Python model.predict() 和 model.fit() 有什么作用？
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/37973005/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
What do model.predict() and model.fit() do?
提问by Soham
I'm going through this reinforcement learning tutorialand It's been really great so far but could someone please explain what
我正在学习这个强化学习教程，到目前为止它真的很棒，但有人可以解释一下吗
newQ = model.predict(new_state.reshape(1,64), batch_size=1)
and
和
model.fit(X_train, y_train, batch_size=batchSize, nb_epoch=1, verbose=1)
mean?
意思？
As in what do the arguments bach_size, nb_epochand verbosedo?
I know neural networks so explaining in terms of that would be helpful.
至于在什么做的论点bach_size，nb_epoch和verbose做什么？我知道神经网络，所以用它来解释会很有帮助。
You could also send me a link where the documentation of these functions can be found.
您还可以向我发送一个链接，其中可以找到这些函数的文档。
采纳答案by nemo
First of all it surprises me that you could not find the documentationbut I guess you just had bad luck while searching.
首先让我惊讶的是你找不到文档，但我猜你在搜索时运气不好。
The documentation states for model.fit:
文档说明model.fit：
fit(self, x, y, batch_size=32, nb_epoch=10, verbose=1, callbacks=[], validation_split=0.0, validation_data=None, shuffle=True, class_weight=None, sample_weight=None)
batch_size: integer. Number of samples per gradient update.nb_epoch: integer, the number of times to iterate over the training data arrays.verbose: 0, 1, or 2. Verbosity mode. 0 = silent, 1 = verbose, 2 = one log line per epoch.
fit(self, x, y, batch_size=32, nb_epoch=10, verbose=1, callbacks=[], validation_split=0.0, validation_data=None, shuffle=True, class_weight=None, sample_weight=None)
batch_size： 整数。每次梯度更新的样本数。nb_epoch：整数，迭代训练数据数组的次数。verbose: 0、1 或 2。详细模式。0 = 无声，1 = 冗长，2 = 每个 epoch 一个日志行。
The batch_sizeparameter in case of model.predictis just the number of samples used for each prediction step. So calling model.predictone time consumes batch_sizenumber of data samples. This helps for devices that can process large matrices quickly (such as GPUs).
batch_size情况下的参数model.predict只是用于每个预测步骤的样本数。所以调用model.predict一次会消耗batch_size大量的数据样本。这有助于可以快速处理大型矩阵的设备（例如 GPU）。
