Train a universal Indigo

Author: francisyyan

dagger/dagger.py

@@ -29,7 +29,7 @@ def __init__(self, cluster, server, worker_tasks):
         self.num_workers = len(worker_tasks)
         self.aggregated_states = []
         self.aggregated_actions = []
-        self.max_eps = 500
+        self.max_eps = 1000
         self.checkpoint_delta = 10
         self.checkpoint = self.checkpoint_delta
         self.learn_rate = 0.01
@@ -56,7 +56,7 @@ def __init__(self, cluster, server, worker_tasks):
         self.sync_op = tf.group(*[v1.assign(v2) for v1, v2 in zip(
             cpu_vars, gpu_vars)])
 
-        self.default_batch_size = 280
+        self.default_batch_size = 300
         self.default_init_state = self.global_network.zero_init_state(
                 self.default_batch_size)
 
@@ -248,14 +248,10 @@ def train(self):
 
         self.sess.run(self.global_network.add_one)
 
-        print 'DaggerLeader:before sync'
-        print 'DaggerLeader:global_network:cnt', self.sess.run(self.global_network.cnt)
-        print 'DaggerLeader:global_network_cpu:cnt', self.sess.run(self.global_network_cpu.cnt)
-
         # copy trained variables from GPU to CPU
         self.sess.run(self.sync_op)
 
-        print 'DaggerLeader:after sync'
+        print 'DaggerLeader:global_network:cnt', self.sess.run(self.global_network.cnt)
         print 'DaggerLeader:global_network_cpu:cnt', self.sess.run(self.global_network_cpu.cnt)
         sys.stdout.flush()
 
@@ -433,13 +429,9 @@ def run(self, debug=False):
                                  (self.task_idx, self.curr_ep))
 
             # Reset local parameters to global
-            print 'DaggerWorker:before sync'
-            print 'DaggerWorker:global_network_cpu:cnt', self.sess.run(self.global_network_cpu.cnt)
-            print 'DaggerWorker:local_network:cnt', self.sess.run(self.local_network.cnt)
-
             self.sess.run(self.sync_op)
 
-            print 'DaggerWorker:after sync'
+            print 'DaggerWorker:global_network_cpu:cnt', self.sess.run(self.global_network_cpu.cnt)
             print 'DaggerWorker:local_network:cnt', self.sess.run(self.local_network.cnt)
             sys.stdout.flush()
 

dagger/worker.py

@@ -47,41 +47,51 @@ def create_env(task_index):
     best_cwnds_file = path.join(project_root.DIR, 'dagger', 'best_cwnds.yml')
     best_cwnd_map = yaml.load(open(best_cwnds_file))
 
-    if task_index <= 15:
-        bandwidth = [10, 20, 50, 100]
+    if task_index == 0:
+        trace_path = path.join(project_root.DIR, 'env', '0.57mbps-poisson.trace')
+        mm_cmd = 'mm-delay 28 mm-loss uplink 0.0477 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=14' % (trace_path, trace_path)
+        best_cwnd = 5
+    elif task_index == 1:
+        trace_path = path.join(project_root.DIR, 'env', '2.64mbps-poisson.trace')
+        mm_cmd = 'mm-delay 88 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=130' % (trace_path, trace_path)
+        best_cwnd = 40
+    elif task_index == 2:
+        trace_path = path.join(project_root.DIR, 'env', '3.04mbps-poisson.trace')
+        mm_cmd = 'mm-delay 130 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=426' % (trace_path, trace_path)
+        best_cwnd = 70
+    elif task_index <= 22:
+        bandwidth = [5, 10, 20, 50, 100]
         delay = [10, 20, 40, 80]
 
         cartesian = [(b, d) for b in bandwidth for d in delay]
-        bandwidth, delay = cartesian[task_index]
+        bandwidth, delay = cartesian[task_index - 3]
 
         uplink_trace, downlink_trace = prepare_traces(bandwidth)
         mm_cmd = 'mm-delay %d mm-link %s %s' % (delay, uplink_trace, downlink_trace)
         best_cwnd = best_cwnd_map[bandwidth][delay]
-
-    elif task_index == 16:
+    elif task_index == 23:
         trace_path = path.join(project_root.DIR, 'env', '100.42mbps.trace')
         mm_cmd = 'mm-delay 27 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=173' % (trace_path, trace_path)
         best_cwnd = 500
-    elif task_index == 17:
+    elif task_index == 24:
         trace_path = path.join(project_root.DIR, 'env', '77.72mbps.trace')
         mm_cmd = 'mm-delay 51 mm-loss uplink 0.0006 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=94' % (trace_path, trace_path)
         best_cwnd = 690
-    elif task_index == 18:
+    elif task_index == 25:
         trace_path = path.join(project_root.DIR, 'env', '114.68mbps.trace')
         mm_cmd = 'mm-delay 45 mm-link %s %s --uplink-queue=droptail --uplink-queue-args=packets=450' % (trace_path, trace_path)
         best_cwnd = 870
-    elif task_index <= 22:
+    elif task_index <= 29:
         bandwidth = [200]
         delay = [10, 20, 40, 80]
 
         cartesian = [(b, d) for b in bandwidth for d in delay]
-        bandwidth, delay = cartesian[task_index - 19]
+        bandwidth, delay = cartesian[task_index - 26]
 
         uplink_trace, downlink_trace = prepare_traces(bandwidth)
         mm_cmd = 'mm-delay %d mm-link %s %s' % (delay, uplink_trace, downlink_trace)
         best_cwnd = best_cwnd_map[bandwidth][delay]
 
-
     env = Environment(mm_cmd)
     env.best_cwnd = best_cwnd
 

env/sender.py

@@ -26,7 +26,7 @@ class Sender(object):
     # RL exposed class/static variables
     max_steps = 1000
     state_dim = 4
-    action_mapping = format_actions(["/2.0", "-10.0", "+10.0", "*2.0"])
+    action_mapping = format_actions(["/2.0", "-10.0", "+0.0", "+10.0", "*2.0"])
     action_cnt = len(action_mapping)
 
     def __init__(self, port=0, train=False):
@@ -140,7 +140,7 @@ def take_action(self, action_idx):
         op, val = self.action_mapping[action_idx]
 
         self.cwnd = apply_op(op, self.cwnd, val)
-        self.cwnd = max(5.0, self.cwnd)
+        self.cwnd = max(2.0, self.cwnd)
 
     def window_is_open(self):
         return self.seq_num - self.next_ack < self.cwnd

helpers/assistant.py

@@ -21,7 +21,7 @@ def run_cmd(args, host, procs):
 
     elif cmd == 'git_pull':
         cmd_in_ssh = ('cd %s && git fetch --all && '
-                      'git checkout indigo-broad && '
+                      'git checkout indigo-universal && '
                       'git reset --hard @~1 && git pull' % args.rlcc_dir)
 
     elif cmd == 'rm_history':