How many images are used during 1 iteration of training?

[bekoju]

Network section of the cfg file

[net]
# Testing
#batch=1
#subdivisions=1
# Training
batch=64
# subdivisions=64
width=416
height=416
channels=3
momentum=0.949
# decay=0.0005
# angle=0
# saturation = 1.5
# exposure = 1.5
# hue=.1

learning_rate=0.001
burn_in=1000
max_batches = 10000
policy=steps
steps=8000,9000
scales=.1,.1

No. of images training images

train_images = 758849
val_images    = 93834

Also, why the next mAP calculation is at 47428

 (next mAP calculation at 47428 iterations) ]2;1/10000: loss=1992.6 hours left=-1.0
 1: 1992.606201, 1992.606201 avg loss, 0.000000 rate, 6.079220 seconds, 64 images, -1.000000 hours left
 MJPEG-stream sent. 
Loaded: 0.000018 seconds