In [1]:

    

import numpy as np
import tensorflow as tf
import tensorflow.contrib.slim as slim


    

In [2]:

    

#List out our bandits. Currently bandit 4 (index#3) is set to most often provide a positive reward.
bandits = [0.2,0,-0.2,-5]
num_bandits = len(bandits)
def pullBandit(bandit):
    #Get a random number.
    result = np.random.randn(1)
    if result > bandit:
        #return a positive reward.
        return 1
    else:
        #return a negative reward.
        return -1


    

In [3]:

    

num_tests = 20
print([ pullBandit(bandits[0]) for _ in range(num_tests) ])
print([ pullBandit(bandits[1]) for _ in range(num_tests) ])
print([ pullBandit(bandits[2]) for _ in range(num_tests) ])
print([ pullBandit(bandits[3]) for _ in range(num_tests) ])


    

    




[1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1]
[1, -1, -1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, -1, -1, 1, 1, -1, 1, -1]
[1, 1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1]
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]

In [4]:

    

tf.reset_default_graph()
value = tf.constant([1, 2, 3, 4], dtype=tf.float32)
indices1 = tf.constant([0], dtype=tf.int32)
indices2 = tf.constant([2], dtype=tf.int32)
output1 = tf.slice(value, indices1, [2])
output2 = tf.slice(value, indices2, [1])
with tf.Session() as sess:
    o1, o2 = sess.run([output1, output2])
    print(o1)
    print(o2)


    

    




[ 1.  2.]
[ 3.]

In [5]:

    

tf.reset_default_graph()

#These two lines established the feed-forward part of the network. This does the actual choosing.
weights = tf.Variable(tf.ones([num_bandits]))
chosen_action = tf.argmax(weights,0)

#The next six lines establish the training proceedure. We feed the reward and chosen action into the network
#to compute the loss, and use it to update the network.
reward_holder = tf.placeholder(shape=[1],dtype=tf.float32)
action_holder = tf.placeholder(shape=[1],dtype=tf.int32)
responsible_weight = tf.slice(weights,action_holder,[1])
loss = -(tf.log(responsible_weight)*reward_holder)
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)
update = optimizer.minimize(loss)


    

In [6]:

    

total_episodes = 1000 #Set total number of episodes to train agent on.
total_reward = np.zeros(num_bandits) #Set scoreboard for bandits to 0.
e = 0.1 #Set the chance of taking a random action.

init = tf.global_variables_initializer()

# Launch the tensorflow graph
with tf.Session() as sess:
    sess.run(init)
    i = 0
    while i < total_episodes:
        
        #Choose either a random action or one from our network.
        if np.random.rand(1) < e:
            action = np.random.randint(num_bandits)
        else:
            action = sess.run(chosen_action)
        
        reward = pullBandit(bandits[action]) #Get our reward from picking one of the bandits.
        
        #Update the network.
        _,resp,ww = sess.run([update,responsible_weight,weights], feed_dict={reward_holder:[reward],action_holder:[action]})
        
        #Update our running tally of scores.
        total_reward[action] += reward
        if i % 50 == 0:
            print("Running reward for the " + str(num_bandits) + " bandits: " + str(total_reward))
        i+=1
print("The agent thinks bandit " + str(np.argmax(ww)+1) + " is the most promising....")
if np.argmax(ww) == np.argmax(-np.array(bandits)):
    print("...and it was right!")
else:
    print("...and it was wrong!")


