crYOLO - training: Training of crYOLO, a deep learning high accuracy particle picking procedure.

Usage in command line

sp_cryolo_train.py particle_diameter training_dir annot_dir --cryolo_train_path=CRYOLO_PATH --architecture=architecture --input_size=input_size --num_patches=num_patches --overlap_patches=overlap_patches --train_times=train_times --pretrained_weights_name=PRETRAINED_NAME --saved_weights_name=SAVE_WEIGHTS_NAME --batch_size=batch_size --learning_rate=learning_rate --np_epoch=np_epoch --object_scale=object_scale --no_object_scale=no_object_scale --coord_scale=coord_scale --valid_image_dir=valid_image_dir --valid_annot_dir=valid_annot_dir --warmup=warmup --gpu=gpu --fine_tune --gpu_fraction=GPU_FRACTION --num_cpu=NUM_CPU

To train crYOLO for a specific dataset, one have to specify the path to training data in the config file. Then the training typcial happens in two steps:

1. Warmup:

sp_cryolo_train.py particle_diameter training_dir annot_dir --architecture="YOLO" --warmup=5

2. Actual training:

sp_cryolo_train.py --conf=config_path --warmup=0 --gpu=0

--cryolo_train_path

crYOLO train executeable

Path to the crYOLO executeable (default none)

particle_diameter

Particle diameter [Pixel]

Particle diameter in pixel. This size will be used for as box size for picking. Should be as small as possible. (default required int)

training_dir

Training image directory

Folder which contain all images. (default required string)

annot_dir

Annotation directory

Box or star files used for training. The should have the same name as the images. (default required string)

--architecture

Network architecture: Type of network that is trained. (default PhosaurusNet)

--input_size

Input image dimension [Pixel]

Dimension of the image used as input to network. (default 1024)

--num_patches

Number of patches

The number of patches (e.g 2×2) the image is divided and classified separately. (default 1)

--overlap_patches: Patch overlap [Pixel]: The amount of overlap the patches will overlap (default 0)

--train_times

Repeat images

How often a images is augmented and repeadet in one epoch. (default 10)

--pretrained_weights_name: Pretrained weights name

Name of the pretrained model (default cryolo_model.h5)

--saved_weights_name: Saved weights name

Name of the model to save (default cryolo_model.h5)

--batch_size

Batch size

How many patches are processed in parallel. (default 5)

--fine_tune

Fine tune mode

Set it to true if you only want to use the fine tune mode. (default False)

--learning_rate

Learning rate

Learning rate used during training. (default 0.0001)

--np_epoch

Number of epochs

Maximum number of epochs. (default 100)

--object_scale

Object loss scale

Loss scale for object. (default 5.0)

--no_object_scale: Background loss scale: Loss scale for background. (default 1.0)

--coord_scale: Coordinates loss scale: Loss scale for coordinates. (default 1.0)

--valid_image_dir

Path to validation images

Images used (default none)

--valid_annot_dir

Path to validation annotations

Path to the validation box files (default none)

--warmup

Warm up epochs

Number of warmup epochs. (default 5)

--gpu: GPUs

List of GPUs to use. (default 0)

--gpu_fraction: GPU memory fraction

Specify the fraction of memory per GPU used by crYOLO during training. Only values between 0.0 and 1.0 are allowed. (default 1.0)

--num_cpu: Number of CPUs

Number of CPUs used during training. By default it will use half of the available CPUs. (default -1)

It will write a .h5 file (default yolo_model.h5) into your project directory.

The training is divided into two parts. 1. Warmup: It prepares the network with a few epochs of training without actually estimating the size of the particle. 2. Actual training: The training will stop when the loss on the validation data stops to improve.

See the reference below.

Training needs a GPU with ~8GB memory. Training on 20 micrographs typicall needs ~20 minutes.

• Initial creation of the document

https://doi.org/10.1101/356584

Thorsten Wagner

Category 1:: APPLICATIONS

sparx/bin/sp_cryolo_train.py

crYOLO

Stable

None right now.