Discussions
question on Convolutional Network example
hi
do you have any CNN example code
for example, CNN for mnist something like that
thx
Posted by kwang nam about 8 years ago
How can I reuse a trained model?
Hello.
I'm writing a network that predicts the probabilities of each symbol(character) from the previous context. I've written codes for training, and it seems work well.
I want to save the total network settings(including trained weights) then reload and use them later.
In other words, I want to separate the training from the part that applies actual data to the trained network. Can I do? If so, what should I do?
Thank you for maintaining such a nice project, anyway!
Posted by Jaeju Kim about 8 years ago
ImportError (MidiOutFile)
Trying to run the Your First Model tutorial, but I get the error that follows below. I've had to install the most recent stable version of Theano (0.7.0) instead of the bleeding edge one indicated in the install instructions due to some compatibility issues with Python 3.2, in case that could somehow be the root of the problem. Unlike this fellow (http://www.opendeep.org/v0.0.8/discuss/55b3b090a96deb1900990d7e) I haven't figured out the cause, though I'll let you know if I work it out. Running Kali Linux, Python 3.2, Pillow 2.7.0, numpy 1.9.2, and scipy 0.16.0 in addition to the aforementioned theano 0.7.0.
Incidentally, thanks for putting this package together. A lot of work goes into something like this.
Traceback (most recent call last):
File "/home/usrname/PycharmProjects/599/OpenDeepTut.py", line 3, in <module>
from opendeep.models.model import Model
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/__init__.py", line 26, in <module>
from . import data
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/data/__init__.py", line 8, in <module>
from .standard_datasets import *
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/data/standard_datasets/__init__.py", line 4, in <module>
from .midi import *
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/data/standard_datasets/midi/__init__.py", line 3, in <module>
from .jsb_chorales import JSBChorales
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/data/standard_datasets/midi/jsb_chorales.py", line 14, in <module>
from opendeep.utils.midi import midiread
File "/home/usrname/Downloads/opendeep/opendeep/opendeep/utils/midi/__init__.py", line 15, in <module>
from MidiOutFile import MidiOutFile
ImportError: No module named MidiOutFile
Posted by Jason Johnson about 8 years ago
please check this
when I run your rbmminst sample I got this error
Traceback (most recent call last):
File "E:\project\python\test\test\opendeep_rbm.py", line 72, in <module>
tile_shape=closest_to_square_factors(rbm.hidden_size),
AttributeError: 'RBM' object has no attribute 'hidden_size'
Posted by kwang nam about 8 years ago
please check this
I got this error, I can't find getSubSet() on the source, Did you change it ? File "E:\project_luna\cuda\open_deep\test\opendeeptest2.py", line 36, in <module>
test_xs, _ = mnist.getSubset(datasets.Dataset.test_inputs)
AttributeError: 'MNIST' object has no attribute 'getSubset'
Posted by kwang nam about 8 years ago
check this one
I got this error, Now I am using phthon 2.7
INFO:opendeep.optimization.optimizer:Initializing optimizer <class 'opendeep.optimization.adadelta.AdaDelta'>
INFO:opendeep.optimization.optimizer:Optimizer config args: {'grad_clip': None, 'save_freq': 1000000, 'min_batch_size': 1, 'stop_threshold': inf, 'decay': 0.95, 'hard_clip': False, 'learning_rate': 1e-06, 'batch_size': 100, 'stop_patience': 1, 'epochs': 10, 'lr_decay': None, 'model': <opendeep.data.standard_datasets.image.mnist.MNIST object at 0x00000000143F7630>, 'dataset': <opendeep.models.single_layer.autoencoder.DenoisingAutoencoder object at 0x00000000233BDA58>, 'lr_decay_factor': None}
Traceback (most recent call last):
File "E:\project_luna\cuda\open_deep\test\opendeeptest2.py", line 29, in <module>
optimizer = AdaDelta(dae, mnist)
File "F:\python27\python-2.7.9.amd64\lib\site-packages\opendeep-0.0.9a0-py2.7.egg\opendeep\optimization\adadelta.py", line 72, in __init__
super(AdaDelta, self).__init__(**initial_parameters)
File "F:\python27\python-2.7.9.amd64\lib\site-packages\opendeep-0.0.9a0-py2.7.egg\opendeep\optimization\optimizer.py", line 123, in __init__
"Found %s" % str(type(model))
AssertionError: Optimizer input model needs to be a Model class! Found <class 'opendeep.data.standard_datasets.image.mnist.MNIST'>
Posted by kwang nam about 8 years ago
please check this
I just git latest one and I got this error please check
2015-07-28 23:08:34,829 - opendeep.optimization.optimizer - INFO - Compiling f_learn 1/1 function for model <class '__main__.DenoisingAutoencoder'>...