    

    




Running reward for the 4 bandits: [ 1.  0.  0.  0.]
Running reward for the 4 bandits: [ -2.  -2.   1.  36.]
Running reward for the 4 bandits: [ -3.  -1.   1.  80.]
Running reward for the 4 bandits: [  -3.   -1.    1.  128.]
Running reward for the 4 bandits: [  -4.   -1.    2.  176.]
Running reward for the 4 bandits: [  -3.   -2.    1.  223.]
Running reward for the 4 bandits: [  -2.   -3.    0.  268.]
Running reward for the 4 bandits: [  -4.   -4.   -2.  313.]
Running reward for the 4 bandits: [  -5.   -5.   -2.  361.]
Running reward for the 4 bandits: [  -4.   -5.    0.  406.]
Running reward for the 4 bandits: [  -3.   -5.    1.  454.]
Running reward for the 4 bandits: [  -3.   -6.    0.  500.]
Running reward for the 4 bandits: [  -3.   -8.   -1.  547.]
Running reward for the 4 bandits: [  -2.   -8.   -1.  596.]
Running reward for the 4 bandits: [  -2.   -7.    1.  641.]
Running reward for the 4 bandits: [  -3.   -7.    1.  690.]
Running reward for the 4 bandits: [  -4.  -10.    1.  736.]
Running reward for the 4 bandits: [  -4.   -9.    4.  782.]
Running reward for the 4 bandits: [  -4.   -8.    5.  830.]
Running reward for the 4 bandits: [  -6.   -8.    5.  878.]
The agent thinks bandit 4 is the most promising....
...and it was right!

In [7]:

    

class contextual_bandit():
    def __init__(self):
        self.state = 0
        #List out our bandits. Currently arms 4, 2, and 1 (respectively) are the most optimal.
        self.bandits = np.array([[0.2,0,-0.0,-5],[0.1,-5,1,0.25],[-5,5,5,5]])
        self.num_bandits = self.bandits.shape[0]
        self.num_actions = self.bandits.shape[1]
        
    def getBandit(self):
        self.state = np.random.randint(0,len(self.bandits)) #Returns a random state for each episode.
        return self.state
        
    def pullArm(self,action):
        #Get a random number.
        bandit = self.bandits[self.state,action]
        result = np.random.randn(1)
        if result > bandit:
            #return a positive reward.
            return 1
        else:
            #return a negative reward.
            return -1


    

In [8]:

    

class agent():
    def __init__(self, lr, s_size,a_size):
        #These lines established the feed-forward part of the network. The agent takes a state and produces an action.
        self.state_in= tf.placeholder(shape=[1],dtype=tf.int32)
        state_in_OH = slim.one_hot_encoding(self.state_in,s_size)
        output = slim.fully_connected(state_in_OH,a_size,\
            biases_initializer=None,activation_fn=tf.nn.sigmoid,weights_initializer=tf.ones_initializer())
        self.output = tf.reshape(output,[-1])
        self.chosen_action = tf.argmax(self.output,0)

        #The next six lines establish the training proceedure. We feed the reward and chosen action into the network
        #to compute the loss, and use it to update the network.
        self.reward_holder = tf.placeholder(shape=[1],dtype=tf.float32)
        self.action_holder = tf.placeholder(shape=[1],dtype=tf.int32)
        self.responsible_weight = tf.slice(self.output,self.action_holder,[1])
        self.loss = -(tf.log(self.responsible_weight)*self.reward_holder)
        optimizer = tf.train.GradientDescentOptimizer(learning_rate=lr)
        self.update = optimizer.minimize(self.loss)


    

In [9]:

    

tf.reset_default_graph() #Clear the Tensorflow graph.

cBandit = contextual_bandit() #Load the bandits.
myAgent = agent(lr=0.001,s_size=cBandit.num_bandits,a_size=cBandit.num_actions) #Load the agent.
weights = tf.trainable_variables()[0] #The weights we will evaluate to look into the network.

total_episodes = 10000 #Set total number of episodes to train agent on.
total_reward = np.zeros([cBandit.num_bandits,cBandit.num_actions]) #Set scoreboard for bandits to 0.
e = 0.1 #Set the chance of taking a random action.

init = tf.global_variables_initializer()

# Launch the tensorflow graph
with tf.Session() as sess:
    sess.run(init)
    i = 0
    while i < total_episodes:
        s = cBandit.getBandit() #Get a state from the environment.
        
