Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Use CTC + tensorflow to OCR
| Date | Stars |
|---|---|
| 2026-07-24 | 351 |
| 2026-07-25 | 351 |
| 2026-07-28 | 351 |
| 2026-07-30 | 351 |
| 2026-08-06 | 351 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
- [old master](https://github.com/ilovin/lstm_ctc_ocr/tree/backup):
- harder to converge compare to the beta version
- both standard ctc and warpCTC
- read data at once
- [dev](https://github.com/ilovin/lstm_ctc_ocr/tree/dev):
- the pipline version of lstm_ctc_ocr, resize to same size
- use tf.records
- [beta](https://github.com/ilovin/lstm_ctc_ocr/tree/beta) (current):
- generate data on the fly
- deal with multi-width image, padding to same width
## How to use
1. ./train.sh
### Dependency
- python 3
- tensorflow 1.0.1
- [captcha](https://pypi.python.org/pypi/captcha)
- [warpCTC tensorflow_binding](https://github.com/baidu-research/warp-ctc/tree/master/tensorflow_binding)
### Some details
The training data:
Notice that,
parameters can be found in `./lstm.yml`(higher priority) and `lib/lstm/utils/config.y`
some parameters need to be fined tune:
- learning rate
- decay step & decay rate
- image_height
- optimizer?
in `./lib/lstm/utils/gen.py`, the height of the images are the same, and I pad the width
to the same for each batch, so
if you want to use your own data, the height of the image shall be the same.
### Result
The accurary can be more that 95%
Read [this blog](https://ilovin.github.io/2017-04-06/tensorflow-lstm-ctc-ocr/) for more details and [this blog](http://ilovin.github.io/2017-04-23/tensorflow-lstm-ctc-input-output/) for how to
use `tf.nn.ctc_loss` or `warpCTC`
Excerpt of 1,555 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a132b45d1fd94c05, name:ocr, desc:ocr, readme:ocr
matched fp:a132b45d1fd94c05, topic:tensorflow