6f74b33696
CNTK_201B_CIFAR-10_ImageHandsOn.ipynb: fix PIL.Image import
2017-02-10 15:57:51 +01:00
a435f5a159
Bump versions
2017-02-10 15:57:35 +01:00
487ba16ba7
fixed a typo in a Jupyter notebook
2017-02-09 23:32:38 -08:00
659e4ae4ee
addressed CR feedback
2017-02-09 19:46:20 -08:00
752c59e02f
Updated docs, test and nb spell mistakes
2017-02-09 19:35:30 -08:00
cee8cd1601
Updated the branch to current master and update CNTK 106B
2017-02-09 18:08:57 -08:00
3b02db423d
renamed Function.save_model() to save(), and restore_model() likewise;
...
made load_model a static method Function.load()
2017-02-09 17:24:39 -08:00
adf0a9238a
changed signature of splice() from list to *args, to make it consistent with all other ops;
...
renamed cntk.ops.reduce_log_sum() to reduce_log_add_exp(), to make it consistent with log_add_exp();
associative binary ops like plus() now accept >2 arguments;
new ops element_max(), element_min();
softmax() now takes axis parameter;
constant() now takes dtype parameter argument
2017-02-09 17:06:09 -08:00
9b87729469
changed LayerStack() to For()
2017-02-09 16:06:36 -08:00
19d2e5f7ea
changed all imports of blocks and models to layers
2017-02-09 15:41:12 -08:00
99c25c0327
changed Trainer() signature from Trainer(model, loss, metric, ...) to Trainer(model, (loss, metric), ...) to prep for Layers lib where (loss, metric) is the criterion function;
...
fixed an existing bug in a doc example
2017-02-09 15:22:45 -08:00
5f938516d0
Integrate thilow/SimpleExampleSSM into master
2017-02-09 09:29:50 -08:00
54c589dd53
addressing cr comments
2017-02-09 16:54:06 +01:00
f2f13194fb
Tutorials/CNTK_104_Finance_Timeseries_Basic_with_Pandas_Numpy.ipynb: fix two links to Wikipedia
...
Closes : #1416
2017-02-09 12:06:51 +01:00
98c2fbdcd1
Merge remote-tracking branch 'origin/master' into thilow/SimpleExampleSSM
2017-02-09 12:01:11 +01:00
50e8e10406
Merge remote-tracking branch 'origin/master' into thilow/SimpleExampleSSM
2017-02-08 16:15:44 +01:00
f0636c902f
Changed methods in Param to be global functions to avoid the differewnt handling of unbound functions in Python 2 vs 3
2017-02-08 14:58:22 +01:00
b1eb518b36
Fixed the pickle file version
2017-02-07 22:11:28 -08:00
937dab5675
Incorporated CR feedback with retry logic
2017-02-07 16:39:17 -08:00
3f68277bdd
Updated caching of test data and retry logic
2017-02-07 16:34:06 -08:00
43ca617425
Fixed a data download condition
2017-02-07 16:26:20 -08:00
d27353e247
Incorporated CR feedback with retry logic
2017-02-07 15:02:59 -08:00
777c9cca07
Adding small data caching for test automation
2017-02-06 16:21:14 -08:00
11e4e4a6ee
Revert "Adding small data caching for test automation"
...
This reverts commit f8383b2511
2017-02-06 16:17:19 -08:00
f8383b2511
Adding small data caching for test automation
2017-02-06 16:14:33 -08:00
ae372d7b09
Update notebooks
2017-02-02 18:23:41 -08:00
6a603b0c79
Update images
2017-02-02 17:40:20 -08:00
9ddf5d9440
Add images to notebooks
2017-02-02 14:14:02 -08:00
ceefaa5458
Fix notebook
2017-02-02 13:44:09 -08:00
473962fe49
Update images
2017-02-02 10:22:53 -08:00
38c1e38596
Test image
2017-02-02 10:06:49 -08:00
7d472b4bff
Test image
2017-02-02 10:01:27 -08:00
d4b49b8208
Bump versions
2017-02-01 18:13:00 +01:00
b121722e8c
CNTK_207: fixing grammar error
2017-02-01 15:45:17 +01:00
7ea2ff157d
CNTK_207_Training_with_Sampled_Softmax: adding test. Adding device-awareness in notebook.