        #Choose either a random action or one from our network.
        if np.random.rand(1) < e:
            action = np.random.randint(cBandit.num_actions)
        else:
            action = sess.run(myAgent.chosen_action,feed_dict={myAgent.state_in:[s]})
        
        reward = cBandit.pullArm(action) #Get our reward for taking an action given a bandit.
        
        #Update the network.
        feed_dict={myAgent.reward_holder:[reward],myAgent.action_holder:[action],myAgent.state_in:[s]}
        _,ww = sess.run([myAgent.update,weights], feed_dict=feed_dict)
        
        #Update our running tally of scores.
        total_reward[s,action] += reward
        if i % 500 == 0:
            print("Mean reward for each of the " + str(cBandit.num_bandits) + " bandits: " + str(np.mean(total_reward,axis=1)))
        i+=1
for a in range(cBandit.num_bandits):
    print("The agent thinks action " + str(np.argmax(ww[a])+1) + " for bandit " + str(a+1) + " is the most promising....")
    if np.argmax(ww[a]) == np.argmin(cBandit.bandits[a]):
        print("...and it was right!")
    else:
        print("...and it was wrong!")


    

    




---------------------------------------------------------------------------
InternalError                             Traceback (most recent call last)
C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in _do_call(self, fn, *args)
   1138     try:
-> 1139       return fn(*args)
   1140     except errors.OpError as e:

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in _run_fn(session, feed_dict, fetch_list, target_list, options, run_metadata)
   1120                                  feed_dict, fetch_list, target_list,
-> 1121                                  status, run_metadata)
   1122 

C:\Anaconda3\envs\xiang\lib\contextlib.py in __exit__(self, type, value, traceback)
     65             try:
---> 66                 next(self.gen)
     67             except StopIteration:

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\framework\errors_impl.py in raise_exception_on_not_ok_status()
    465           compat.as_text(pywrap_tensorflow.TF_Message(status)),
--> 466           pywrap_tensorflow.TF_GetCode(status))
    467   finally:

InternalError: Blas GEMM launch failed : a.shape=(1, 3), b.shape=(3, 4), m=1, n=4, k=3
	 [[Node: fully_connected/MatMul = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/gpu:0"](OneHotEncoding/one_hot/_5, fully_connected/weights/read)]]
	 [[Node: ArgMax/_7 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_24_ArgMax", tensor_type=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

During handling of the above exception, another exception occurred:

InternalError                             Traceback (most recent call last)
<ipython-input-9-4901022e0d57> in <module>()
     22             action = np.random.randint(cBandit.num_actions)
     23         else:
---> 24             action = sess.run(myAgent.chosen_action,feed_dict={myAgent.state_in:[s]})
     25 
     26         reward = cBandit.pullArm(action) #Get our reward for taking an action given a bandit.

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in run(self, fetches, feed_dict, options, run_metadata)
    787     try:
    788       result = self._run(None, fetches, feed_dict, options_ptr,
--> 789                          run_metadata_ptr)
    790       if run_metadata:
    791         proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)
    995     if final_fetches or final_targets:
    996       results = self._do_run(handle, final_targets, final_fetches,
--> 997                              feed_dict_string, options, run_metadata)
    998     else:
    999       results = []

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)
   1130     if handle is None:
   1131       return self._do_call(_run_fn, self._session, feed_dict, fetch_list,
-> 1132                            target_list, options, run_metadata)
   1133     else:
   1134       return self._do_call(_prun_fn, self._session, handle, feed_dict,

C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\client\session.py in _do_call(self, fn, *args)
   1150         except KeyError:
   1151           pass
-> 1152       raise type(e)(node_def, op, message)
   1153 
   1154   def _extend_graph(self):

InternalError: Blas GEMM launch failed : a.shape=(1, 3), b.shape=(3, 4), m=1, n=4, k=3
	 [[Node: fully_connected/MatMul = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/gpu:0"](OneHotEncoding/one_hot/_5, fully_connected/weights/read)]]
	 [[Node: ArgMax/_7 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_24_ArgMax", tensor_type=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

