我正在尝试按原样运行ECM代码。
作者声称可以正确运行代码。但是当我尝试运行它时，会出现此错误。
标志变量如下：
tf.app.flags.DEFINE_float("learning_rate", 0.5, "Learning rate.")
tf.app.flags.DEFINE_float("learning_rate_decay_factor", 0.99,"Learning rate decays by this much.")
tf.app.flags.DEFINE_float("max_gradient_norm", 5.0,gradients to this norm.")
tf.app.flags.DEFINE_integer("batch_size", 128,size to use during training.")
tf.app.flags.DEFINE_integer("size", 256, "Size of each model layer.")
tf.app.flags.DEFINE_integer("load_model", 0, "which model to load.")
tf.app.flags.DEFINE_integer("beam_size", 20, "Size of beam.")
tf.app.flags.DEFINE_integer("embedding_size", 100, "Size of word embedding.")
tf.app.flags.DEFINE_integer("emotion_size", 100, "Size of emotion embedding.")
tf.app.flags.DEFINE_integer("imemory_size", 256, "Size of imemory.")
tf.app.flags.DEFINE_integer("category", 6, "category of emotions.")
tf.app.flags.DEFINE_integer("num_layers", 2, "Number of layers in the model.")
tf.app.flags.DEFINE_integer("response_vocab_size", 40000, "response vocabulary size.")
tf.app.flags.DEFINE_integer("post_vocab_size", 40000, "post vocabulary size.")
tf.app.flags.DEFINE_string("data_dir", "/content", "Data directory")
tf.app.flags.DEFINE_string("train_dir", "train", "Training directory.")
tf.app.flags.DEFINE_string("pretrain_dir", "pretrain", "Pretraining directory.")
tf.app.flags.DEFINE_integer("pretrain", -1, "pretrain model number")
tf.app.flags.DEFINE_integer("max_train_data_size", 0,"Limit on the size of training data (0: no limit).")
tf.app.flags.DEFINE_integer("steps_per_checkpoint", 1000, "How many training steps to do per checkpoint.")
此错误是否可能是由于我正在对其进行训练的数据集的大小所致？还是错误和数据集的大小之间没有联系？
PS：张量流版本为0.12