mbinkowski / nntimeseries Goto Github PK

View Code? Open in Web Editor NEW

207.0 207.0 67.0 23.58 MB

Python 29.21% Jupyter Notebook 70.79%

nntimeseries's People

Contributors

Stargazers

Watchers

nntimeseries's Issues

GPU Version and run on Amazon AWS

Is there a GPU version of the code. i have tried running it my Intel Xeon server with 4 x Titan X Pascal GPUs , unfortunately it only uses CPU . Can it be run on Amazon AWS or on GPUs ,

where's cluster_results.pkl?

In results_display.py, all_results = pd.read_pickle(WDIR + '/results/cluster_results.pkl'). But there's no cluster_results.pkl in the data file.

Questions about `LSTM2.py`

Hello,

Thanks for making your code public on Github. I was wondering if there was something wrong with param_dict you defined in LSTM2.py:

# dictionary of hyperparameter lists across which the grid search will run
param_dict = dict(
    #input parameters
    ......
    target_cols=[1],    # 'default' or list of names of columns to predict  
    ......
)

When I run your code, I got following mistakes:

using <class '__main__.LSTMmodel'> to build the model
using <class 'nnts.artificial.ArtificialGenerator'> to draw samples
Traceback (most recent call last):
  File "LSTM2.py", line 106, in <module>
    runner.run(LSTMmodel, log=log, limit=1)
  File "/Users/jack-cheng/Desktop/毕业设计/时间序列/SOCNN/nntimeseries/nnts/utils.py", line 128, in run
    model = model_class(data, params, os.path.join(WDIR, 'tensorboard'))
  File "/Users/jack-cheng/Desktop/毕业设计/时间序列/SOCNN/nntimeseries/nnts/utils.py", line 228, in __init__
    self.idim, self.odim = self.G.get_dims(cols=self.target_cols)   
  File "/Users/jack-cheng/Desktop/毕业设计/时间序列/SOCNN/nntimeseries/nnts/utils.py", line 365, in get_dims
    return self.get_dim(), len(self.get_target_col_ids(cols=cols))
  File "/Users/jack-cheng/Desktop/毕业设计/时间序列/SOCNN/nntimeseries/nnts/artificial.py", line 223, in get_target_col_ids
    assert hasattr(cols, '__iter__')
AssertionError

When I set target_cols=['default'], it worked and produced result similar to that you proposed in the paper. Could you please tell me what do you mean by setting target_cols=[1]?

Pytorch version implementation

https://github.com/Fangyh09/Autoregressive-Convolutional-Neural-Networks.Pytorch

Reshape cannot infer the missing input size for an empty tensor

There was a typo also on line 159 of SOCNN.py when trying the shared_final_weights option

out = TimeDistributed(Dense(output_length ...

that I changed to

out = TimeDistributed(Dense(self.output_length ...

And got the training started with:

I tensorflow/core/common_runtime/gpu/gpu_device.cc:975] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 1070, pci bus id: 0000:01:00.0)
Total model parameters: 7196
---current learning rate: 0.00100000
Epoch 1/1000

But then the following error happend then:

Caused by op 'softmax/Reshape_1', defined at:
  File "nntimeseries/nnts/models/SOCNN.py", line 189, in <module>
    runner.run(SOCNNmodel, log=log, limit=6)
  File "nntimeseries/nnts/utils.py", line 150, in run
    model = model_class(data, params, os.path.join(WDIR, 'tensorboard'))
  File "nntimeseries/nnts/utils.py", line 263, in __init__
    self.nn, self.io_func, self.callbacks = self.build()
  File "nntimeseries/nnts/models/SOCNN.py", line 152, in build
    sig = TimeDistributed(Activation('softmax'), name='softmax')(sig)
  File "anaconda3/lib/python3.6/site-packages/keras/engine/topology.py", line 578, in __call__
    output = self.call(inputs, **kwargs)
  File "anaconda3/lib/python3.6/site-packages/keras/layers/wrappers.py", line 180, in call
    y = K.reshape(y, (-1, input_length) + output_shape[2:])
  File "anaconda3/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 1568, in reshape
    return tf.reshape(x, shape)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gen_array_ops.py", line 2630, in reshape
    name=name)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 763, in apply_op
    op_def=op_def)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 2327, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1226, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): Reshape cannot infer the missing input size for an empty tensor unless all specified input sizes are non-zero
	 [[Node: softmax/Reshape_1 = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _device="/job:localhost/replica:0/task:0/gpu:0"](softmax/Softmax, softmax/Reshape_1/shape)]]
	 [[Node: add/_445 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_10390_add", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

I guess I have to explicitly set shape for the sig of sig = TimeDistributed(Activation('softmax'), name='softmax')(sig) ?