Caused by op 'fully_connected/MatMul', defined at:
  File "C:\Anaconda3\envs\xiang\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "C:\Anaconda3\envs\xiang\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel_launcher.py", line 16, in <module>
    app.launch_new_instance()
  File "C:\Anaconda3\envs\xiang\lib\site-packages\traitlets\config\application.py", line 658, in launch_instance
    app.start()
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\kernelapp.py", line 477, in start
    ioloop.IOLoop.instance().start()
  File "C:\Anaconda3\envs\xiang\lib\site-packages\zmq\eventloop\ioloop.py", line 177, in start
    super(ZMQIOLoop, self).start()
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tornado\ioloop.py", line 888, in start
    handler_func(fd_obj, events)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tornado\stack_context.py", line 277, in null_wrapper
    return fn(*args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\zmq\eventloop\zmqstream.py", line 440, in _handle_events
    self._handle_recv()
  File "C:\Anaconda3\envs\xiang\lib\site-packages\zmq\eventloop\zmqstream.py", line 472, in _handle_recv
    self._run_callback(callback, msg)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\zmq\eventloop\zmqstream.py", line 414, in _run_callback
    callback(*args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tornado\stack_context.py", line 277, in null_wrapper
    return fn(*args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\kernelbase.py", line 283, in dispatcher
    return self.dispatch_shell(stream, msg)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\kernelbase.py", line 235, in dispatch_shell
    handler(stream, idents, msg)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\kernelbase.py", line 399, in execute_request
    user_expressions, allow_stdin)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\ipkernel.py", line 196, in do_execute
    res = shell.run_cell(code, store_history=store_history, silent=silent)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\ipykernel\zmqshell.py", line 533, in run_cell
    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\IPython\core\interactiveshell.py", line 2698, in run_cell
    interactivity=interactivity, compiler=compiler, result=result)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\IPython\core\interactiveshell.py", line 2802, in run_ast_nodes
    if self.run_code(code, result):
  File "C:\Anaconda3\envs\xiang\lib\site-packages\IPython\core\interactiveshell.py", line 2862, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-9-4901022e0d57>", line 4, in <module>
    myAgent = agent(lr=0.001,s_size=cBandit.num_bandits,a_size=cBandit.num_actions) #Load the agent.
  File "<ipython-input-8-6ea9a1d89ad6>", line 6, in __init__
    output = slim.fully_connected(state_in_OH,a_size,            biases_initializer=None,activation_fn=tf.nn.sigmoid,weights_initializer=tf.ones_initializer())
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\contrib\framework\python\ops\arg_scope.py", line 181, in func_with_args
    return func(*args, **current_args)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\contrib\layers\python\layers\layers.py", line 1469, in fully_connected
    outputs = layer.apply(inputs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\layers\base.py", line 492, in apply
    return self.__call__(inputs, *args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\layers\base.py", line 441, in __call__
    outputs = self.call(inputs, *args, **kwargs)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\layers\core.py", line 141, in call
    outputs = standard_ops.matmul(inputs, self.kernel)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\ops\math_ops.py", line 1816, in matmul
    a, b, transpose_a=transpose_a, transpose_b=transpose_b, name=name)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\ops\gen_math_ops.py", line 1217, in _mat_mul
    transpose_b=transpose_b, name=name)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 767, in apply_op
    op_def=op_def)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\framework\ops.py", line 2506, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "C:\Anaconda3\envs\xiang\lib\site-packages\tensorflow\python\framework\ops.py", line 1269, in __init__
    self._traceback = _extract_stack()

InternalError (see above for traceback): Blas GEMM launch failed : a.shape=(1, 3), b.shape=(3, 4), m=1, n=4, k=3
	 [[Node: fully_connected/MatMul = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/job:localhost/replica:0/task:0/gpu:0"](OneHotEncoding/one_hot/_5, fully_connected/weights/read)]]
	 [[Node: ArgMax/_7 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_24_ArgMax", tensor_type=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

In [ ]: