ValueError running spatial_stream_cnn.py #9
Comments
|
See #8. The dimension of the input should be of the format [batch_size, X, Y, no_of_channels]. |
|
Can I alter the number of keys in temporal data pickle file?
…________________________________
From: Sahil Wadhwa <[email protected]>
Sent: Monday, February 27, 2017 12:57 AM
To: wadhwasahil/Video-Classification-2-Stream-CNN
Cc: yclai1114; Author
Subject: Re: [wadhwasahil/Video-Classification-2-Stream-CNN] ValueError running spatial_stream_cnn.py (#9)
#4<#4> The dimension of the input should be of the format [batch_size, X, Y, no_of_channels] i.e of dimension 4.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_RDISykqknrNKLhI0rqhH-Re5jsVks5rga7hgaJpZM4MMTjV>.
|
|
you can but still you have to load the pickle into the memory and then increase/decrease the no of keys. |
|
How I should reduce it? Because I want to try your training but I used one day for only running one epoch.
Thanks very much
On 1 Mar 2017, at 1:08 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
you can but still you have to load the pickle into the memory and then increase/decrease the no of keys.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_czvzKTLeq8HA2QGiywQzFvOOGtEks5rhP0pgaJpZM4MMTjV>.
|
|
How would this training result output?
On 1 Mar 2017, at first 1:08 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
you can but still you have to load the pickle into the memory and then increase/decrease the no of keys.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_czvzKTLeq8HA2QGiywQzFvOOGtEks5rhP0pgaJpZM4MMTjV>.
|
|
You can remove according to your choice.
or maybe try this? |
|
Is the main.py file being removed? I couldn't find it. If yes, Can you send me the main.py file?
…Sent from my iPhone
On 1 Mar 2017, at 5:26 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
You can remove according to your choice.
* Load the pickle in the memory. Let's say it has 1K keys.
* Remove 500 keys using python. To do this, iterate through the dictionary and pick only 500 keys and create another pickle from this.
or maybe try this?
http://stackoverflow.com/questions/17521379/python-remove-entry-from-pickle-saved-data
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_dCcj9_oNpVySeQUVTQCYq-MIEYeks5rhTm7gaJpZM4MMTjV>.
|
|
what's main.py, what does it do? |
|
I saw an issue about main.py file which I guess is for testing dataset after done training?
…Sent from my iPhone
On 1 Mar 2017, at 5:45 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
what's main.py, what does it do?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_RT20oPUWyH8YPanjvjOMf06s5Z3ks5rhT4zgaJpZM4MMTjV>.
|
|
Can you please refer the issue here? |
|
Sorry I couldn't send link here. I just make a screenshot of it.
[image1.PNG]
On 1 Mar 2017, at 5:49 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
Can you please refer the issue here?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_adOlNaCFQ9p4DAYTstNRXmOrzoAks5rhT8MgaJpZM4MMTjV>.
|
|
The link is broken 😢 |
|
Can u send me the main.py file for testing after training done?
On 1 Mar 2017, at 8:19 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
The link is broken 😢
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_YwCwfXTgCIkpYQoN9IojsPliMYCks5rhWJTgaJpZM4MMTjV>.
|
|
#2
error in main.py · Issue #2 · wadhwasahil/Video-Classification-2-Stream-CNN · GitHub<#2>
github.com
Video-Classification-2-Stream-CNN - Video Classification using 2 stream CNN
Here is the link a person asked about main.py file.
…________________________________
From: yingchau lai <[email protected]>
Sent: Wednesday, March 1, 2017 8:47:07 PM
To: wadhwasahil/Video-Classification-2-Stream-CNN
Subject: Re: [wadhwasahil/Video-Classification-2-Stream-CNN] ValueError running spatial_stream_cnn.py (#9)
Can u send me the main.py file for testing after training done?
On 1 Mar 2017, at 8:19 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
The link is broken 😢
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_YwCwfXTgCIkpYQoN9IojsPliMYCks5rhWJTgaJpZM4MMTjV>.
|
|
I think she is talking about this file |
|
I see. Does your code here provide code for testing after both spatial and temporal training?
On 2 Mar 2017, at 1:27 AM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
I think she is talking about this file
spatial_stream_cnn.py. This is the main file for training the spatial CNN.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_W5Ziro6Kf81BLJ1yLbbC7K9znxJks5rhapqgaJpZM4MMTjV>.
|
|
No. But you can write it yourself, it'll be very easy. |
|
Ok. Thank you
…Sent from my iPhone
On 2 Mar 2017, at 2:43 PM, Sahil Wadhwa <[email protected]<mailto:[email protected]>> wrote:
No. But you can write it yourself, it'll be very easy.
Thanks.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#9 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AY1N_TL1dYJeGydxvG-a_eVPzcV3bvBIks5rhmUSgaJpZM4MMTjV>.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Using Theano backend.
Making model...
Compiling model...
Loading train dictionary...
Loading test dictionary...
Training model...
spatial_stream_cnn.py:23: FutureWarning: comparison to
Nonewill result in an elementwise object comparison in the future.if (X_train!=None and Y_train!=None):
(0,) <--------print(X_sample.shape)
Traceback (most recent call last):
File "spatial_stream_cnn.py", line 168, in
train(model, nb_epoch, spatial_train_data, spatial_test_data, chunk_size, nb_classes, img_rows, img_cols, batch_size)
File "spatial_stream_cnn.py", line 67, in train
datagen.fit(X_sample)
File "/usr/local/lib/python2.7/dist-packages/keras/preprocessing/image.py", line 578, in fit
'Got array with shape: ' + str(x.shape))
ValueError: Input to
.fit()should have rank 4. Got array with shape: (0,)The text was updated successfully, but these errors were encountered: