ResourceExhaustedError：テンソルをシェイプ[256,100000000]に割り当てるときのOOM

Question

私は分類問題にTensorFlowの変数を使用しています。出力クラスの数は1e8です。

n_inputs = 5000 
n_classes = 1e8 
features = tf.placeholder(tf.float32, [None, n_inputs]) 
labels = tf.placeholder(tf.float32, [None, n_classes]) 

h_layer = 256 

weights = { 
'hidden_weights' : tf.Variable(tf.random_normal([n_inputs, h_layer])), 
'out_weights' : tf.Variable(tf.random_normal([h_layer, int(n_classes)])) 
} 

bias = { 
'hidden_bias' : tf.Variable(tf.random_normal([h_layer])), 
'out_bias' : tf.Variable(tf.random_normal([int(n_classes)])) 
} 

このコードを実行している間、私は（256,100000000）と 'out_weights' を割り当てるためのResourceExhaustedErrorを取得しています。とにかく私はこの問題を克服することができますか？

FYI：このコードをCPUで実行しています。短い答えはノーある

--------------------------------------------------------------------------- 
ResourceExhaustedError     Traceback (most recent call last) 
C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in _do_call(self, fn, *args) 
    1021  try: 
-> 1022  return fn(*args) 
    1023  except errors.OpError as e: 

C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in _run_fn(session, feed_dict, fetch_list, target_list, options, run_metadata) 
    1003         feed_dict, fetch_list, target_list, 
-> 1004         status, run_metadata) 
    1005 

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

C:\Anaconda\envs\tensorflow\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: 

ResourceExhaustedError: OOM when allocating tensor with shape[256,100000000] 
    [[Node: random_normal_5/RandomStandardNormal = RandomStandardNormal[T=DT_INT32, dtype=DT_FLOAT, seed=0, seed2=0, _device="/job:localhost/replica:0/task:0/cpu:0"](random_normal_5/shape)]] 

During handling of the above exception, another exception occurred: 

ResourceExhaustedError     Traceback (most recent call last) 
<ipython-input-26-d5491564869f> in <module>() 
    39 init = tf.global_variables_initializer() 
    40 with tf.Session() as sess: 
---> 41  sess.run(init) 
    42  total_batches = batches(batchSize, train_features, train_labels) 
    43 

C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in run(self, fetches, feed_dict, options, run_metadata) 
    765  try: 
    766  result = self._run(None, fetches, feed_dict, options_ptr, 
--> 767       run_metadata_ptr) 
    768  if run_metadata: 
    769   proto_data = tf_session.TF_GetBuffer(run_metadata_ptr) 

C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in _run(self, handle, fetches, feed_dict, options, run_metadata) 
    963  if final_fetches or final_targets: 
    964  results = self._do_run(handle, final_targets, final_fetches, 
--> 965        feed_dict_string, options, run_metadata) 
    966  else: 
    967  results = [] 

C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata) 
    1013  if handle is None: 
    1014  return self._do_call(_run_fn, self._session, feed_dict, fetch_list, 
-> 1015       target_list, options, run_metadata) 
    1016  else: 
    1017  return self._do_call(_prun_fn, self._session, handle, feed_dict, 

C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py in _do_call(self, fn, *args) 
    1033   except KeyError: 
    1034   pass 
-> 1035  raise type(e)(node_def, op, message) 
    1036 
    1037 def _extend_graph(self): 

ResourceExhaustedError: OOM when allocating tensor with shape[256,100000000] 
    [[Node: random_normal_5/RandomStandardNormal = RandomStandardNormal[T=DT_INT32, dtype=DT_FLOAT, seed=0, seed2=0, _device="/job:localhost/replica:0/task:0/cpu:0"](random_normal_5/shape)]] 

Caused by op 'random_normal_5/RandomStandardNormal', defined at: 
    File "C:\Anaconda\envs\tensorflow\lib\runpy.py", line 193, in _run_module_as_main 
    "__main__", mod_spec) 
    File "C:\Anaconda\envs\tensorflow\lib\runpy.py", line 85, in _run_code 
    exec(code, run_globals) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\__main__.py", line 3, in <module> 
    app.launch_new_instance() 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\traitlets\config\application.py", line 658, in launch_instance 
    app.start() 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\kernelapp.py", line 474, in start 
    ioloop.IOLoop.instance().start() 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\zmq\eventloop\ioloop.py", line 177, in start 
    super(ZMQIOLoop, self).start() 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tornado\ioloop.py", line 887, in start 
    handler_func(fd_obj, events) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tornado\stack_context.py", line 275, in null_wrapper 
    return fn(*args, **kwargs) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\zmq\eventloop\zmqstream.py", line 440, in _handle_events 
    self._handle_recv() 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\zmq\eventloop\zmqstream.py", line 472, in _handle_recv 
    self._run_callback(callback, msg) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\zmq\eventloop\zmqstream.py", line 414, in _run_callback 
    callback(*args, **kwargs) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tornado\stack_context.py", line 275, in null_wrapper 
    return fn(*args, **kwargs) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\kernelbase.py", line 276, in dispatcher 
    return self.dispatch_shell(stream, msg) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\kernelbase.py", line 228, in dispatch_shell 
    handler(stream, idents, msg) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\kernelbase.py", line 390, in execute_request 
    user_expressions, allow_stdin) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\ipkernel.py", line 196, in do_execute 
    res = shell.run_cell(code, store_history=store_history, silent=silent) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\ipykernel\zmqshell.py", line 501, in run_cell 
    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\IPython\core\interactiveshell.py", line 2717, in run_cell 
    interactivity=interactivity, compiler=compiler, result=result) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\IPython\core\interactiveshell.py", line 2821, in run_ast_nodes 
    if self.run_code(code, result): 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\IPython\core\interactiveshell.py", line 2881, in run_code 
    exec(code_obj, self.user_global_ns, self.user_ns) 
    File "<ipython-input-17-f183ffda50a1>", line 10, in <module> 
    'out_weights' : tf.Variable(tf.random_normal([h_layer, int(n_classes)])) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\ops\random_ops.py", line 77, in random_normal 
    seed2=seed2) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\ops\gen_random_ops.py", line 189, in _random_standard_normal 
    name=name) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 763, in apply_op 
    op_def=op_def) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\framework\ops.py", line 2327, in create_op 
    original_op=self._default_original_op, op_def=op_def) 
    File "C:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\framework\ops.py", line 1226, in __init__ 
    self._traceback = _extract_stack() 

ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[256,100000000] 
    [[Node: random_normal_5/RandomStandardNormal = RandomStandardNormal[T=DT_INT32, dtype=DT_FLOAT, seed=0, seed2=0, _device="/job:localhost/replica:0/task:0/cpu:0"](random_normal_5/shape)]] 

Answer 1

：

以下のスタックトレースを見つけてください。 256と1e8ニューロンの間に完全に接続されたレイヤーを作成したい場合は、メモリに256 * 1e8の数字が入りますが、何もできません。これはむしろ間違ったモデルと誤ったコードのように思えます。どうして1e8の出力クラスがありますか？それらの間に非常に強い相関関係があっても、最初にそれを訓練するためには、少なくとも1e10（10のビリオンのサンプル）ポイントが必要になるでしょう。手元にあるタスクにどのようにアプローチするかを再考する必要があります。本当に1e8の独立した出力が本当に必要だとは思えません。