Can you please post your trained weights?

[jordanw421]

Thank you very much!

[sbillburg]

I’d love to help you, but the weight files are in my lab computer, and I am taking an internship far away from my college during the entire Aug.

If you need the weights these days, the quickest way is to train the model yourself.

My code is easy and fast to train, you just need to download the dataset that I gave in Readme, modify the paths in batch_generator.py, and if you have a Nvidia GPU like GTX 1080, just run the CRNN_with_STN.py. You will get 90% accuracy in 24 hours.

If you get any problem, please contact me.

BTW. If you like my project, pls give a star (＾▽＾)

[sniper0110]

Hello,

Thanks for the code, can you please post the weights if you're back? I don't have a good enough computer to train from scratch, I am thinking of doing some transfer learning so the weights will be very helpful.

Thanks in advance!

[sbillburg]

Hello,

Thanks for the code, can you please post the weights if you're back? I don't have a good enough computer to train from scratch, I am thinking of doing some transfer learning so the weights will be very helpful.

Thanks in advance!

sure, give me your email address.

[WrathofBhuvan11]

Bro can you send me the weight files hdf5 of crnn+stn on synth90K

[sbillburg]

Bro can you send me the weight files hdf5 of crnn+stn on synth90K

sure, send me a email and i will rely you with a hdf5 file.
junbo_liu@foxmail.com

[un-lock-me]

that would be great if you could upload your learned weights here.

Thanks in advance!

[sbillburg]

that would be great if you could upload your learned weights here.

Thanks in advance!

send me your email address

[un-lock-me]

goudarzvand.s@gmail.com

Thank you!

[un-lock-me]

Hello again,

I am still waiting for your model as I am getting 85% accuracy.
Thanks for taking the time.

[sbillburg]

Hello again,

I am still waiting for your model as I am getting 85% accuracy.
Thanks for taking the time.

sry for answering you too late, I just sent the weight files to you.

[no-1ne]

If it permits, do you mind uploading it to google drive or Baidu

Thanks

[sbillburg]

If it permits, do you mind uploading it to google drive or Baidu

Thanks

baidu url：https://pan.baidu.com/s/1ENS5vwRO3CsKQiM4iWwndA
password：rsvk

[argosang]

can you upload to google drive? i can't access baidu.

[german0]

can you upload to google drive? i can't access baidu.

Wasn't able to create a baidu account as well. Search for baidu netdisk, it's a workaround

[liuzhuang1024]

I’d love to help you, but the weight files are in my lab computer, and I am taking an internship far away from my college during the entire Aug.

If you need the weights these days, the quickest way is to train the model yourself.

My code is easy and fast to train, you just need to download the dataset that I gave in Readme, modify the paths in batch_generator.py, and if you have a Nvidia GPU like GTX 1080, just run the CRNN_with_STN.py. You will get 90% accuracy in 24 hours.

If you get any problem, please contact me.

BTW. If you like my project, pls give a star (＾▽＾)

Thank you for your answer.

[sbillburg]

can you upload to google drive? i can't access baidu.

Sry for reply so late, I have added a Google drive share link in README

[sbillburg]

can you upload to google drive? i can't access baidu.

Wasn't able to create a baidu account as well. Search for baidu netdisk, it's a workaround

You can find the download link in README now