I want to infer with my dataset and do not want to download the training sets, is ther

<a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/us

Inference only about clip4str HOT 1 OPEN

Ly-Lynn commented on August 20, 2024

Inference only

from clip4str.

Comments (1)

mzhaoshuai commented on August 20, 2024

@Ly-Lynn

Please check

bash scripts/read.sh 0 clip4str_base16x16_d70bde1f2d.ckpt misc/test_images

Output:
image_1576.jpeg: Chicken

It should be helpful for testing images.

If you want to test on lmdb format, please check

CLIP4STR/strhub/data/module.py

Lines 28 to 33 in ef592fe

 TEST_BENCHMARK_SUB = ('IIIT5k', 'SVT', 'IC13_1015', 'IC15_1811', 'IC15_2077', 'SVTP', 'CUTE80', 'HOST', 'WOST') 

 TEST_BENCHMARK = ('IIIT5k', 'SVT', 'IC13_1015', 'IC15_1811', 'IC15_2077', 'SVTP', 'CUTE80', 'HOST', 'WOST') 

 # TEST_BENCHMARK_SUB = ('HOST',) 

 # TEST_BENCHMARK = ('HOST',) 

 TEST_NEW = ('ArT', 'COCOv1.4', 'Uber') 

 TEST_ALL = tuple(set(TEST_BENCHMARK_SUB + TEST_BENCHMARK + TEST_NEW))

You should change them accordingly.

from clip4str.

Related Issues (20)

Bug in run bash script HOT 12
Error locating target for VL4STR HOT 1
The provided lr scheduler `OneCycleLR` doesn't follow PyTorch's LRScheduler API HOT 2
请问训练时图片的尺寸大小是会resize成224*224还是128*32呢？ HOT 9
clip4str attention graph HOT 2
Inference time on CPU/GPU HOT 2
Is there a way to detect spaces? HOT 1
Convert to ONNX HOT 2
Issue with inference HOT 13
bad filename format for base checkpoint pre-trained on DC-1B HOT 9
Using the LaTeX dataset to train CLIP4STR HOT 1
Issue with inference ! HOT 9
Could training Parseq directly on image data containing text content sampled from the 400 million images used to train CLIP potentially yield better results?
Recognize French accents HOT 2
Error While Inferencing HOT 2
inference error HOT 3
how to Finetuning in korean (or other language) HOT 2
Training other language HOT 3
Error Running Code: Missing clip4str_base16x16_d70bde1f2d.ckpt Checkpoint File HOT 1

Inference only about clip4str HOT 1 OPEN

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

	TEST_BENCHMARK_SUB = ('IIIT5k', 'SVT', 'IC13_1015', 'IC15_1811', 'IC15_2077', 'SVTP', 'CUTE80', 'HOST', 'WOST')
	TEST_BENCHMARK = ('IIIT5k', 'SVT', 'IC13_1015', 'IC15_1811', 'IC15_2077', 'SVTP', 'CUTE80', 'HOST', 'WOST')
	# TEST_BENCHMARK_SUB = ('HOST',)
	# TEST_BENCHMARK = ('HOST',)
	TEST_NEW = ('ArT', 'COCOv1.4', 'Uber')
	TEST_ALL = tuple(set(TEST_BENCHMARK_SUB + TEST_BENCHMARK + TEST_NEW))