training_data
- target_data
- hyperparameters
- model
- training_log
- metrics
◄experiment_nametraining_experiment►
◄batch_size32►
◄epochs10►
◄optimizerAdam►
◄learning_rate0.0010►
◄weight_decay0.0001►
◄momentum0.9►
◄loss_functionCrossEntropyLoss►
◄reductionmean►
◄weight_enabledFalse►
◄class_weights[]►
◄margin0.0►
◄use_lr_schedulerTrue►
◄scheduler_typeReduceLROnPlateau►
◄scheduler_step_size3►
◄scheduler_gamma0.10►
◄min_lr0.0000►
◄use_early_stoppingTrue►
◄patience5►
◄min_delta0.0010►
◄reshape_inputFalse►
◄input_reshape_dim[-1]►
◄flatten_inputFalse►
CategoryNNT Neural Network Toolkit/Models
Inputs (27)
| Name | Type | Default | Description |
|---|---|---|---|
| MODEL | MODEL | — | |
| training_data | TENSOR | — | |
| target_data | TENSOR | — | |
| experiment_name | STRING | training_experiment | — |
| batch_size | INT | 321–512 | — |
| epochs | INT | 101–1000 | — |
| optimizer | COMBO | Adam | 12 options: Adadelta, Adagrad, Adam, AdamW, SparseAdam, Adamax, +6 |
| learning_rate | FLOAT | 0.00100.000001–1 | — |
| weight_decay | FLOAT | 0.00010–0.1 | — |
| momentum | FLOAT | 0.90–1 | — |
| loss_function | COMBO | CrossEntropyLoss | 21 options: L1Loss, MSELoss, CrossEntropyLoss, CTCLoss, NLLLoss, PoissonNLLLoss, +15 |
| reduction | COMBO | mean | 3 options: mean, sum, none |
| weight_enabled | COMBO | False | 2 options: True, False |
| class_weights | STRING | [] | — |
| margin | FLOAT | 0.0-1–1 | — |
| use_lr_scheduler | COMBO | True | 2 options: True, False |
| scheduler_type | COMBO | ReduceLROnPlateau | 3 options: StepLR, ReduceLROnPlateau, CosineAnnealingLR |
| scheduler_step_size | INT | 31–100 | — |
| scheduler_gamma | FLOAT | 0.100.01–1 | — |
| min_lr | FLOAT | 0.00001e-7–0.1 | — |
| use_early_stopping | COMBO | True | 2 options: True, False |
| patience | INT | 51–50 | — |
| min_delta | FLOAT | 0.00100.0001–0.1 | — |
| reshape_input | COMBO | False | 2 options: True, False |
| input_reshape_dim | STRING | [-1] | — |
| flatten_input | COMBO | False | 2 options: True, False |
| hyperparametersopt | DICT | — |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| model | MODEL | — |
| training_log | STRING | — |
| metrics | DICT | — |