從零基礎入門Tensorflow2.0 ----五、25TF1.0自定義estimator

every blog every motto:

0. 前言

tf1.0 自定義estimator

1. 代碼部分

1. 導入模塊

import matplotlib as mpl
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import sklearn
import pandas as pd
import os
import sys
import time
import tensorflow as tf
from tensorflow import keras

print(tf.__version__)
print(sys.version_info)
for module in mpl,np,pd,sklearn,tf,keras:
    print(module.__name__,module.__version__)

2. 讀取數據

# 讀取數據
train_file = './data/train.csv'
eval_file = './data/eval.csv'

train_df = pd.read_csv(train_file)
eval_df = pd.read_csv(eval_file)

print(train_df.head())
print(eval_df.head())

y_train = train_df.pop('survived')
y_eval = eval_df.pop('survived')

print(train_df.head())
print(eval_df.head())
print(y_train.head())
print(y_eval.head())

3. feature_columns

# 離散特徵
categorical_colums = ['sex','n_siblings_spouses','parch','class','deck','embark_town','alone']
numerical_columns = ['age','fare']

feature_columns = []
# 離散
for categorical_column in categorical_colums:
    vocab = train_df[categorical_column].unique() # 獲取所有可能的值
    print(categorical_column,vocab)
    # one-hot 編碼，
    feature_columns.append(
        tf.feature_column.indicator_column(tf.feature_column.categorical_column_with_vocabulary_list(categorical_column,vocab)))

# 連續
for categorical_column in numerical_columns:
    feature_columns.append(tf.feature_column.numeric_column(categorical_column,dtype=tf.float32))

4. 構建dataset

# 構建dataset
def make_dataset(data_df,label_df,epochs=10,shuffle=True,batch_size = 32):
    dataset = tf.data.Dataset.from_tensor_slices((dict(data_df),label_df))
    if shuffle:
        dataset = dataset.shuffle(10000)
    dataset = dataset.repeat(epochs).batch(batch_size)
    return dataset.make_one_shot_iterator().get_next()

5. 自定義estimator

output_dir = "customized_estimator"
if not os.path.exists(output_dir):
    os.mkdir(output_dir)

def model_fn(features,labels,mode,params):
    # model runtime state : Train Eval,Predict
    input_for_next_layer = tf.feature_column.input_layer(features,params['feature_columns'])

    for n_unit in params['hidden_units']:
        input_for_next_layer = tf.layers.dense(input_for_next_layer,units=n_unit,activation=tf.nn.relu)

    logits = tf.layers.dense(input_for_next_layer,params['n_classes'],activation = None)

    predicted_classes = tf.argmax(logits,1)
    if mode == tf.estimator.ModeKeys.PREDICT:
        predictions = {
            "classes_ids":predicted_classes[:tf.newaxis],
            "probabilities":tf.nn.softmax(logits),
            "logits":logits
                }
        
        return tf.estimator.EstimatorSpec(mode,predictions=predictions)
    
    loss = tf.losses.sparse_softmax_cross_entropy(labels=labels,logits=logits)
    
    accuracy = tf.metrics.accuracy(labels=labels,predictions=predicted_classes,name="acc_op")
    metrics = {"accuracy":accuracy}
        
    if mode == tf.estimator.ModeKeys.EVAL:
        return tf.estimator.EstimatorSpec(mode,loss=loss,eval_metric_ops=metrics)
    
    optimizer = tf.train.AdamOptimizer()
    train_op = optimizer.minimzer(loss,global_step=tf.train.get_global_step())
    
    if mode == tf.estimator.ModeKeys.TRAINS:
        return tf.estimator.EstimatorSpec(mode,loss=loss,train_op = train_op)
    
   

estimator = tf.estimator.Estimator(
    model_fn = model_fn,
    model_dir = output_dir,
    params = {
        "feature_columns":feature_columns,
        "hidden_units":[100,100],
        "n_classes":2
    }
)

estimator.train(input_fn = lambda : make_dataset(train_df,y_train,epochs=100))