2017-02-01 15:39:07 +01:00
72f1b6ff0a
Changed parameters of speed measurement for CNTK_207
2017-02-01 12:10:10 +01:00
59b41e4ead
Tutorial CNTK_207: adding to samples.json. Adding 'from __future__'
2017-02-01 10:20:36 +01:00
9efb369344
changed settings of per measurement plot
2017-01-31 22:07:22 +01:00
5a1720d9e1
Extended text
2017-01-31 18:15:12 +01:00
51b3c03a6a
Adding speed measurement to CNTK_207
2017-01-31 18:04:14 +01:00
842570d4ef
Changed some parameter attributes to lambdas
2017-01-31 17:33:27 +01:00
cd223e88ba
Fixing comment
2017-01-31 16:08:59 +01:00
4102c20ca2
Fixed spelling errors in CNTK_207_...
2017-01-31 10:39:15 +01:00
aa884114cd
added cntk-106b
2017-01-30 13:27:56 -08:00
57cf46fdc5
Reamed from CNTK_206... to CNTK_207...
2017-01-30 16:18:32 +01:00
87ab0d3651
merge from master
2017-01-30 16:16:57 +01:00
68e9cbfb35
Added references to notebook
2017-01-30 16:14:20 +01:00
2d1782d8bb
Renamed notebook to CNTK_206_Training_with_Sampled_Softmax
2017-01-30 15:54:23 +01:00
1319c698f8
Fixed bugs in notebook
2017-01-30 15:51:33 +01:00
dd4398fb33
added plotting to notebook
2017-01-30 14:38:20 +01:00
655937f94f
Clean up FF tutorial
2017-01-30 13:43:16 +01:00
29d9969e20
Fixed some errors in notebook
2017-01-30 13:41:19 +01:00
877c9aa7a3
re-factored notebook
2017-01-29 19:19:43 +01:00
74b978b5be
update notebook
2017-01-27 22:21:36 +01:00
1cd29d2de9
Increased iterations in isFast mode
2017-01-27 11:55:31 -08:00
9108aa35df
Merge branch 'master' of https://github.com/Microsoft/CNTK into sayanpa/gantut
2017-01-26 21:43:42 -08:00
469aea6c9e
Added GAN tut + test data path
2017-01-26 21:00:07 -08:00
abb201ef03
Added GAN tutorial updates
2017-01-26 18:58:13 -08:00
faa973b8f5
Added GAN tutorial and fixed test threshold
2017-01-26 18:40:28 -08:00
9964293f8f
Added GAN tutorial with CR feedback
2017-01-26 17:57:57 -08:00
02709cafc1
Merge branch 'master' into vistepan/tensorboard-support
2017-01-26 15:53:51 -08:00
e393240779
Undo renaming ProgressPrinter.
2017-01-26 15:49:18 -08:00
57d228c45d
Updated notebook
2017-01-26 18:09:12 +01:00
70626b421f
adding notebook
2017-01-26 17:36:07 +01:00
f94ebcf4f6
Adding basic GAN tutorial
2017-01-25 16:07:19 -08:00
bf8e3ef3ab
Carrying forward git pr 1311 - render cartpole + clear conda kernel to default
2017-01-23 15:16:12 -08:00
e927c166b6
Carrying forward git pr 1311 changes + disable cartpole render for test
2017-01-23 15:08:30 -08:00
7ebd5d011c
Merged the fixes from pr1311 and turn off cart-pole rendering for test
2017-01-23 14:49:43 -08:00
e9aa3f83ed
Revert "Merge branch 'eveselov-master'"
...
