Tensorflow InvalidArgumentError：断言失败：[标签必须是 <= n_classes - 1]

Question

import pandas as pd
import matplotlib.pyplot as plt
import tensorflow as tf
df = pd.read_csv('pokemon_data.csv')

df['Total'] = df['HP'] + df['Attack'] + df['Defense'] + df['Sp. Atk'] + df['Sp. Def'] + df['Speed']

df = df.loc[df['Total'] > 450]
df = df.loc[~df['Name'].str.contains('Mega')]
df = df.loc[~df['Name'].str.contains('Primal')]

df = df.drop(columns = ['Name'])
df = df.drop(columns = ['Generation'])
df = df.drop(columns = ['Legendary'])
df = df.drop(columns = ['Type 2'])
df = df.drop(columns = ['#'])

df.loc[df['Type 1'] == 'Fire', 'Type 1'] = 0
df.loc[df['Type 1'] == 'Normal', 'Type 1'] = 1
df.loc[df['Type 1'] == 'Water', 'Type 1'] = 2
df.loc[df['Type 1'] == 'Grass', 'Type 1'] = 3
df.loc[df['Type 1'] == 'Electric', 'Type 1'] = 4
df.loc[df['Type 1'] == 'Ice', 'Type 1'] = 5
df.loc[df['Type 1'] == 'Fighting', 'Type 1'] = 6
df.loc[df['Type 1'] == 'Poison', 'Type 1'] = 7
df.loc[df['Type 1'] == 'Ground', 'Type 1'] = 8
df.loc[df['Type 1'] == 'Flying', 'Type 1'] = 9
df.loc[df['Type 1'] == 'Psychic', 'Type 1'] = 10
df.loc[df['Type 1'] == 'Rock', 'Type 1'] = 11
df.loc[df['Type 1'] == 'Bug', 'Type 1'] = 12
df.loc[df['Type 1'] == 'Ghost', 'Type 1'] = 13
df.loc[df['Type 1'] == 'Dark', 'Type 1'] = 14
df.loc[df['Type 1'] == 'Dragon', 'Type 1'] = 15
df.loc[df['Type 1'] == 'Steel', 'Type 1'] = 16
df.loc[df['Type 1'] == 'Fairy', 'Type 1'] = 17

TEMP = ['Type 1']
for col in TEMP:
    df[col] = pd.to_numeric(df[col])

df_eval_sub = df.loc[df['Total'] < 500]
df_eval_over = df.loc[df['Total'] > 500]                      
y_train = df.pop('Type 1')
y_eval_sub = df_eval_sub.pop('Type 1')
y_eval_over = df_eval_over.pop('Type 1')                     
                      
feature_columns = []

TO_INT = ['HP', 'Attack', 'Defense', 'Sp. Atk', 'Sp. Def', 'Speed', 'Total']
for col in TO_INT:
    df[col] = pd.to_numeric(df[col])


NUMERIC_COLUMNS = ['HP', 'Attack', 'Defense', 'Sp. Atk', 'Sp. Def', 'Speed', 'Total']
for feature_name in NUMERIC_COLUMNS:
    feature_columns.append(tf.feature_column.numeric_column(feature_name, dtype=tf.float32))

def make_input_fn(data_df, label_df, num_epochs = 10, shuffle = True, batch_size=32):
    def input_function():
        ds = tf.data.Dataset.from_tensor_slices((dict(data_df), label_df))
        if shuffle:
            ds = ds.shuffle(1000)
        ds = ds.batch(batch_size).repeat(num_epochs)
        return ds
    return input_function

train_input_fn = make_input_fn(df, y_train)
eval_input_fn = make_input_fn(df_eval_sub, y_eval_sub, num_epochs = 1, shuffle = False)

linear_est = tf.estimator.LinearClassifier(feature_columns=feature_columns)
linear_est.train(train_input_fn)
result = linear_est.evaluate(eval_input_fn)

错误：

InvalidArgumentError: assertion failed: [Labels must be <= n_classes - 1] [Condition x <= y did not hold element-wise:] [x (head/losses/Cast:0) = ] [[16][0][11]...] [y (head/losses/check_label_range/Const:0) = ] [1]
     [[{{node Assert}}]]

我假设 n_classes 表示列数，所以在清理数据框后，我还剩下 8 列，然后我删除了 1 列，这是我用于 y_train 的列。错误触发于

linear_est.train(train_input_fn)

我在 y_train 中使用的列有 18 个不同的值，但不确定这是否相关。

Answer 1

n_classes不是列数，是tf.estimator.LinearClassifier中的参数，你必须指定，标签中的类必须n_classes，在你的情况下你应该设置linear_est = tf.estimator.LinearClassifier(feature_columns=feature_columns, n_classes=18)