GenomeFlowInterface Settings:
shuffle: True
pos_sampling_rate: 0.05
validation_chroms: ['chr22']
holdout_chroms: ['chr1', 'chr8', 'chr21']
optimizer: adam
learning rate: 0.0003
batch size: 256
epoch size: 2500000
max num of epochs: 100
early stopping metrics: auPRC
early stopping patience: 4

Epoch 1:
Metrics across all datasets:
Balanced Accuracy: 92.00%	auROC: 0.996	 auPRC: 0.564
Recall at 5% | 10% | 25% | 50% FDR: 0.4% | 0.4% | 43.0% | 64.2%	Num Positives: 1415	 Num Negatives: 996701


New best auPRC. Saving model.


Epoch 2:
Metrics across all datasets:
Balanced Accuracy: 90.80%	auROC: 0.996	 auPRC: 0.581
Recall at 5% | 10% | 25% | 50% FDR: 0.1% | 0.1% | 45.2% | 68.1%	Num Positives: 1415	 Num Negatives: 996701


New best auPRC. Saving model.


Epoch 3:
Metrics across all datasets:
Balanced Accuracy: 89.41%	auROC: 0.991	 auPRC: 0.558
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 41.9% | 64.0%	Num Positives: 1415	 Num Negatives: 996701



Epoch 4:
Metrics across all datasets:
Balanced Accuracy: 91.51%	auROC: 0.990	 auPRC: 0.586
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 45.0% | 67.4%	Num Positives: 1415	 Num Negatives: 996701


New best auPRC. Saving model.


Epoch 5:
Metrics across all datasets:
Balanced Accuracy: 85.05%	auROC: 0.975	 auPRC: 0.552
Recall at 5% | 10% | 25% | 50% FDR: 0.1% | 0.1% | 41.3% | 66.4%	Num Positives: 1415	 Num Negatives: 996701



Epoch 6:
Metrics across all datasets:
Balanced Accuracy: 86.41%	auROC: 0.970	 auPRC: 0.552
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 38.7% | 65.2%	Num Positives: 1415	 Num Negatives: 996701



Epoch 7:
Metrics across all datasets:
Balanced Accuracy: 85.18%	auROC: 0.973	 auPRC: 0.533
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 37.5% | 60.8%	Num Positives: 1415	 Num Negatives: 996701



Epoch 8:
Metrics across all datasets:
Balanced Accuracy: 85.92%	auROC: 0.985	 auPRC: 0.541
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 35.0% | 64.7%	Num Positives: 1415	 Num Negatives: 996701



Epoch 9:
Metrics across all datasets:
Balanced Accuracy: 87.30%	auROC: 0.988	 auPRC: 0.559
Recall at 5% | 10% | 25% | 50% FDR: 0.0% | 0.0% | 39.8% | 66.8%	Num Positives: 1415	 Num Negatives: 996701


Finished training after 9 epochs.
The best model's architecture and weights (from epoch 4) were saved to /users/marinovg/2018-01-25-histone-mods/tfdragonn-SequenceAndDnaseClassifier-K562-ATAC-cuts-CTCFL/model.arch.json and /users/marinovg/2018-01-25-histone-mods/tfdragonn-SequenceAndDnaseClassifier-K562-ATAC-cuts-CTCFL/model.weights.h5