This reverts commit db4e8a46cf373a436cb1
2017-01-23 10:13:58 -08:00
f51db94793
Merge branch 'master' of https://github.com/eveselov/CNTK into eveselov-master
2017-01-23 10:02:20 -08:00
45e429d0f3
Initial support for writing TensorBoard event files.
2017-01-20 22:51:32 -08:00
0451daa7c3
Bump versions
2017-01-20 17:26:07 +01:00
3ac55535fc
Integrate alrezni/v2_default_unit_gain into master
2017-01-20 02:48:55 -08:00
9b6c6bde34
Add a default (true) for the unit-gain flag value
2017-01-20 09:27:36 +01:00
f514609836
Updated tutorial with CR feedback
2017-01-19 13:05:33 -08:00
f51a85050c
Merge branch 'master' of https://github.com/Microsoft/CNTK into sayanpa/cntk206
2017-01-19 08:42:06 -08:00
2d91bad7df
CNTK_203_Reinforcement_Learning_Basics_test.py: bump timeout
...
Also, allow a bit more total episodes in fast mode of the notebook.
2017-01-19 13:51:43 +01:00
cf6f50cc26
addressed comments from CR
2017-01-18 19:36:30 -08:00
ba42b6b1a4
Updates to CNTK 102 and 103B with CR feedback
2017-01-18 13:58:58 -08:00
c6147735a9
Added timeseries with LSTM basics tutorial
2017-01-18 13:12:24 -08:00
76ad89a6b9
A couple of bugs fixed in CNTK_203_Reinforcement_Learning_Basics tutorial:
...
a) The order of loss function arguments was incorrect, causing crash in training
b) env resetting moved from outside to inside the loop to allow visualization on all iterations
c) env rendering uncomented to allow visualization of the model
2017-01-17 23:10:33 -08:00
21d0bbf614
Updated the notebook to use layers library
2017-01-17 21:22:34 -08:00
3dd737f500
Bump version
2017-01-16 20:15:06 +01:00
3ab246855b
Add unit-gain momentum flag
...
This commit contains a breaking change: it adds a flag to both
C++ and python Learner API that indicates if the momentum should be applied in
the regular fashion or as a unit-gain filter. This flag is a required
parameter for learners that use momentum (momentum_sgd, nesterov_sgd and
adam_sgd).
2017-01-16 09:39:39 +01:00
43b30a1c41
Integrate chazhang/reader_fix into master
2017-01-13 17:58:30 -08:00
80321d9148
Added CNTK105 autoencoder tutorial
2017-01-13 11:12:47 -08:00
4c03e2c0a5
Revision based on CR.
2017-01-13 08:52:37 -08:00
b28f579131
Bug fix. More config change.
2017-01-13 08:52:36 -08:00
abc1fc92a7
Bug fixes on crop transform. Change CNTK config files to new crop transform.
2017-01-13 08:52:36 -08:00
0fc7cdc3ae
fix an issue on CNTK 202 introduced with new handling of sequence inputs
2017-01-12 18:15:30 -08:00
88b6c328a8
Address CR comments
2017-01-12 15:56:52 +01:00
48d79a3518
CNTK_203_Reinforcement_Learning_Basics.ipynb: just spacing
2017-01-12 13:59:56 +01:00
cb66befcf2
CNTK_203_Reinforcement_Learning_Basics.ipynb: add test
2017-01-12 13:59:56 +01:00
4665583810
Make Jupyter Notebook tests device-aware
2017-01-12 13:59:56 +01:00
6c73818bf7
[CNTK Core] Change BatchNormalization default to CuDnn in both BrainScript and V2 C++/Python, and add fallback to cntk engine when not supported by cudnn
...
Fix epsilon in CuDnnBatchNormalization to be consistent in forward/backward
Fix BatchNormalization parameter gradient accumulation
Add message on CuDnnBatchNormalization forward infer batch size limitation, and adjust CNTK_202_Language_Understanding eval minibatch size to avoid CUDA failure
2017-01-11 01:00:09 -08:00
4b31eda24f
Merge remote-tracking branch 'origin/master' into mahilleb/Mnist13
2017-01-10 08:05:44 +01:00
6dc09565bb
CNTK_103B_MNIST_FeedForwardNetwork.ipynb: add test
...
