This section describes the built-in algorithms supported by ModelArts and the running parameters supported by each algorithm. You can set running parameters for a training job as required.

yolo_v3

**Table 1** Algorithm description
Parameter	Description
Name	yolo_v3
Usage	Object detection and locating
Engine Type	MXNet, MXNet-1.2.1-python2.7
Precision	81.7%(mAP) mAP is an indicator that measures the effect of an object detection algorithm. For object detection tasks, the precision rate (Precision) and recall rate (Recall) can be calculated for each class of object. The rates can be calculated and tested multiple times for each class of object based on different thresholds, and a P-R curve is obtained accordingly. The area under the curve is the average value.
Training Dataset	Camvid
Data Format	shape: [H>=224, W>=224, C>=1]; type: int8
Running Parameter	lr=0.0001 ; mom=0.9 ; wd=0.0005 For more available running parameters, see Table 2.

**Table 2** Running parameters
Optional Parameter	Parameter Description	Default Value
lr	Learning rate	0.0001
mom	Momentum of the training network	0.9
wd	Parameter weight decay coefficient, L2	0.0005
num_classes	Total number of image classes used in training. If you add images of other classes, this parameter will be automatically identified and matched. You do not need to manually change the parameter value.	None
split_spec	Split ratio of the training set and validation set	0.8
batch_size	Total number of training images updated each time	16
eval_frequence	Frequency for validating the model. By default, validation is performed every epoch.	1
num_epoch	Number of training epochs	10

**Table 3** Running parameters that can be added
Optional Parameter	Parameter Description	Recommended Value
num_examples	Total number of images used for training. For example, if the total number of images is 1,000, the images used for training is 800 based on the split ratio.	16551
disp_batches	The loss and training speed of the model is displayed every N batches.	20
warm_up_epochs	Number of epochs when the target learning rate of the warm-up strategy is reached	0
lr_steps	Number of epochs when the learning rate attenuates in the multi-factor strategy. By default, the learning rate attenuates to 0.1 times of the original value at the 10th and 15th epochs.	10,15

retinanet_resnet_v1_50

**Table 4** Algorithm description
Parameter	Description
Name	retinanet_resnet_v1_50
Usage	Object detection and locating
Engine Type	TensorFlow, TF-1.8.0-python2.7
Precision	83.15%(mAP) mAP is an indicator that measures the effect of an object detection algorithm. For object detection tasks, the precision rate (Precision) and recall rate (Recall) can be calculated for each class of object. The rates can be calculated and tested multiple times for each class of object based on different thresholds, and a P-R curve is obtained accordingly. The area under the curve is the average value.
Training Dataset	Pascal VOC2007, detection of 20 classes of objects
Data Format	shape: [H, W, C>=1]; type: int8
Running Parameter	For more available running parameters, see Table 5.

**Table 5** Running parameters
Optional Parameter	Parameter Description	Default Value
split_spec	Split ratio of the training set and validation set	train:0.8,eval:0.2
num_gpus	Number of used GPUs	1
batch_size	Number of images for each iteration (standalone). To ensure the algorithm precision, you are advised to use the default value.	32
eval_batch_size	Number of images read each step during validation (standalone)	32
learning_rate_strategy	Learning rate strategy. The value ranges from 0 to 1. For example, the value can be set to 0.001.	0.002
evaluate_every_n_epochs	A validation is performed after N epochs are trained.	1
save_interval_secs	Interval for saving the model. The unit is second. If model running time is greater than 2,000,000s, the model is saved once every 2,000,000s by default. If model running time is less than 2,000,000s, the model is saved when the running is complete.	2000000
max_epoches	Maximum number of training epochs	100
log_every_n_steps	Logs are printed every N steps. By default, logs are printed every 10 steps.	10
save_summaries_steps	Summary information is saved every five steps, including the model gradient update value and training parameters.	5

inception_v3

**Table 6** Algorithm description
Parameter	Description
Name	inception_v3
Usage	Image Classification
Engine Type	TensorFlow, TF-1.8.0-python2.7
Precision	78.00%(top1), 93.90%(top5) top1 indicates that the classification is considered correct only when the image with the maximum probability is the correct image. top5 indicates that the classification is considered correct only when the correct image is within the top 5 images.
Training Dataset	ImageNet, classification of 1,000 image classes
Data Format	shape: [H, W, C>=1]; type: int8
Running Parameter	batch_size=32 ; split_spec=train:0.8,eval:0.2 ; For more available running parameters, see Table 7.

**Table 7** Running parameters
Optional Parameter	Parameter Description	Default Value
split_spec	Split ratio of the training set and validation set	train:0.8,eval:0.2
num_gpus	Number of used GPUs	1
batch_size	Number of images for each iteration (standalone). To ensure the algorithm precision, you are advised to use the default value.	32
eval_batch_size	Number of images read each step during validation (standalone)	32
learning_rate_strategy	Learning rate strategy. For example, 10:0.001,20:0.0001 indicates that the learning rate for 0 to 10 epochs is 0.001, and that for 10 to 20 epochs is 0.0001.	0.002
evaluate_every_n_epochs	A validation is performed after N epochs are trained.	1
save_interval_secs	Interval for saving the model. The unit is second. If model running time is greater than 2,000,000s, the model is saved once every 2,000,000s by default. If model running time is less than 2,000,000s, the model is saved when the running is complete.	2000000
max_epoches	Maximum number of training epochs	100
log_every_n_steps	Logs are printed every N steps. By default, logs are printed every 10 steps.	10
save_summaries_steps	Summary information is saved every five steps, including the model gradient update value and training parameters.	5

**Table 8** Running parameters that can be added
Optional Parameter	Parameter Description	Recommended Value
weight_decay	L2 regularization weight decay	0.00004
optimizer	Optimizer. The options are as follows: dymomentumw sgd adam momentum	momentum
momentum	Optimizer parameter momentum	0.9
patience	After training of N epochs, if the precision (mAP for object detection and accuracy for image classification) does not increase compared with the previous maximum value, that is, the difference between the precision and the maximum precision is less than the value of decay_min_delta, the learning rate attenuates to one tenth of the original value. The default value of N is 8.	8
decay_patience	After training of extra M epochs on the basis of the preceding patience, if the precision (mAP for object detection and accuracy for image classification) does not increase, that is, the difference between the precision and the maximum precision is less than the value of decay_min_delta, training will be terminated early. The default value of M is 1.	1
decay_min_delta	Minimum difference between the precision (mAP for object detection and accuracy for image classification) corresponding to different learning rates. If the parameter value is greater than 0.001, the precision is increased. Otherwise, the precision is not increased.	0.001
image_size	Size of the input image. If this parameter is set to None, the default image size prevails.	None
lr_warmup_strategy	Warm-up strategy (linear or exponential)	linear
num_readers	Number of threads for reading data	64
fp16	Whether to use FP16 for training	FALSE
max_lr	Maximum learning rate for the dymomentum and dymomentumw optimizers, or when use_lr_schedule is used	6.4
min_lr	Minimum learning rate for the dymomentum and dymomentumw optimizers, or when use_lr_schedule is used	0.005
warmup	Proportion of warm-up in total training steps. This parameter is valid when use_lr_schedule is lcd or poly.	0.1
cooldown	Minimum learning rate in the warm-up	0.05
max_mom	Maximum momentum. This parameter is valid for dynamic momentum.	0.98
min_mom	Minimum momentum. This parameter is valid for dynamic momentum.	0.85
use_lars	Whether to use LARS	FALSE
use_nesterov	Whether to use Nesterov Momentum	TRUE
preprocess_threads	Number of threads for image preprocessing	12
use_lr_schedule	Learning rate adjustment policy ('lcd':linear_cosine_decay, 'poly':polynomial_decay)	None

darknet_53

**Table 9** Algorithm description
Parameter	Description
Name	darknet_53
Usage	Image Classification
Engine Type	MXNet, MXNet-1.2.1-python2.7
Precision	78.56%(top1), 94.43%(top5) top1 indicates that the classification is considered correct only when the image with the maximum probability is the correct image. top5 indicates that the classification is considered correct only when the correct image is within the top 5 images.
Training Dataset	ImageNet, classification of 1,000 image classes
Data Format	shape: [H>=224, W>=224, C>=1]; type: int8
Running Parameter	split_spec=0.8 ; batch_size=4 ; For more available running parameters, see Table 10.

**Table 10** Running parameters
Optional Parameter	Parameter Description	Default Value
split_spec	Split ratio of the training set and validation set	0.8
batch_size	Total amount of input data each time the parameters are updated	4
lr	Learning rate of the updated parameters	0.0001
save_frequency	Interval for saving the model, indicating that the model is saved every N epochs	1
num_classes	Total number of image classes in training	None
num_epoch	Number of training epochs	10

SegNet_VGG_BN_16

**Table 11** Algorithm description
Parameter	Description
Name	SegNet_VGG_BN_16
Usage	Image semantic segmentation
Engine Type	MXNet, MXNet-1.2.1-python2.7
Precision	89%(pixel acc) pixel acc indicates the ratio of correct pixels to total pixels.
Training Dataset	Camvid
Data Format	shape: [H=360, W=480, C==3]; type: int8
Running Parameter	deploy_on_terminal=False; For more available running parameters, see Table 12.