Traceback (most recent call last):
File "E:\project\python\test\test\opendeeptest.py", line 83, in <module>
optimizer.train()
File "E:\program\python34\python-3.4.3.amd64\lib\site-packages\opendeep\optimization\optimizer.py", line 330, in train
outputs=[train_cost] + self.train_monitors_dict.values(),
TypeError: can only concatenate list (not "dict_values") to list
Posted by kwang nam about 8 years ago
question on example
http://www.opendeep.org/docs/monitors-and-live-plotting
on the sample,
from opendeep.models.single_layer.basic import BasicLayer, SoftmaxLayer
there is no basic (packge) and
can't find BasicLayer , Which one I have to use ?
Posted by kwang nam about 8 years ago
please update the source code on this page
http://www.opendeep.org/docs/tutorial-first-steps
On Example Code
# grab 100 test examples
test_xs, _ = mnist.getSubset(datasets.TEST)
--> change like this
mnist.getSubset(datasets.Dataset.test_inputs)
is test_inputs is right attribute ?
Posted by kwang nam about 8 years ago
error lists I found
hi
thanks again your quick response
I got latest one and I still got these error Please check
1) print () issues
test_cifar10.py,
example)
print len(self.cifar.train_inputs) --> print (len(self.cifar.train_inputs))
2) raise () issues
MidFileParser.py,EventDispatcher.py
example)
raise TypeError, "It is not a valid midi file!" -> raise (TypeError, "It is not a valid midi file!")
3) packaage Path issues
midi.py, MidiOutFile.py,RawOutstreamFile,MidiInFile,RawInstreamFile,MidiFileParser, EventDispatcher
example)
from MidiOutFile import MidiOutFile --> from opendeep.utils.midi.MidiOutFile import MidiOutFile
4) unused packages in python 3 issues
RawOutstreamFile.py, RawInstreamFile, MIdiFileParser
from types import StringType
from cStringIO import StringIO
5) MidiOutFile,EventDispatcher
from constants import * --> from . import constants
6) rnn_gsn,rnn_rbm
cPickle is not used in python 3.4
import cPickle as pickle --> import pickle
7) Optimizer.py
not used I commented
from itertools import izip
8) mnist
I got this error
File "E:\project_luna\cuda\open_deep\test\opendeeptest.py", line 23, in <module>
mnist = MNIST()
File "F:\python34\python-3.4.3.amd64\lib\site-packages\opendeep-0.0.9a0-py3.4.egg\opendeep\data\standard_datasets\image\mnist.py", line 68, in __init__
gzip.open(self.path, 'rb')
UnicodeDecodeError: 'ascii' codec can't decode byte 0x90 in position 614: ordinal not in range(128)
so I changed the code like this gzip.open(self.path, 'rb'),encoding='iso-8859-1'
9) I still get this error
from test import opendeeptest
File "E:\project_luna\cuda\open_deep\test\opendeeptest.py", line 39, in <module>
dae = DenoisingAutoencoder(**config)
File "F:\python34\python-3.4.3.amd64\lib\site-packages\opendeep-0.0.9a0-py3.4.egg\opendeep\models\single_layer\autoencoder.py", line 152, in __init__
**{arg: val for (arg, val) in locals().iteritems() if arg is not 'self'}
AttributeError: 'dict' object has no attribute 'iteritems'
Posted by kwang nam about 8 years ago