训练后如何预测？

Question

我的csv数据如下：https://storage.googleapis.com/download.tensorflow.org/data/abalone_train.csv，我想根据其他专栏预测“年龄”。培训守则如下：

import pandas as pd
import numpy as np
# Make numpy values easier to read.
np.set_printoptions(precision=3, suppress=True)
import tensorflow as tf
from tensorflow.keras import layers
from tensorflow.keras.layers.experimental import preprocessing
abalone_train = pd.read_csv("https://storage.googleapis.com/download.tensorflow.org/data/abalone_train.csv", header=None, 
names=["Length", "Diameter", "Height", "Whole weight", "Shucked weight","Viscera weight", "Shell weight", "Age"])
abalone_train.head()
abalone_features = abalone_train.copy()
abalone_labels = abalone_features.pop('Age')
abalone_features = np.array(abalone_features)
abalone_features
abalone_model = tf.keras.Sequential([
layers.Dense(64),
layers.Dense(1)
])

abalone_model.compile(loss = tf.losses.MeanSquaredError(),optimizer = tf.optimizers.Adam())
abalone_model.fit(abalone_features, abalone_labels, epochs=10)

输出：

Epoch 1/10 104/104 ============================== -s 1ms/step损失: 63.1474 Epoch 5/10 104/104 ============================== -s 924 0s /step损失: 11.8933 Epoch 3/10 104/104 ============================== -s 920 0s /step损失: 8.4037 Epoch 4/10 104/104 ============================== -s 885 0s /step: 7.9656 Epoch 5/10 104/104 ============================== -90 0 0s /step: 7.5481 Epoch 6/10 104/104 ============================== -s 908 0s /step: 7.2339 Epoch 7/10 104/104 ============================== -s 926 0s /阶跃损失: 6.9871时代8/10 104/104 ============================== -s 919 0s /步进损失: 6.7886 Epoch 9/10 104/104 ============================== -s 956 0s /步进损失: 6.6482 Epoch 10/10 104/104 ============================== -s 953 0s /step损失: 6.5404

现在，我想上传另一个csv文件，其中有空白的“年龄”栏，并看到预测，但我被卡住了。我得到了一些教训，但在所有的课程，直到“时代”阶段。在“时代”阶段之后，我如何导入我的“空白年龄”csv文件并查看“年龄预测”？

Answer 1

根据文档(https://www.tensorflow.org/api_docs/python/tf/keras/Sequential#predict)，Sequential对象有一个predict方法。输入数据可以是：

• NumPy数组
• a TensorFlow张量或张量
• a tf.data dataset

列表

您可以使用abalone_model.predict(YourData)，其中YourData是提到的数据类型之一。当然，您可以在您自己的培训数据上使用predict()，这可能会过于合适。如果提供或拆分了可用的数据集，请在断开的验证或测试集上进行尝试。在这里，您可以找到一个很好的回归问题的例子，比如您所面临的问题：https://www.tensorflow.org/tutorials/keras/regression