Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Provide trained mode #12

Closed
goodrahstar opened this issue Oct 12, 2016 · 28 comments
Closed

Provide trained mode #12

goodrahstar opened this issue Oct 12, 2016 · 28 comments

Comments

@goodrahstar
Copy link

Hi,
Can someone please provide the trained model.

Thanks

@usamaoDATA
Copy link

i too would like access to the trained weights, i dont have the resources to run the training scripts

@ChadTaljaardt
Copy link

I will train it and provide it for you in a few days 👍

@ChadTaljaardt
Copy link

Im having issues installing everything required (tensor flow) but i will have it resolved soon.

@ChadTaljaardt
Copy link

Okay i have a working system out now, training the file today, will upload it in around 7 hours from now 👍

@goodrahstar
Copy link
Author

Thanks a lot @ChadTaljaardt .. w8ting for your training model.

@ISKU
Copy link

ISKU commented Oct 21, 2016

gj 😄 i want to.

@ChadTaljaardt
Copy link

ChadTaljaardt commented Oct 21, 2016

Here is the file.

There was a error when i clicked ctrl + c to end the training after around 180,000 batches,

Can someone try this and tell me if it works : http://www.mediafire.com/file/j08uh8e4f26u110/weights.npz

@ChadTaljaardt
Copy link

Got it to work, was using a bad image.

Here is a result of the model.

out

@usamaoDATA
Copy link

thank you for this

@usamaoDATA
Copy link

can you tell me how long did it take to train and what were the specs of the machine doing the training

@ChadTaljaardt
Copy link

my specs are:

AMD Phenom x6 1090T
12GB Ram
GTX 1070
Ubuntu 16.10
CUDA 8.0
CUDNN 5.1.5

And i think it took like 4-5 hours to converge at around 100,000 but i ran it around 7 hours.

@goodrahstar
Copy link
Author

Thanks a lot @ChadTaljaardt ... it was really helpful :)
Cheers (Y)

@pentateu
Copy link

@ChadTaljaardt the file is no longer available

@ChadTaljaardt
Copy link

@sxs4337
Copy link

sxs4337 commented Feb 24, 2017

I am unable to load the weights.npz file and get the below error. I have the latest version of numpy with Python 2.7. Any help is appreciated.

Traceback (most recent call last):
File "detect.py", line 184, in
f = numpy.load(sys.argv[2])
File "/usr/local/lib/python2.7/dist-packages/numpy/lib/npyio.py", line 429, in load
"Failed to interpret file %s as a pickle" % repr(file))
IOError: Failed to interpret file 'weights.npz' as a pickle

@chinmayips
Copy link

HI , @ChadTaljaardt , I am trying to run train , but not on a GPU , but on a normal CPU .
After about 3 hours , It did not generate any weights.npz. I agree it will not be well trained but it should have at least generated a weights.npz??? Does it generate weights.npz only after it is well trained or maybe there is some other glitch???

@Abhi001vj
Copy link

@Chinmayi-Prasad I think it will be generated if you do ctrl-c ,that's what i think from this code snippet from 249 in train.py.

`

        249 except KeyboardInterrupt:

        last_weights = [p.eval() for p in params]

        numpy.savez("weights.npz", *last_weights)

        return last_weights`

@JuanjoLopez
Copy link

What database did you use?

@chinmayips
Copy link

@JuanjoLopez , The one generated using the generate.py code .

@prabhjot-singh-gogana
Copy link

Hi @ChadTaljaardt , I really appreciate your affords to create an awesome model. Just want to know can we convert it into .caffemodel or .mlmodel. Thanks in advance

@QueenJuliaZxx
Copy link

hi,ChadTaljaardt,what is the result come from,is it use the weight file you commented on Oct 22, 2016 ,or another weight file you trained!because I use the weight file you "ChadTaljaardt commented on Oct 22, 2016 ",it work still so bad!

@QueenJuliaZxx
Copy link

hi,ChadTaljaardt,what is the result come from,is it use the weight file you commented on Oct 22, 2016 ,or another weight file you trained!because I use the weight file you "ChadTaljaardt commented on Oct 22, 2016 ",it work still so @ChadTaljaardt

@preetj
Copy link

preetj commented Apr 6, 2019

I'm unable to load 'weights.npz' and having below error. Python version= 3.6.7 and numpy version= 1.16.2. Please, somebody, help me solve this error. Thankyou

Traceback (most recent call last):
File "detect.py", line 183, in
f = numpy.load("/weights.npz", allow_pickle=True, fix_imports=True)
File "/./././././lib/python3.6/site-packages/numpy/lib/npyio.py", line 415, in load
fid = open(os_fspath(file), "rb")
FileNotFoundError: [Errno 2] No such file or directory: '/weights.npz'

@preetj
Copy link

preetj commented Apr 6, 2019

I'm unable to load 'weights.npz' and having below error. Python version= 3.6.7 and numpy version= 1.16.2. Please, somebody, help me solve this error. Thankyou

Traceback (most recent call last):
File "detect.py", line 183, in
f = numpy.load("/weights.npz", allow_pickle=True, fix_imports=True)
File "/./././././lib/python3.6/site-packages/numpy/lib/npyio.py", line 415, in load
fid = open(os_fspath(file), "rb")
FileNotFoundError: [Errno 2] No such file or directory: '/weights.npz'

@ChadTaljaardt @goodrahstar @usamaoDATA @ISKU @pentateu

@ChadTaljaardt
Copy link

Hello everyone.

Sorry that i have not been active in this thread, i haven't used this project in a long time and don't even have the computer i used anymore to be able to train another model.

Please remember that the project was "experimental" and wasn't designed to match license plates that didn't match a specific format.

@ChadTaljaardt
Copy link

@preetj Can you show us your directory structure?

@preetj
Copy link

preetj commented Apr 8, 2019

@preetj Can you show us your directory structure?

ThankYou so much @ChadTaljaardt for replying.
I found the error just now it was because I extracted the weights.npz folder instead of using the zip file.

@haha363
Copy link

haha363 commented Apr 18, 2020

@preetj Can you show us your directory structure?

ThankYou so much @ChadTaljaardt for replying.
I found the error just now it was because I extracted the weights.npz folder instead of using the zip file.

Hi @preetj
I am encountering the same error as you, can you show your code on how you solve this problem? Thanks a lot

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests