- Is there a way in Keras to apply different weights to a cost function? · Issue #2115 · keras-team/keras is a really nice discussion with options, especially the latter part of it.
- ValueError: No gradients provided for any variable in Tensorflow - Stack Overflow
When you do
annotation_pred = tf.to_float(tf.argmax(out, dimension=4, name='prediction')), you get an index of the max value in your tensor. This index can’t be derivated, thus the gradient can’t flow throught this operation.
So as your loss is only defined by this value, and the gradient can’t flow throught it, no gradient can be calculated for your network.
- GitHub - fchollet/keras-resources: Directory of tutorials and open-source code repositories for working with Keras, the Python deep learning library
- Query = [kwee ree]
- Paradigm = [pa ruh daim] or American [peh ruh daim]
CRF and probability tutorial / explanation / presentation
I really should resurrect my link DB.Read more...
Tensorflow custom cost functions with different weights
Taskwarrior modifier .isnt for “not equal”
Edited my “someday” report:Read more...
Moving to a new apt
Ausstattung für die erste eigene Wohnung - Checkliste is a nice checklist :)
Remark.js presentation change font size inside slideRead more...
Tensorflow gradient propagation
Intellij ideavim toggle case
Added the following to
map <leader>c :action EditorToggleCase<CR>
Tensorflow metrics ignored if loss doesn’t get defined
So I don’t forget, Metrics ignored when using
model.add_loss() (like in VAE example) · Issue #9459 · keras-team/keras · GitHub currently happens.
gt command to search tabs
:buffer works as a best-match tab filtering thingy, this in nice with 30+ tabs when tab number doesn’t cut it anymore
ML/Tensorflow logits meaning
Logits are the inputs to the softmax function:
the vector of raw (non-normalized) predictions that a classification model generates, which is ordinarily then passed to a normalization function. If the model is solving a multi-class classification problem, logits typically become an input to the softmax function. The softmax function then generates a vector of (normalized) probabilities with one value for each possible class.
apt-get purge and zsh
zsh does its own wildcard stuff, and
apt-get purge nvidia* doesn’t work because of this.
apt-get purge nvidia\* does (or with ‘’s). Same story as with scp, I’m surprised I keep having issues with this.
Google has nice animations for this!
Most of this while I’m reading the “Attention is all you need” paper. The most important resources will be The Illustrated Transformer – Jay Alammar – Visualizing machine learning one concept at a time and 9.3. Transformer — Dive into Deep Learning 0.7 documentation.Read more...
Pyplot / matplotlib creating multiple subplotsRead more...
Python redirect stdout/output to file
This is really really neat!Read more...
Unpack Java jars
Apparently java files are archives! Laut Extracting the Contents of a JAR File (The Java™ Tutorials > Deployment > Packaging Programs in JAR Files) it’s
jar xf jar-file [archived-file(s) to extract].
Jekyll hide page from navigation
Out of all the ways I’ve found and described here, this is so far the cleanest one.Read more...