In the COCO dataset, the number of validation images is 1000, but I found many papers

The number of COCO validation images about vsepp HOT 1 CLOSED

fartashf commented on June 14, 2024

The number of COCO validation images

from vsepp.

Comments (1)

fartashf commented on June 14, 2024

There are 5000 test images as well as more than 5000 validation images in the original split of MSCOCO. Please see Section 3.1 of the paper for the details and references (https://arxiv.org/pdf/1707.05612.pdf) and the corresponding lines in the code:

vsepp/data.py

Lines 51 to 59 in 7f8ffd5

 ids['train'] = np.load(os.path.join(capdir, 'coco_train_ids.npy')) 

 ids['val'] = np.load(os.path.join(capdir, 'coco_dev_ids.npy'))[:5000] 

 ids['test'] = np.load(os.path.join(capdir, 'coco_test_ids.npy')) 

 ids['trainrestval'] = ( 

 ids['train'], 

 np.load(os.path.join(capdir, 'coco_restval_ids.npy'))) 

 if use_restval: 

 roots['train'] = roots['trainrestval'] 

 ids['train'] = ids['trainrestval']

from vsepp.

Related Issues (20)

How to caculate the scores on MSCOCO 1k test images? HOT 3
Metrics for 1k test images on MS COCO HOT 1
Loss stuck, not decreasing HOT 2
The question about loss function HOT 1
How to build vocab? HOT 2
Can't reproduce the result using pytorch 0.4.1 branch HOT 3
questions on dataset construction HOT 3
encoding data
about use dataset HOT 1
Runs file too large HOT 1
FileNotFoundError when try to reproduce results of pretrained model HOT 1
train on synthetic dataset HOT 2
Reproducing results HOT 5
Same meanr being logged by tb_logger during validation HOT 1
loss gap between train and test HOT 1
Question about your model ? HOT 1
Where are your model weights stored? HOT 1
Doubt
RuntimeError: mat1 and mat2 shapes cannot be multiplied (4608x2048 and 4096x1024) HOT 1

The number of COCO validation images about vsepp HOT 1 CLOSED

Comments (1)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	ids['train'] = np.load(os.path.join(capdir, 'coco_train_ids.npy'))
	ids['val'] = np.load(os.path.join(capdir, 'coco_dev_ids.npy'))[:5000]
	ids['test'] = np.load(os.path.join(capdir, 'coco_test_ids.npy'))
	ids['trainrestval'] = (
	ids['train'],
	np.load(os.path.join(capdir, 'coco_restval_ids.npy')))
	if use_restval:
	roots['train'] = roots['trainrestval']
	ids['train'] = ids['trainrestval']