Also, let run MNIST tests against the default (end-user) data location.
2017-01-10 05:42:30 +01:00
fe24966f7e
CNTK v2 library: Change BlockFunction implementation to store the mapping of block function arguments and outputs in the variables themselves instead of maps on teh side
2017-01-05 10:38:57 -08:00
fa975f8476
Integrate sayanpa/fixcntk103b into master
2017-01-04 19:42:10 -08:00
8a07efdfec
bug fix in testing model data source
2017-01-04 13:58:36 -08:00
d0f5250e43
Fix git hub rendering of CNTK 204 notebook
2017-01-02 11:35:39 -08:00
b83484401a
CNTK v2 library: BlockFunction feature to create heirarchical function graphs
2016-12-24 00:28:22 -08:00
f9ad4aafcf
Remove text_minibatch_source
2016-12-23 02:35:20 -08:00
a01e609450
Integrate nikosk/highlight into master
2016-12-22 17:28:57 -08:00
69fb965fd3
Update file to work for both Python 2 and 3.
2016-12-22 17:06:05 -08:00
a8fa4554ee
Removed annoying ?highlight from pythondocs links
2016-12-22 16:41:28 -08:00
89ffdf4ee4
Bump version
2016-12-22 10:52:53 +01:00
1e67e8cd13
figured out the discrepancy between test and local
2016-12-21 12:08:12 -08:00
0233ff98e9
Merge branch 'wilrich/amitaga/tensorLibPerf' into nikosk/neuralstyle
2016-12-21 11:26:38 -08:00
e54fbc7589
update baseline with correct images
2016-12-21 11:24:53 -08:00
bb7cd37b8e
Matplotlib inline
2016-12-21 18:17:54 +01:00
d5b93430ba
Fix shape changes
2016-12-21 18:12:13 +01:00
74968295d9
Cleaned up imports; fixed error handling
2016-12-21 13:30:23 +01:00
b90fad0689
support for python 2.7
2016-12-20 22:04:48 -08:00
acb1209399
switched kernel back to generic
2016-12-20 21:28:05 -08:00
7addd34aff
Merge remote-tracking branch 'origin/master' into nikosk/neuralstyle
2016-12-20 18:33:59 -08:00
3b1ac91a0d
updated tutorial with approved style and content
...
updated test
2016-12-20 18:32:59 -08:00
8f4cad8f40
Enable Python 2.7
2016-12-19 13:49:05 +01:00
1d5f5a7d06
fixed the notebook/test to work on cpu and gpu
...
removed images from tutorials folder
2016-12-16 10:26:29 -08:00
0a21b3644a
addressed all CR feedback
2016-12-15 19:12:38 -08:00
ecaa6cf423
addressed most CR feedback
2016-12-15 18:47:02 -08:00
ae653c6fba
Incorporated CR feedback
2016-12-14 12:29:33 -08:00
867e08271d
Fix CNTK_203_Reinforcement_Learning_Basics.ipynb
...
Remove an extra argument from save_model calls.
2016-12-14 11:38:45 +01:00
51db935f01
neural artistic style transfer
2016-12-13 18:21:40 -08:00
80f70b638d
Bump version
2016-12-12 18:59:15 +01:00
5a36689a4f
Integrate t-ivrodr/value_to_csr into master
2016-12-09 06:17:28 -08:00
f7d0150221
Add test for CNTK_202_Language_Understanding.ipynb
2016-12-09 13:20:12 +01:00
2b1237cd5a
CNTK_204_Sequence_To_Sequence.ipynb: use local data if present
...
Also, minor changes + drop output.
