Problem with neural network

[maksimliubimov]

Hello, can you help me? I tried to learn neural network but I have this result: Using TensorFlow backend. /usr/local/lib/python2.7/dist-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20. "This module will be removed in 0.20.", DeprecationWarning) Loading training data... ['train_labels', 'train'] (424, 38400) (424, 3) Shape of feature array: (424, 38400) Shape of label array: (424, 3) Total time taken to load image data: 0.713379859924 seconds Training... Train on 339 samples, validate on 85 samples Epoch 1/20 339/339 [==============================] - 0s - loss: 1.1011 - acc: 0.3186 - val_loss: 1.0821 - val_acc: 0.4588 Epoch 2/20 339/339 [==============================] - 0s - loss: 1.0958 - acc: 0.3333 - val_loss: 1.0769 - val_acc: 0.5412 Epoch 3/20 339/339 [==============================] - 0s - loss: 1.0874 - acc: 0.3982 - val_loss: 1.0711 - val_acc: 0.4941 Epoch 4/20 339/339 [==============================] - 0s - loss: 1.0786 - acc: 0.4602 - val_loss: 1.0648 - val_acc: 0.5176 Epoch 5/20 339/339 [==============================] - 0s - loss: 1.0733 - acc: 0.4926 - val_loss: 1.0591 - val_acc: 0.4941 Epoch 6/20 339/339 [==============================] - 0s - loss: 1.0589 - acc: 0.4985 - val_loss: 1.0538 - val_acc: 0.4941 Epoch 7/20 339/339 [==============================] - 0s - loss: 1.0525 - acc: 0.4985 - val_loss: 1.0494 - val_acc: 0.4941 Epoch 8/20 339/339 [==============================] - 0s - loss: 1.0420 - acc: 0.4985 - val_loss: 1.0461 - val_acc: 0.4941 Epoch 9/20 339/339 [==============================] - 0s - loss: 1.0414 - acc: 0.5015 - val_loss: 1.0441 - val_acc: 0.4941 Epoch 10/20 339/339 [==============================] - 0s - loss: 1.0314 - acc: 0.5015 - val_loss: 1.0431 - val_acc: 0.4941 Epoch 11/20 339/339 [==============================] - 0s - loss: 1.0239 - acc: 0.5044 - val_loss: 1.0429 - val_acc: 0.4941 Epoch 12/20 339/339 [==============================] - 0s - loss: 1.0169 - acc: 0.5015 - val_loss: 1.0433 - val_acc: 0.4941 Epoch 13/20 339/339 [==============================] - 0s - loss: 1.0154 - acc: 0.5015 - val_loss: 1.0440 - val_acc: 0.4941 Epoch 14/20 339/339 [==============================] - 0s - loss: 1.0181 - acc: 0.5015 - val_loss: 1.0447 - val_acc: 0.4941 Epoch 15/20 339/339 [==============================] - 0s - loss: 1.0019 - acc: 0.5015 - val_loss: 1.0451 - val_acc: 0.4941 Epoch 16/20 339/339 [==============================] - 0s - loss: 1.0068 - acc: 0.5015 - val_loss: 1.0448 - val_acc: 0.4941 Epoch 17/20 339/339 [==============================] - 0s - loss: 1.0000 - acc: 0.5015 - val_loss: 1.0446 - val_acc: 0.4941 Epoch 18/20 339/339 [==============================] - 0s - loss: 0.9944 - acc: 0.5015 - val_loss: 1.0434 - val_acc: 0.4941 Epoch 19/20 339/339 [==============================] - 0s - loss: 0.9801 - acc: 0.5015 - val_loss: 1.0419 - val_acc: 0.4941 Epoch 20/20 339/339 [==============================] - 0s - loss: 0.9913 - acc: 0.5015 - val_loss: 1.0399 - val_acc: 0.4941

Total time taken to train model: 5.78856301308 secucationds

Evaluation of model on test holdout set: 85/85 [==============================] - 0s

Loss score: 1.03991830349 Accuracy score: 0.494117647409 Also training was instantaneous. It would be really good if you give me a question. Thank you

[AiBuddy]

@LyubimovMaxim i wonder if u ever got ur answer

[muskansingla12]

@LyubimovMaxim can you provide me the dataset @muskansingla1999@gmail.com?