Wormbase Results

/usr/bin/python2.7 /home/keiron/.projects/lethality_prediction/python_files/wormbase_predict.py
Using Theano backend.
Loading the dataset...
	- Loading the raw dataset from ../data/Worm_Dropshilla_Lethality.arff
	- Vectorising the raw dataset into a format suitable for Neural Networks
	- Randomly shuffling the dataset, to ensure proper results
	- Splitting the dataset into separate training and testing sets
Now for the Deep Learning bit...
	- Modelling the Neural Network
	- Training the model

Train on 300 samples, validate on 75 samples
Epoch 1/8
300/300 [==============================] - 4s - loss: 0.5245 - acc: 0.6633 - val_loss: 0.3111 - val_acc: 0.8667
Epoch 2/8
300/300 [==============================] - 4s - loss: 0.3041 - acc: 0.9000 - val_loss: 0.2125 - val_acc: 0.8933
Epoch 3/8
300/300 [==============================] - 4s - loss: 0.2455 - acc: 0.9033 - val_loss: 0.1930 - val_acc: 0.8933
Epoch 4/8
300/300 [==============================] - 4s - loss: 0.2162 - acc: 0.9167 - val_loss: 0.1547 - val_acc: 0.9333
Epoch 5/8
300/300 [==============================] - 4s - loss: 0.2028 - acc: 0.9300 - val_loss: 0.1586 - val_acc: 0.9200
Epoch 6/8
300/300 [==============================] - 4s - loss: 0.1921 - acc: 0.9267 - val_loss: 0.1661 - val_acc: 0.9200
Epoch 7/8
300/300 [==============================] - 4s - loss: 0.1895 - acc: 0.9267 - val_loss: 0.1318 - val_acc: 0.9200
Epoch 8/8
300/300 [==============================] - 4s - loss: 0.1818 - acc: 0.9300 - val_loss: 0.1452 - val_acc: 0.9200
	- Testing the model
	- Returning the classification report of the test
             precision    recall  f1-score   support

          0       0.76      0.94      0.84        17
          1       0.98      0.91      0.95        58

avg / total       0.93      0.92      0.92        75

	- Returning the confusion matrix of the test
[[16  1]
 [ 5 53]]

Process finished with exit code 0