 mandlm
|
c5fbe764a6
|
Limited ErrorPlotter buffer size
|
2015-11-16 17:27:42 +01:00 |
|
mandlm
|
0f7c617f9c
|
Back to learning all samples
|
2015-11-15 16:09:25 +01:00 |
|
mandlm
|
5e32724b1a
|
Layers average their output sums
|
2015-11-15 16:08:49 +01:00 |
|
mandlm
|
44760d0402
|
Debugging-setup with a single training sample to check the lack of learning progress
|
2015-11-01 15:01:43 +01:00 |
|
mandlm
|
1d343a079a
|
Moved MNIST sample to QImage conversion to MNIST loader
|
2015-11-01 15:00:32 +01:00 |
|
mandlm
|
eecd7a0fe6
|
Querying of a single MNIST sample
|
2015-11-01 14:49:02 +01:00 |
|
mandlm
|
ab9dcfbd35
|
Clean learner-thread termination on program exit
|
2015-11-01 12:36:23 +01:00 |
mandlm
|
1a0d2b9ea7
|
Merge pull request #1 from mandlm/digits
Digits
|
2015-10-31 15:08:01 +01:00 |
|
mandlm
|
8d01edb7a1
|
Removed naked double * calls
|
2015-10-31 14:59:10 +01:00 |
|
mandlm
|
cd1101dfe2
|
Learning from and displaying of digit samples
|
2015-10-31 14:58:49 +01:00 |
|
mandlm
|
d98ec63fbd
|
Completed loading of MNIST Databases
|
2015-10-31 12:28:35 +01:00 |
|
mandlm
|
87e267d65f
|
Implemented parsing of MNIST files
|
2015-10-29 16:00:58 +01:00 |
|
mandlm
|
83b4562a29
|
Removed useless training images, added MNIST database instead
See http://yann.lecun.com/exdb/mnist/
|
2015-10-29 13:06:30 +01:00 |
|
mandlm
|
5778afa121
|
Display training digits in UI while training
|
2015-10-27 18:20:50 +01:00 |
|
mandlm
|
f9be5ca717
|
Load and train handwritten digits
|
2015-10-27 15:33:54 +01:00 |
|
mandlm
|
650b4be9fc
|
Added forward-feeding of double-ptrs
|
2015-10-27 15:33:10 +01:00 |
|
mandlm
|
ec654c0e5f
|
Improved icon visibility
|
2015-10-27 15:32:46 +01:00 |
|
mandlm
|
81d1f54c98
|
Started loader-class for 8x8 pixel digit training data
|
2015-10-26 22:05:50 +01:00 |
|
mandlm
|
80da55a2b8
|
Added handwritten digits as training data
Source: http://www.cs.nyu.edu/~roweis/data.html
|
2015-10-26 21:34:13 +01:00 |
|
mandlm
|
aa61792911
|
Fixed a performance problem with a QListWidget while inserting and scrolling
|
2015-10-26 20:51:36 +01:00 |
|
mandlm
|
37b5153d6a
|
Cleaned up code to work with MinGW/GCC Kit
|
2015-10-26 20:34:26 +01:00 |
|
mandlm
|
1e716979a9
|
Reduced the progress update messages to take load from the UI,
implemented load-or-create in test funtion
|
2015-10-26 19:50:01 +01:00 |
|
mandlm
|
9bb927d2d2
|
Fixed a problem with multiple connects
|
2015-10-26 17:54:06 +01:00 |
|
mandlm
|
22dba2cb27
|
Added time measurement
|
2015-10-26 09:19:48 +01:00 |
|
mandlm
|
24b6969ab7
|
Added a error-plotter widget to NeuroUI
|
2015-10-26 08:41:59 +01:00 |
|
mandlm
|
b899c6f55e
|
Replaced a few int with size_t to fix 64 bit warnings.
|
2015-10-26 07:33:45 +01:00 |
|
mandlm
|
99ef63e019
|
Added simple (de-)serialization of (trained) nets
|
2015-10-25 17:40:22 +01:00 |
|
mandlm
|
249bd22b67
|
Update README.md
|
2015-10-25 11:49:45 +01:00 |
|
mandlm
|
91427954c7
|
Added an icon to NeuroUI
|
2015-10-25 10:00:48 +01:00 |
|
mandlm
|
2b84b8a054
|
Added a progress bar to NeuroUI
|
2015-10-25 09:51:09 +01:00 |
|
mandlm
|
6943fc0116
|
Threaded learning and signaling in the Qt UI
|
2015-10-24 18:03:07 +02:00 |
|
mandlm
|
c9fb9b9fa8
|
Qt GUI initial import
|
2015-10-24 14:44:33 +02:00 |
|
mandlm
|
4eb232b1e9
|
Working net, calculates the mean value of two inputs
|
2015-10-23 22:16:12 +02:00 |
|
mandlm
|
6ed30e56c4
|
Finished the max-value net (2/3/1 neurons) with 10k learning iterations. No good.
|
2015-10-22 22:09:35 +02:00 |
|
mandlm
|
d4a22ecae7
|
Simple max(a, b) test-net without iterations
|
2015-10-22 16:03:10 +02:00 |
|
mandlm
|
f22e4221a1
|
Fixed feed-forward algo
|
2015-10-22 16:02:27 +02:00 |
|
mandlm
|
3d30346f2d
|
Implemented dynamic learning
|
2015-10-18 22:05:18 +02:00 |
|
mandlm
|
6ef1f9657c
|
Backprop seems to be working, yay
|
2015-10-18 21:20:37 +02:00 |
|
mandlm
|
a79abb5db1
|
First implementation of weight updates. Very slow rate of change in the output value.
|
2015-10-17 22:05:27 +02:00 |
|
mandlm
|
de06daaad3
|
Fixed gradient computation
|
2015-10-17 21:02:10 +02:00 |
|
mandlm
|
26a51ce476
|
Hidden neuron gradients complete
|
2015-10-16 23:23:27 +02:00 |
|
mandlm
|
370451c2e6
|
Calculation of hidden neuron gradients (partial)
|
2015-10-16 22:59:04 +02:00 |
|
mandlm
|
ea454b58c6
|
Added 64-bit configuration and support
|
2015-10-15 22:37:13 +02:00 |
|
mandlm
|
953db0a17e
|
Added README to vcxproj
|
2015-10-15 22:30:53 +02:00 |
|
mandlm
|
a7c0fb8b33
|
Create README.md
|
2015-10-15 22:19:48 +02:00 |
|
mandlm
|
ce88f690cf
|
More back-propagation code, calculation of output-neuron gradients
|
2015-10-15 22:16:34 +02:00 |
|
mandlm
|
7ba16e9e9d
|
Renamed a few things, started working on back-propagation
|
2015-10-15 19:18:26 +02:00 |
|
mandlm
|
2f556d1b92
|
Added a (hacky) bias neuron
|
2015-03-24 13:45:38 +01:00 |
|
mandlm
|
e3a804242c
|
Split up to different source files, entry-point for back propagation
|
2015-03-23 21:58:30 +01:00 |
|
mandlm
|
c8e08193f1
|
Initial forward-feeding net
|
2015-03-23 19:28:29 +01:00 |