qurator-spk
/
sbb_pixelwise_segmentation
mirror of https://github.com/qurator-spk/sbb_pixelwise_segmentation.git
1
0
Fork 0
Code Issues Releases Wiki Activity
You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
unifying-training-models
master
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '7661080899'
${ noResults }
sbb_pixelwise_segmentation/train_no_patches_448x448.json

53 lines
1.6 KiB
JSON
Raw Normal View History Unescape Escape

LR Warmup and Optimization Implementation # Learning Rate Warmup and Optimization Implementation ## Overview Added learning rate warmup functionality to improve training stability, especially when using pretrained weights. The implementation uses TensorFlow's native learning rate scheduling for better performance. ## Changes Made ### 1. Configuration Updates (`runs/train_no_patches_448x448.json`) Added new configuration parameters for warmup: ```json { "warmup_enabled": true, "warmup_epochs": 5, "warmup_start_lr": 1e-6 } ``` ### 2. Training Script Updates (`train.py`) #### A. Optimizer and Learning Rate Schedule - Replaced fixed learning rate with dynamic scheduling - Implemented warmup using `tf.keras.optimizers.schedules.PolynomialDecay` - Maintained compatibility with existing ReduceLROnPlateau and EarlyStopping ```python if warmup_enabled: lr_schedule = tf.keras.optimizers.schedules.PolynomialDecay( initial_learning_rate=warmup_start_lr, decay_steps=warmup_epochs * steps_per_epoch, end_learning_rate=learning_rate, power=1.0 # Linear decay ) optimizer = Adam(learning_rate=lr_schedule) else: optimizer = Adam(learning_rate=learning_rate) ``` #### B. Learning Rate Behavior - Initial learning rate: 1e-6 (configurable via `warmup_start_lr`) - Target learning rate: 5e-5 (configurable via `learning_rate`) - Linear increase over 5 epochs (configurable via `warmup_epochs`) - After warmup, learning rate remains at target value until ReduceLROnPlateau triggers ## Benefits 1. Improved training stability during initial epochs 2. Better handling of pretrained weights 3. Efficient implementation using TensorFlow's native scheduling 4. Configurable through JSON configuration file 5. Maintains compatibility with existing callbacks (ReduceLROnPlateau, EarlyStopping) ## Usage To enable warmup: 1. Set `warmup_enabled: true` in the configuration file 2. Adjust `warmup_epochs` and `warmup_start_lr` as needed 3. The warmup will automatically integrate with existing learning rate reduction and early stopping To disable warmup: - Set `warmup_enabled: false` or remove the warmup parameters from the configuration file
1 week ago
{
"backbone_type" : "nontransformer",
"task": "segmentation",
"n_classes" : 3,
"n_epochs" : 50,
"input_height" : 448,
"input_width" : 448,
"weight_decay" : 1e-4,
"n_batch" : 4,
"learning_rate": 5e-5,
"patches" : false,
"pretraining" : true,
"augmentation" : true,
"flip_aug" : false,
"blur_aug" : false,
"scaling" : true,
"degrading": true,
"brightening": false,
"binarization" : false,
"scaling_bluring" : false,
"scaling_binarization" : false,
"scaling_flip" : false,
"rotation": false,
"rotation_not_90": false,
"blur_k" : ["blur","guass","median"],
"scales" : [0.6, 0.7, 0.8, 0.9],
"brightness" : [1.3, 1.5, 1.7, 2],
"degrade_scales" : [0.2, 0.4],
"flip_index" : [0, 1, -1],
"thetha" : [10, -10],
"continue_training": false,
"index_start" : 0,
"dir_of_start_model" : " ",
"weighted_loss": false,
"is_loss_soft_dice": true,
"data_is_provided": true,
"dir_train": "/home/incognito/sbb_pixelwise_segmentation/dataset/sam_41_mss/dir_train/train",
"dir_eval": "/home/incognito/sbb_pixelwise_segmentation/dataset/sam_41_mss/dir_train/eval",
"dir_output": "runs/sam_41_mss_npt_448x448",
"reduce_lr_enabled": true,
"reduce_lr_monitor": "val_loss",
"reduce_lr_factor": 0.5,
"reduce_lr_patience": 4,
"reduce_lr_min_lr": 1e-6,
"early_stopping_enabled": true,
"early_stopping_monitor": "val_loss",
"early_stopping_patience": 6,
"early_stopping_restore_best_weights": true,
"warmup_enabled": true,
"warmup_epochs": 5,
"warmup_start_lr": 1e-6
}