**Table 12** Running parameters
Optional Parameter	Parameter Description	Default Value
lr	Learning rate of the updated parameters	0.0001
mom	Momentum of the training network	0.9
wd	Attenuation coefficient	0.0005
num_classes	Total number of image classes in training. You do not need to plus 1 here.	11
batch_size	Total number of training images updated each time	8
num_epoch	Number of training epochs	15
save_frequency	Interval for saving the model, indicating that the model is saved every N epochs	1
num_examples	Total number of images used for training, which indicates the number of files in train.txt	2953

ResNet_v2_50

**Table 13** Algorithm description
Parameter	Description
Name	ResNet_v2_50
Usage	Image Classification
Engine Type	MXNet, MXNet-1.2.1-python2.7
Precision	75.55%(top1), 92.6%(top5) top1 indicates that the classification is considered correct only when the image with the maximum probability is the correct image. top5 indicates that the classification is considered correct only when the correct image is within the top 5 images.
Training Dataset	ImageNet, classification of 1,000 image classes
Data Format	shape: [H>=32, W>=32, C>=1]; type: int8
Running Parameter	split_spec=0.8 ; batch_size=4 ; The available running parameters are the same as those for the darknet_53 algorithm. For details, see Table 10.

ResNet_v1_50

**Table 14** Algorithm description
Parameter	Description
Name	ResNet_v1_50
Usage	Image Classification
Engine Type	TensorFlow, TF-1.8.0-python2.7
Precision	74.2%(top1), 91.7%(top5) top1 indicates that the classification is considered correct only when the image with the maximum probability is the correct image. top5 indicates that the classification is considered correct only when the correct image is within the top 5 images.
Training Dataset	ImageNet, classification of 1,000 image classes
Data Format	shape: [H>=600,W<=1024,C>=1];type:int8
Running Parameter	batch_size=32 ; split_spec=train:0.8,eval:0.2 ; The available running parameters are the same as those for the inception_v3 algorithm. For details, see Table 7.

Faster_RCNN_ResNet_v2_101

To achieve a satisfactory training effect, fine-tuning of parameters is required when this algorithm is used. Otherwise, the result may not meet the expectation. Alternatively, you can use other algorithms.

**Table 15** Algorithm description
Parameter	Description
Name	Faster_RCNN_ResNet_v2_101
Usage	Object detection and locating
Engine Type	MXNet, MXNet-1.2.1-python2.7
Precision	80.05%(mAP) mAP is an indicator that measures the effect of an object detection algorithm. For object detection tasks, the precision rate (Precision) and recall rate (Recall) can be calculated for each class of object. The rates can be calculated and tested multiple times for each class of object based on different thresholds, and a P-R curve is obtained accordingly. The area under the curve is the average value.
Training Dataset	PASCAL VOC2007, PASCAL VOC2012
Data Format	shape: [H, W, C==3]; type: int8
Running Parameter	lr=0.0001 ; eval_frequence=1 ; For more available running parameters, see Table 16.

**Table 16** Running parameters
Optional Parameter	Parameter Description	Default Value
num_classes	Total number of image classes in training. The value must plus 1 because there is a background class.	None
eval_frequence	Frequency for validating the model. By default, validation is performed every epoch.	1
lr	Learning rate	0.0001
mom	Momentum of the training network	0.9
wd	Parameter weight decay coefficient, L2	0.0005
split_spec	Split ratio of the training set and validation set	0.8

Faster_RCNN_ResNet_v1_50

**Table 17** Algorithm description
Parameter	Description
Name	Faster_RCNN_ResNet_v1_50
Usage	Object detection and locating
Engine Type	TensorFlow, TF-1.8.0-python2.7
Precision	73.6%(mAP) mAP is an indicator that measures the effect of an object detection algorithm. For object detection tasks, the precision rate (Precision) and recall rate (Recall) can be calculated for each class of object. The rates can be calculated and tested multiple times for each class of object based on different thresholds, and a P-R curve is obtained accordingly. The area under the curve is the average value.
Training Dataset	Pascal VOC2007, detection of 20 classes of objects
Data Format	shape: [H>=600,W<=1024,C>=1];type:int8
Running Parameter	For details about the parameters and default values, see Table 5.

Algorithms and Their Running Parameters