TensorFlow 2.x: Cannot save trained model in h5 format (OSError: Unable to create link (name already exists))

Question

我的模型使用预处理数据来预测客户是私人客户还是非私人客户。预处理步骤使用 feature_column.bucketized_column(…)、feature_column.embedding_column(…) 等步骤。训练结束后，我试图保存模型，但出现以下错误：

File "h5py_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
File "h5py_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
File "h5py\h5o.pyx", line 202, in h5py.h5o.link
OSError: Unable to create link (name already exists)

我尝试了以下方法来解决我的问题：

我试图排除这里提到的优化器：https://github.com/tensorflow/tensorflow/issues/27688。
我尝试了不同版本的 TensorFlow，例如 2.2 和 2.3。
我尝试按照此处所述重新安装 h5py：RuntimeError: Unable to create link (name already exists) when I append hdf5 file?。

一事无成！

模型相关代码如下：

(feature_columns, train_ds, val_ds, test_ds) = preprocessing.getPreProcessedDatasets(args.data, args.zip, args.batchSize)

feature_layer = tf.keras.layers.DenseFeatures(feature_columns, trainable=False)

model = tf.keras.models.Sequential([
        feature_layer,
        tf.keras.layers.Dense(1, activation=tf.nn.sigmoid)
    ])

model.compile(optimizer='sgd',
        loss='binary_crossentropy',
        metrics=['accuracy'])

paramString = "Arg-e{}-b{}-z{}".format(args.epoch, args.batchSize, bucketSizeGEO)

...

model.fit(train_ds,
              validation_data=val_ds,
              epochs=args.epoch,
              callbacks=[tensorboard_callback])


model.summary()

loss, accuracy = model.evaluate(test_ds)
print("Accuracy", accuracy)

paramString = paramString + "-a{:.4f}".format(accuracy)

outputName = "logReg" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S") + paramStrin

if args.saveModel:
       filepath = "./saved_models/" + outputName + ".h5"
       model.save(filepath, save_format='h5')

预处理模块中调用的函数：

def getPreProcessedDatasets(filepath, zippath, batch_size, bucketSizeGEO):
    print("start preprocessing...")

    path = filepath
    data = pd.read_csv(path, dtype={
    "NAME1": np.str_, 
    "NAME2": np.str_, 
    "EMAIL1": np.str_, 
    "ZIP": np.str_, 
    "STREET": np.str_, 
    "LONGITUDE":np.floating, 
    "LATITUDE": np.floating, 
    "RECEIVERTYPE": np.int64}) 

    feature_columns = []

    data = data.fillna("NaN")

    data = __preProcessName(data)
    data = __preProcessStreet(data)
    
    train, test = train_test_split(data, test_size=0.2, random_state=0)
    train, val = train_test_split(train, test_size=0.2, random_state=0)

    train_ds = __df_to_dataset(train, batch_size=batch_size)
    val_ds = __df_to_dataset(val, shuffle=False, batch_size=batch_size)
    test_ds = __df_to_dataset(test, shuffle=False, batch_size=batch_size)


    __buildFeatureColums(feature_columns, data, zippath, bucketSizeGEO, True)

    print("preprocessing completed")

    return (feature_columns, train_ds, val_ds, test_ds)

调用特征的不同预处理函数：

def __buildFeatureColums(feature_columns, data, zippath, bucketSizeGEO, addCrossedFeatures):
    
    feature_columns.append(__getFutureColumnLon(bucketSizeGEO))
    feature_columns.append(__getFutureColumnLat(bucketSizeGEO))
    
    (namew1_one_hot, namew2_one_hot) = __getFutureColumnsName(__getNumberOfWords(data, 'NAME1PRO'))
    feature_columns.append(namew1_one_hot)
    feature_columns.append(namew2_one_hot)
    
    feature_columns.append(__getFutureColumnStreet(__getNumberOfWords(data, 'STREETPRO')))
    
    feature_columns.append(__getFutureColumnZIP(2223, zippath))
    
    if addCrossedFeatures:
        feature_columns.append(__getFutureColumnCrossedNames(100))
        feature_columns.append(__getFutureColumnCrossedZIPStreet(100, 2223, zippath))

与嵌入相关的函数：

def __getFutureColumnsName(name_num_words):
    vocabulary_list = np.arange(0, name_num_words + 1, 1).tolist()

    namew1_voc = tf.feature_column.categorical_column_with_vocabulary_list(
        key='NAME1W1', vocabulary_list=vocabulary_list, dtype=tf.dtypes.int64)
    namew2_voc = tf.feature_column.categorical_column_with_vocabulary_list(
        key='NAME1W2', vocabulary_list=vocabulary_list, dtype=tf.dtypes.int64)

    dim = __getNumberOfDimensions(name_num_words)

    namew1_embedding = feature_column.embedding_column(namew1_voc, dimension=dim)
    namew2_embedding = feature_column.embedding_column(namew2_voc, dimension=dim)

    return (namew1_embedding, namew2_embedding)

def __getFutureColumnStreet(street_num_words):
    vocabulary_list = np.arange(0, street_num_words + 1, 1).tolist()

    street_voc = tf.feature_column.categorical_column_with_vocabulary_list(
        key='STREETW', vocabulary_list=vocabulary_list, dtype=tf.dtypes.int64)

    dim = __getNumberOfDimensions(street_num_words)

    street_embedding = feature_column.embedding_column(street_voc, dimension=dim)

    return street_embedding

def __getFutureColumnZIP(zip_num_words, zippath):
    zip_voc = feature_column.categorical_column_with_vocabulary_file(
    key='ZIP', vocabulary_file=zippath, vocabulary_size=zip_num_words,
    default_value=0)

    dim = __getNumberOfDimensions(zip_num_words)

    zip_embedding = feature_column.embedding_column(zip_voc, dimension=dim)

    return zip_embedding

Answer 1

以h5格式保存模型时的错误OSError: Unable to create link (name already exists)是由于一些重复的变量名引起的。 for i, w in enumerate(model.weights): print(i, w.name) 检查表明它们是 embedding_weights 个名字。

通常，在构建 feature_column 时，传入每个特征列的不同 key 将用于构建不同的变量 name。这在 TF 2.1 中工作正常，但在 TF 2.2 和 2.3 中出现故障，据推测 fixed in TF 2.4 nigthly.

Answer 2

我的 TF 2.3 解决方法是基于@SajanGohil 的评论，但我的问题是 weight 名称（不是 layer 名称）：

for i in range(len(model.weights)):
    model.weights[i]._handle_name = model.weights[i].name + "_" + str(i)

同样的警告适用：这种方法操纵 TF 内部结构，因此不是面向未来的。

Answer 3

我发现当我从模型检查点加载模型时也会出现这种情况，model.compile 它具有相同的优化器、指标和损失函数，并对其进行训练。但是如果我避免用相同的参数再次编译它，这个错误信息就不会再出现了。

TensorFlow 2.x: Cannot save trained model in h5 format (OSError: Unable to create link (name already exists))

TensorFlow 2.x: Cannot save trained model in h5 format (OSError: Unable to create link (name already exists))

python

tensorflow

machine-learning

keras

h5py