2016-12-09 13:20:12 +01:00
fd5e3d4e20
Addressing CR comments in 2nd iteration
2016-12-09 12:58:23 +01:00
2a9d000f06
Rename find_nodes_by_name to find_all_with_name and add find_by_name
2016-12-09 09:47:25 +01:00
253e0a1400
Integrate alrezni/v2_save_model into master
2016-12-05 05:03:37 -08:00
3a085d66bf
Remove legacy format flag from SaveModel
...
Use protobuf-based serialization for model saving in V2. Refactor
saving models in legacy format to a separate function (in Internal
namespace).
2016-12-05 13:16:40 +01:00
17e4a07cff
Added CR comments and updated test
2016-12-02 01:15:45 -08:00
6f89bfed7b
Adding a time series tutorial with Pandas and Numpy interop for finance data
2016-12-02 01:15:45 -08:00
e2b7fee2d9
Incorporated CR feedback
2016-12-01 11:24:52 -08:00
e6c96f8dd0
Added couple of clarifying comments without any test
2016-12-01 11:24:51 -08:00
0c50788e41
Fix image hand on tutorial.
2016-11-28 18:56:43 -08:00
baad5f3630
Bump version
2016-11-25 20:31:14 +01:00
494e39f76a
Added Sequence to sequence tutorial and test
2016-11-23 14:18:14 -08:00
ce301e2ec2
Bump version
2016-11-21 15:21:43 +01:00
b052cece03
Tutorials: more fixes
2016-11-18 19:15:15 +01:00
e442dd4273
Tutorials: more fixes
2016-11-18 14:47:17 +01:00
49125b5e4c
Adapting to stricter shape checks
2016-11-18 12:21:17 +01:00
3ec23a54d9
Fix Tutorials/CNTK_{101,102} wrt learning rate schedule.
...
Add test for them.
2016-11-17 20:39:10 +01:00
e8c9865977
Fix some typos, tune
2016-11-17 20:39:10 +01:00
4923b404bb
Applying comments
2016-11-17 08:48:23 -08:00
e189d23c0d
Fix
2016-11-17 02:01:00 -08:00
6344a95ca7
Remove outputs
2016-11-17 01:57:52 -08:00
98b445c9f2
Fix RL tutorial
2016-11-17 01:53:17 -08:00
cbc27a94ca
fixed 202 and 203 tutorials
2016-11-16 11:09:01 +01:00
99389b3032
Update CNTK_203_Reinforcement_Learning_Basics.ipynb
...
Fixed the "unknown character" problems.
Fixed the problem which "Bellmann" was not part of the hyperlink.
2016-11-14 22:23:40 -08:00
7f5c78d4f9
Update README.md
2016-11-14 21:19:20 +01:00
babf078b49
Integrate alrezni/v2_scratch into master
2016-11-14 10:59:42 -08:00
90ff04f254
fixed tutorials and VS links
2016-11-14 16:24:56 +01:00
3d1e93badf
Addressed CR comments
2016-11-14 16:24:52 +01:00
fba28bf0ed
Fixed python commons and eval clients
2016-11-14 16:24:50 +01:00
9231ece8ce
Restructuring examples and tutorials
2016-11-14 16:24:45 +01:00
e58abac01c
Move { -> Examples/}Tutorials, Examples/{Tutorial -> Tutorials/LogisticRegressionAndMultiClass}
2016-10-21 10:42:20 +02:00
0bbfdbef99
cherry-picked: numGradientBits is now a vector; simplified logging of MB scaling
2016-09-16 19:59:33 -07:00
3d5d78fbb7
updated pretrained model
2016-09-15 22:24:05 -07:00
922cc963f9
updated ImageHandsOn to new defaults
2016-09-15 22:15:42 -07:00
ec0adbe79b
bug fix in SLUHandsOn_Solution4; updated pre-trained models for this tutorial
2016-09-15 20:56:46 -07:00
92f49753d7
updated SLUHandsOn to after fsAdagrad and default patch
2016-09-15 20:00:13 -07:00