FileNotFoundError: [Errno 2] No such file or directory: SOCNN.pkl

Thanks for sharing the code @mbinkowski of your interesting paper. However I had some issues trying to make the model run for the household data

For example, I got the following:

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/petteri/anaconda3/lib/python3.6/site-packages/pandas/io/pickle.py", line 56, in try_read
    with open(path, 'rb') as fh:
FileNotFoundError: [Errno 2] No such file or directory: '/home/petteri/Dropbox/Data/nntimeseries/results/household_SOCNN.pkl'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "nntimeseries/nnts/models/SOCNN.py", line 158, in <module>
    runner.run(SOCNNmodel, limit=1)
  File "nntimeseries/nnts/utils.py", line 110, in run
    self.cresults = self._read_results()
  File "/nntimeseries/nnts/utils.py", line 64, in _read_results
    results = [v for k, v in pd.read_pickle(self.save_file).T.to_dict().items()]
  File "anaconda3/lib/python3.6/site-packages/pandas/io/pickle.py", line 68, in read_pickle
    return try_read(path, encoding='latin1')
  File "anaconda3/lib/python3.6/site-packages/pandas/io/pickle.py", line 61, in try_read
    with open(path, 'rb') as fh:
FileNotFoundError: [Errno 2] No such file or directory: '/Data/nntimeseries/results/household_SOCNN.pkl'

And on line 110 of utils.py, there is an attempt to read some results from pickled file that I don't have as I have never run the model yet successfully?

I replaced that with empty dictionary: self.cresults = {}, and had to add WDIR manually to some files, and then hit that when putting limit higher than 1:

using <class '__main__.SOCNNmodel'> to build the model
using <class 'nnts.household.HouseholdGenerator'> to draw samples
Traceback (most recent call last):
  File "nntimeseries/nnts/models/SOCNN.py", line 159, in <module>
    runner.run(SOCNNmodel, log=log, limit=6)
  File "nntimeseries/nnts/utils.py", line 150, in run
    model = model_class(data, params, os.path.join(WDIR, 'tensorboard'))
  File "nntimeseries/nnts/utils.py", line 263, in __init__
    self.nn, self.io_func, self.callbacks = self.build()
  File "nntimeseries/nnts/models/SOCNN.py", line 136, in build
    out = outL(main)
  File "anaconda3/lib/python3.6/site-packages/keras/engine/topology.py", line 578, in __call__
    output = self.call(inputs, **kwargs)
  File "anaconda3/lib/python3.6/site-packages/keras/layers/local.py", line 159, in call
    x_aggregate = K.concatenate(xs, axis=0)
  File "anaconda3/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 1553, in concatenate
    return tf.sparse_concat(axis, tensors)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/sparse_ops.py", line 227, in sparse_concat
    inds, vals, shapes, axis, name=name))
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gen_sparse_ops.py", line 428, in _sparse_concat
    concat_dim=concat_dim, name=name)
  File "anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 546, in apply_op
    (input_name, op_type_name, len(values), num_attr.minimum))
ValueError: List argument 'indices' to 'SparseConcat' Op with length 0 shorter than minimum length 2.

Any idea of the errors?

Your paper was an interesting read and thank you for making available the code. After following example.ipynb, I was wondering how I can print/visualize the weightings of the different 'candidate predictors'. It should be the case that B and D have high weightings for target variable E since E is a sum of last 10 values of B multiplied by D.
Which variable in SOCNN.py is the weighting frame? And do you have any suggestions what the best way is to print this?

Thank you in advance

ValueError: If printing histograms, validation_data must be provided, and cannot be a generator.

I am running

TF = 1.6 , Tensorboard = 1.5.0 and Keras 2.1.5 but i am getting this error below:

Developer@DEEPGPU:~/Downloads/nntimeseries-master/nnts/models# python ./CNN.py --dataset=artificial
....
...
...
..
...
Total model parameters: 3451
WARNING:tensorflow:From /home/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/base.py:198: retry (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.
Instructions for updating:
Use the retry module or similar alternatives.
Epoch 1/1000
---current learning rate: 0.00100000
124/124 [==============================] - 13s 103ms/step - loss: 1.2770 - val_loss: 2.7545

---current best val_loss: 2.75454
960/960 [==============================] - 0s 113us/step

--- test_loss: 2.290810
Traceback (most recent call last):
File "./CNN.py", line 109, in
runner.run(CNNmodel, log=log, limit=1)
File "/root/Downloads/nntimeseries-master/nnts/utils.py", line 130, in run
model_results, nn = model.run()
File "/root/Downloads/nntimeseries-master/nnts/utils.py", line 287, in run
verbose=self.verbose
File "/home/anaconda3/lib/python3.6/site-packages/keras/legacy/interfaces.py", line 91, in wrapper
return func(*args, **kwargs)
File "/home/anaconda3/lib/python3.6/site-packages/keras/engine/training.py", line 2262, in fit_generator
callbacks.on_epoch_end(epoch, epoch_logs)
File "/home/anaconda3/lib/python3.6/site-packages/keras/callbacks.py", line 77, in on_epoch_end
callback.on_epoch_end(epoch, logs)
File "/home/anaconda3/lib/python3.6/site-packages/keras/callbacks.py", line 820, in on_epoch_end
raise ValueError('If printing histograms, validation_data must be '
ValueError: If printing histograms, validation_data must be provided, and cannot be a generator.

mbinkowski / nntimeseries Goto Github PK

nntimeseries's People

Contributors

Stargazers

Watchers

Forkers

nntimeseries's Issues

GPU Version and run on Amazon AWS

where's cluster_results.pkl?

Questions about `LSTM2.py`

Pytorch version implementation

Reshape cannot infer the missing input size for an empty tensor

FileNotFoundError: [Errno 2] No such file or directory: SOCNN.pkl

Printing weighting frame

ValueError: If printing histograms, validation_data must be provided, and cannot be a generator.

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent