lmjohns3/theanets

Issues

• Proposing a PR to fix a few small typos

  #136 opened 4 years ago by timgates42
  0

• xor example not working

  #135 opened 7 years ago by chriscamacho
  0

• Problem: Cannot pass theano.shared to .itertrain()

  #134 opened 7 years ago by Philipp777
  0

• Error in Quick Start: Recurrent Models

  #133 opened 8 years ago by saddy001
  2

• Preserve recurrent hidden state across training examples

  #132 opened 8 years ago by feynmanliang
  2

• Problem with reproducibility for release 0.6.2

  #102 opened 8 years ago by tlikhomanenko
  2

• add feed_forward-like method for computing loss value

  #117 opened 9 years ago by lmjohns3
  1

• create a model for jointly optimizing multiple models

  #88 opened 9 years ago by lmjohns3
  4

• Add 3D convolution layers and corresponding models

  #84 opened 10 years ago by lmjohns3
  0

• New loss function: ROC AUC score

  #129 opened 9 years ago by saddy001
  1

• Problem saving a network with compound activations

  #116 opened 9 years ago by juliettema
  3

• feature: use file-like objects for net saving and loading

  #99 opened 9 years ago by majidaldo
  5

• autoencoder stop training

  #130 opened 9 years ago by Duanexiao
  3

• use shape information in layers where possible

  #126 opened 9 years ago by lmjohns3
  0

• UnsupervisedPretrainer broken

  #128 opened 9 years ago by droid666
  1

• Is there any way to get feedback during training?

  #127 opened 9 years ago by gtzo
  1

• The current clockwork RNN implementation

  #125 opened 9 years ago by ypxie
  7

• for clockwork RNN

  #124 opened 9 years ago by ypxie
  0

• After finishing training, get the number of iterations

  #122 opened 9 years ago by SadatAnwar
  1

• Hidden_l1 problem

  #121 opened 9 years ago by jamona
  1

• add "bayesian rnn" implementation

  #120 opened 9 years ago by lmjohns3
  1

• add

  #119 opened 9 years ago by lmjohns3
  0

• memory leaks when calling predict() and load()

  #118 opened 9 years ago by lmjohns3
  0

• [DOCS] Update MNIST example to use 2D Convolutions

  #115 opened 9 years ago
  1

• Unable to load a theanets.Regressor from file using the load

  #114 opened 9 years ago by SadatAnwar
  2

• 'SGD' object has no attribute 'iterate'

  #113 opened 9 years ago by pannous
  2

• RNN not converging

  #108 opened 9 years ago by joetigger
  6

• unexpected keyword argument 'rng' in graph.py

  #94 opened 9 years ago by nkundiushuti
  7

• Error in build_graph (The updates parameter must be an OrderedDict/dict)

  #95 opened 9 years ago by qyouurcs
  3

• feature: automatically switch b/w GPU and CPU for performance

  #96 opened 9 years ago by majidaldo
  5

• theanets package on PyPI is broken or obsolete

  #97 opened 9 years ago by Franck-Dernoncourt
  1

• hidden dropout w/ 'predict' not performing as expected

  #100 opened 9 years ago by white54503
  1

• new feature request: Allow theanets.recurrent.batches(...) to accept a pandas dataframe

  #101 opened 9 years ago by kzk2000
  0

• does theanets support gradient of subtensor?

  #104 opened 9 years ago by fedorajzf
  1

• Any plans for supporting multi dimensional RNN?

  #105 opened 9 years ago by ypxie
  1

• Classification with two classes

  #106 opened 9 years ago by fordanic
  1

• Bug in MUT1

  #109 opened 9 years ago by joetigger
  1

• add model-wide rng parameter

  #110 opened 9 years ago by lmjohns3
  0

• Model-rng only for Regressor

  #112 opened 9 years ago by talpay
  1

• RNN not able to predict using existing state

  #111 opened 9 years ago by joetigger
  1

• python and ipython crashes when running examples on Win 8 x64

  #103 opened 9 years ago by fordanic
  6

• Layerwise GPU memory use

  #98 opened 9 years ago by abramhindle
  1

• CPU/GPU saving/conversion -- what actually needs to be serialized?

  #93 opened 9 years ago by abramhindle
  3

• Perilous pretraining, poorly named outputs causing key errors? 'pretrain' issue.

  #92 opened 9 years ago by abramhindle
  4

• Can't deserialize Regressor networks: AttributeError: 'Regressor' object has no attribute '_functions' as of 33775e7c96

  #91 opened 9 years ago by abramhindle
  2

• Using theanets with ipython notebook

  #90 opened 9 years ago by spudzee
  7

• Looks like save_progress and save_every are not working in v0.6.2

  #87 opened 9 years ago by ibeltagy
  2

• Problems continuing to train Experiment loaded from file.

  #85 opened 9 years ago by atnelson
  3

• specifying layers -- output layer

  #82 opened 10 years ago by XuesongYang
  1

• Why do I get this error message? IndexError: index 4 is out of bounds for axis 1 with size 4

  #83 opened 10 years ago by gribeiro2004
  3