Add random reshuffle

Author: lio-wong

graphs

@@ -7,5 +7,5 @@ checkpoint19='experimentOutputs/list_aic=1.0_arity=3_BO=True_CO=True_ET=720_HR=0
 
 srun --job-name=graphsHeldout$1 --output=data/batch_12_2_2018/graphs --ntasks=1 --mem-per-cpu=5000 --cpus-per-task 1 --time=5:00 --qos=tenenbaum \
   singularity exec -B /om2  sklearn-container.img \
-  python graphs.py --checkpoints $checkpoint19  --showEpochs --export data/batch_12_2_2018/test.png  \
+  python graphs.py --checkpoints $checkpoint19  --showEpochs --showTraining --export data/batch_12_2_2018/test.png  \
   &

taskBatcher.py

@@ -38,7 +38,7 @@ def getTaskBatch(self, ec_result, tasks, taskBatchSize, currIteration):
 
 class RandomShuffleTaskBatcher:
 	"""Randomly shuffles the task batch first, and then iterates through task batches of the specified size like DefaultTaskBatcher.
-	   Uses a fixed shuffling across iterations - intended as benchmark comparison to test the task ordering."""
+	   Reshuffles across iterations - intended as benchmark comparison to test the task ordering."""
 	def __init__(self):
 		pass
 
@@ -49,13 +49,19 @@ def getTaskBatch(self, ec_result, tasks, taskBatchSize, currIteration):
 			eprint("Task batch size is greater than total number of tasks, aborting.")
 			assert False
 
-		# Shuffles tasks with a set seed across iterations.
+		# Reshuffles tasks in a fixed way across epochs for reproducibility.
+		baseSeed = 0
+		currEpoch = int(int(currIteration * taskBatchSize) / int(len(tasks)))
+
 		shuffledTasks = tasks.copy() # Since shuffle works in place.
-		random.Random(0).shuffle(shuffledTasks)
+		random.Random(baseSeed + currEpoch).shuffle(shuffledTasks)
+
+		shuffledTasksWrap = tasks.copy() # Since shuffle works in place.
+		random.Random(baseSeed + currEpoch + 1).shuffle(shuffledTasksWrap)
 
 		start = (taskBatchSize * currIteration) % len(shuffledTasks)
 		end = start + taskBatchSize
-		taskBatch = (shuffledTasks + shuffledTasks)[start:end] # Handle wraparound.
+		taskBatch = (shuffledTasks + shuffledTasksWrap)[start:end] # Wraparound nicely.
 
 		return taskBatch