-
Notifications
You must be signed in to change notification settings - Fork 159
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[onert] Add unit tests for training Softmax op #13197
Conversation
_context->addTrainCase( | ||
uniformTCD<float>({{{1, 3}}, {{2, 1}}}, // inputs | ||
{{{0, 1, 0, 0, 0, 0, 0, 0}}, {{0, 0, 0, 0, 0, 1, 0, 0}}}, // expected | ||
{0.1092f} // loss |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The result of executing the same model with the same data on tensorflow:
Epoch 1/5
2/2 [==============================] - 0s 3ms/step - loss: 0.1094 - categorical_accuracy: 0.0000e+00
Epoch 2/5
2/2 [==============================] - 0s 1ms/step - loss: 0.1093 - categorical_accuracy: 0.5000
Epoch 3/5
2/2 [==============================] - 0s 1ms/step - loss: 0.1092 - categorical_accuracy: 0.5000
Epoch 4/5
2/2 [==============================] - 0s 1ms/step - loss: 0.1092 - categorical_accuracy: 0.5000
Epoch 5/5
2/2 [==============================] - 0s 1ms/step - loss: 0.1091 - categorical_accuracy: 0.5000
@nnfw-bot test onert-x64-release |
This commit adds some tests to validate training of Softmax op. ONE-DCO-1.0-Signed-off-by: ragmani <[email protected]>
831368c
to
5dc9e35
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM 👍
This commit adds some tests to validate training of Softmax op.
ONE-DCO-1.0-Signed-off-by: ragmani [email protected]