move dnn parameters to constant file

server/website/website/settings/constants.py

@@ -42,20 +42,37 @@ NUM_THREADS = 4
 #  the maximum iterations of gradient descent
 MAX_ITER = 500
 
+DEFAULT_LEARNING_RATE = 0.01
+
+# ---GRADIENT DESCENT FOR GPR---
 #  a small bias when using training data points as starting points.
 GPR_EPS = 0.001
 
 DEFAULT_RIDGE = 0.01
 
-DEFAULT_LEARNING_RATE = 0.01
-
 DEFAULT_EPSILON = 1e-6
 
 DEFAULT_SIGMA_MULTIPLIER = 3.0
 
 DEFAULT_MU_MULTIPLIER = 1.0
 
-# ---CONSTRAINTS CONSTANTS---
+# ---GRADIENT DESCENT FOR DNN---
+DNN_TRAIN_ITER = 500
+
+DNN_EXPLORE = False
+
+DNN_EXPLORE_ITER = 500
+
+# noise scale for paramater space exploration
+DNN_NOISE_SCALE_BEGIN = 0.1
+
+DNN_NOISE_SCALE_END = 0.0
+
+DNN_DEBUG = True
+
+DNN_DEBUG_INTERVAL = 100
+
+# ---DDPG CONSTRAINTS CONSTANTS---
 #  Batch size in DDPG model
 DDPG_BATCH_SIZE = 32
 

server/website/website/tasks/async_tasks.py

@@ -31,7 +31,10 @@ from website.settings import (DEFAULT_LENGTH_SCALE, DEFAULT_MAGNITUDE,
                               DEFAULT_EPSILON, MAX_ITER, GPR_EPS,
                               DEFAULT_SIGMA_MULTIPLIER, DEFAULT_MU_MULTIPLIER,
                               DDPG_BATCH_SIZE, ACTOR_LEARNING_RATE,
-                              CRITIC_LEARNING_RATE, GAMMA, TAU)
+                              CRITIC_LEARNING_RATE, GAMMA, TAU,
+                              DNN_TRAIN_ITER, DNN_EXPLORE, DNN_EXPLORE_ITER,
+                              DNN_NOISE_SCALE_BEGIN, DNN_NOISE_SCALE_END,
+                              DNN_DEBUG, DNN_DEBUG_INTERVAL)
 
 from website.settings import INIT_FLIP_PROB, FLIP_PROB_DECAY
 from website.settings import MODEL_DIR
@@ -553,12 +556,14 @@ def configuration_recommendation(recommendation_input):
         model_nn = NeuralNet(weights_file=full_path,
                              n_input=X_samples.shape[1],
                              batch_size=X_samples.shape[0],
-                             explore_iters=500,
-                             noise_scale_begin=0.1,
-                             noise_scale_end=0,
-                             debug=True)
-        model_nn.fit(X_scaled, y_scaled)
-        res = model_nn.recommend(X_samples, X_min, X_max, explore=True)
+                             explore_iters=DNN_EXPLORE_ITER,
+                             noise_scale_begin=DNN_NOISE_SCALE_BEGIN,
+                             noise_scale_end=DNN_NOISE_SCALE_END,
+                             debug=DNN_DEBUG,
+                             debug_interval=DNN_DEBUG_INTERVAL)
+        model_nn.fit(X_scaled, y_scaled, fit_epochs=DNN_TRAIN_ITER)
+        res = model_nn.recommend(X_samples, X_min, X_max,
+                                 explore=DNN_EXPLORE, recommend_epochs=MAX_ITER)
     elif algorithm == 'gpr':
         # default gpr model
         model = GPRGD(length_scale=DEFAULT_LENGTH_SCALE,

server/website/website/views.py

@@ -539,11 +539,13 @@ def handle_result_files(session, files):
         response = chain(aggregate_target_results.s(result.pk, 'dnn'),
                          map_workload.s(),
                          configuration_recommendation.s()).apply_async()
+
     taskmeta_ids = []
     current_task = response
     while current_task:
-        taskmeta_ids.append(current_task.id)
+        taskmeta_ids.insert(0, current_task.id)
         current_task = current_task.parent
+
     result.task_ids = ','.join(taskmeta_ids)
     result.save()
     return HttpResponse("Result stored successfully! Running tuner...(status={})  Result ID:{} "