9a04db5536
updated ImageHandsOn
2016-08-25 21:16:14 -07:00
bc60bbea4a
updated ImageHandsOn parallelization setup
2016-08-25 21:03:18 -07:00
6412e9b796
updated ImageHandsOn parallelization setup
2016-08-25 21:00:17 -07:00
769b2602a2
updated SLUHandsOn tests
2016-08-24 18:15:09 -07:00
d9c7e82031
OptimizedRNNStackNode: renamed some variables, renamed recurrentOps to camelCase, added weigth inference
2016-08-24 17:17:23 -07:00
54f096083d
merged from mahilleb/CuDnn5Test
2016-08-22 18:51:22 -07:00
363f1e5809
changed some more baselines from ErrorPrediction to ClassificationError
2016-08-22 13:27:40 -07:00
536e264749
renamed ErrorPrediction to ClassificationError, second attempt
2016-08-22 11:16:39 -07:00
9bd57387bb
renamed ErrorPrediction to ClassificationError
2016-08-22 10:30:17 -07:00
d2391a986d
changed EmbeddingLayer{} and LSTMBlock{} to 'heNormal' initialization
2016-08-20 00:02:10 -07:00
db74d6b468
changed ImageHandsOn from "gaussian" to "heNormal" initialization, and also most layers defaults in CNTK.core.bs
2016-08-19 23:34:17 -07:00
832a685d29
Merge branch 'fseide/kdd' of https://github.com/Microsoft/cntk into fseide/kdd
2016-08-17 09:54:09 -07:00
0dcc4b0844
updated KDD Tutorial package
2016-08-15 08:45:09 -07:00
453371a361
added a new pretrained model
2016-08-14 13:34:35 -07:00
e54b974fa1
added KDD Tutorial ZIP
2016-08-14 13:05:23 -07:00
c3855959d2
reduced #epochs in ImageHandsOn from 30 to 10 due to runtime
2016-08-14 13:00:05 -07:00
2ff67d266d
updated samples
2016-08-14 12:15:42 -07:00
5812c26c90
two missing files
2016-08-14 11:40:14 -07:00
ac381271dc
updated ResNet solutions
2016-08-14 11:37:48 -07:00
83ba229429
fixed up Tutorials in Solution;
...
modified ImageHandsOn ResNet
2016-08-14 10:33:01 -07:00
42adf0c64b
added SLUHandsOn Solution4
2016-08-14 10:03:08 -07:00
b6742c23ff
added _Solution versions of SLUHandsOn up to Solution4
2016-08-13 22:11:28 -07:00
7e764a0122
moved away the .wl files
2016-08-13 21:13:29 -07:00
4c61d0c385
SLUHandsOn starting model
2016-08-13 20:14:49 -07:00
f483334ff0
created SLUHandsOn.md
2016-08-13 20:04:44 -07:00
66c82d478f
adapted ImageHangsOn to tutorial style
2016-08-13 15:20:00 -07:00
05a86bb32b
moved ATIS tutorial to Tutorials location
2016-08-13 15:02:16 -07:00
67368ff186
added ResNet pretrained model
2016-08-13 14:48:06 -07:00
af709d4cfa
added Solution versions of ImageHandsOn
2016-08-13 14:38:11 -07:00
0d0a4d8148
added Solution variants of ImageHandsOn
2016-08-13 12:48:18 -07:00
f5e77e4efb
minor fixes
2016-08-13 12:39:27 -07:00
c7df94d68c
creating final version of ImageHandsOn
2016-08-13 11:21:37 -07:00
752ac631da
created new folder hierarchy for Tutorials
2016-08-13 10:11:17 -07:00
Mark Hillebrand
Frank Seide
REDMOND\sayanpa
Project Philly
Thilo Will
Chris Basoglu
Guenther Schmuelling
Willi Richert
NORTHAMERICA\vistepan
Alexey Reznichenko
Eugene Veselov
Cha Zhang
Nikos Karampatziakis
KeDengMS
Amit Agarwal
Aayush Garg
Ivan Rodriguez
Emad Barsoum
Philipp Kranen
Matthew Chan