-
Notifications
You must be signed in to change notification settings - Fork 370
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
Failure to read model in Tensorflow r1.0 #6
Comments
Here's the updated code by the way
|
I guess you have to retrain model with tensorflow 1.0? |
Yeah it looks like the variable names changed with Tensorflow 1.0 (this repo was created under v0.11). I haven't looked into this in depth yet but I expect the right dictionary used in connection with a saver will make it possible to fix this without retraining the model from scratch. |
I got the same problem.I retrained the model with tensorflow1.0.It worked.That`s all. |
Did you modify your code to make it work on TensorFlow 1.0?
On Thu, Mar 2, 2017, 3:00 AM aguang1201 ***@***.***> wrote:
I got the same problem.I retrained the model with tensorflow1.0.It
worked.That`s all.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#6 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAPfBKGXbZPvNGzbDdYmYUvQexgDB1idks5rhncHgaJpZM4MMpA2>
.
--
…Sent from my phone
|
Anyone who success to convert pre-trained model working on TF >1.0 without retraining? |
I am getting the same error with tesorflow 1.3 . Can anyone help on how to resolve this issue |
I tried this and it works. Thank you. But can you please confirm if what i have done is correct? I used the below list
Then in sample_main(args) function I assigned this list to variable tensors to form a dict to be passed to tf.train.Saver.
Placed the whole code here https://github.com/NarahariKaushik/toy_chatbot . Not sure how to contribute to this repo(still new to git hub) |
@aguang1201 It would be of great help to me if you can upload your trained model here. |
@NarahariKaushik This solved the problem for me. Thank you. |
Nice job @NarahariKaushik ! I hoped that would work. I've now updated the whole repo to TF 1.4 and Python3, and I retrained a new (and much larger) model from scratch for several months. I also simplified the vocabulary so that every net trained with this version of the code will use the same ASCII output to make it possible to initially train on one data set and then fine-tune the same model on another one. All that is to say that the old model is probably no longer usable. But, a new one is available! :) |
Hey I'm getting this error in Tensorflow 1.0 after modifying the code to get working
The text was updated successfully, but these errors were encountered: