filippomb / tck_ae Goto Github PK
View Code? Open in Web Editor NEWHome Page: HomePage
tck_ae's People
Contributors
Stargazers
Watchers
tck_ae's Issues
Minor change in loss function is not working as expected!
Hi there!, thank you for sharing this code. I am trying to add an entropy function as density estimation in addition to the reconstruction loss and kernel alignment loss.
$ \mathcal{L} = \frac{1}{N} \sum_{N=1}^{N} L(X, Z) + \min_{x} (H(X) + \lambda D_{KL}(X||Z)) $
I have made the following changes in the code.
prob = prior_K_norm / tf.reduce_sum(prior_K_norm)
tf_entrpy = -tf.reduce_sum(prob * tf.log(prob))
tot_loss = reconstruct_loss + args.w_reg * reg_loss + args.a_reg * (k_loss + tf_entrpy)
_, train_loss, train_kloss, train_entropy_loss = sess.run([update_step, reconstruct_loss, k_loss, tf_entrpy], fdtr)
entrpyloss_track.append(train_entropy_loss)
outvs, lossvs, klossvs, entrpy_loss_vs, vs_code_K, summary = sess.run(
[dec_out, reconstruct_loss,k_loss, tf_entrpy, code_K, merged_summary], fdvs)
train_writer.add_summary(summary, ep)
print('VS r_loss=%.3f, k_loss=%.3f ,entrpy_loss= %.3f -- TR r_loss=%.3f, k_loss=%.3f, entrpy_loss= %.3f' % (
lossvs, klossvs, entrpy_loss_vs, np.mean(loss_track[-100:]),
np.mean(kloss_track[-100:]),np.mean((entrpyloss_track[-100:]))))
But the entropy loss is stuck at a value. I am not sure if I am correctly calculating backpropagation on the entropy loss. Can you kindly have a look? Thank you in advance!
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.