Keras 전처리 레이어를 사용한 구조적 데이터 분류

import numpy as np
import pandas as pd
import tensorflow as tf

from tensorflow.keras import layers

2022-12-14 23:21:31.042474: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
2022-12-14 23:21:31.042605: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
2022-12-14 23:21:31.042617: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

tf.__version__

'2.11.0'

dataset_url = 'http://storage.googleapis.com/download.tensorflow.org/data/petfinder-mini.zip'
csv_file = 'datasets/petfinder-mini/petfinder-mini.csv'

tf.keras.utils.get_file('petfinder_mini.zip', dataset_url,
                        extract=True, cache_dir='.')
dataframe = pd.read_csv(csv_file)

Downloading data from http://storage.googleapis.com/download.tensorflow.org/data/petfinder-mini.zip
1668792/1668792 [==============================] - 0s 0us/step

dataframe.head()

# In the original dataset, `'AdoptionSpeed'` of `4` indicates
# a pet was not adopted.
dataframe['target'] = np.where(dataframe['AdoptionSpeed']==4, 0, 1)

# Drop unused features.
dataframe = dataframe.drop(columns=['AdoptionSpeed', 'Description'])

train, val, test = np.split(dataframe.sample(frac=1), [int(0.8*len(dataframe)), int(0.9*len(dataframe))])

print(len(train), 'training examples')
print(len(val), 'validation examples')
print(len(test), 'test examples')

9229 training examples
1154 validation examples
1154 test examples

def df_to_dataset(dataframe, shuffle=True, batch_size=32):
  df = dataframe.copy()
  labels = df.pop('target')
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}
  ds = tf.data.Dataset.from_tensor_slices((dict(df), labels))
  if shuffle:
    ds = ds.shuffle(buffer_size=len(dataframe))
  ds = ds.batch(batch_size)
  ds = ds.prefetch(batch_size)
  return ds

batch_size = 5
train_ds = df_to_dataset(train, batch_size=batch_size)

/tmpfs/tmp/ipykernel_775031/3479182957.py:4: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}

[(train_features, label_batch)] = train_ds.take(1)
print('Every feature:', list(train_features.keys()))
print('A batch of ages:', train_features['Age'])
print('A batch of targets:', label_batch )

Every feature: ['Type', 'Age', 'Breed1', 'Gender', 'Color1', 'Color2', 'MaturitySize', 'FurLength', 'Vaccinated', 'Sterilized', 'Health', 'Fee', 'PhotoAmt', 'target']
A batch of ages: tf.Tensor(
[[3]
 [7]
 [3]
 [5]
 [1]], shape=(5, 1), dtype=int64)
A batch of targets: tf.Tensor([0 0 1 1 0], shape=(5,), dtype=int64)

def get_normalization_layer(name, dataset):
  # Create a Normalization layer for the feature.
  normalizer = layers.Normalization(axis=None)

  # Prepare a Dataset that only yields the feature.
  feature_ds = dataset.map(lambda x, y: x[name])

  # Learn the statistics of the data.
  normalizer.adapt(feature_ds)

  return normalizer

photo_count_col = train_features['PhotoAmt']
layer = get_normalization_layer('PhotoAmt', train_ds)
layer(photo_count_col)

WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/autograph/pyct/static_analysis/liveness.py:83: Analyzer.lamba_check (from tensorflow.python.autograph.pyct.static_analysis.liveness) is deprecated and will be removed after 2023-09-23.
Instructions for updating:
Lambda fuctions will be no more assumed to be used in the statement where they are used, or at least in the same block. https://github.com/tensorflow/tensorflow/issues/56089
<tf.Tensor: shape=(5, 1), dtype=float32, numpy=
array([[ 0.7534282 ],
       [ 6.423487  ],
       [-0.19158155],
       [-0.5065848 ],
       [-0.8215881 ]], dtype=float32)>

def get_category_encoding_layer(name, dataset, dtype, max_tokens=None):
  # Create a layer that turns strings into integer indices.
  if dtype == 'string':
    index = layers.StringLookup(max_tokens=max_tokens)
  # Otherwise, create a layer that turns integer values into integer indices.
  else:
    index = layers.IntegerLookup(max_tokens=max_tokens)

  # Prepare a `tf.data.Dataset` that only yields the feature.
  feature_ds = dataset.map(lambda x, y: x[name])

  # Learn the set of possible values and assign them a fixed integer index.
  index.adapt(feature_ds)

  # Encode the integer indices.
  encoder = layers.CategoryEncoding(num_tokens=index.vocabulary_size())

  # Apply multi-hot encoding to the indices. The lambda function captures the
  # layer, so you can use them, or include them in the Keras Functional model later.
  return lambda feature: encoder(index(feature))

test_type_col = train_features['Type']
test_type_layer = get_category_encoding_layer(name='Type',
                                              dataset=train_ds,
                                              dtype='string')
test_type_layer(test_type_col)

<tf.Tensor: shape=(5, 3), dtype=float32, numpy=
array([[0., 0., 1.],
       [0., 1., 0.],
       [0., 0., 1.],
       [0., 1., 0.],
       [0., 1., 0.]], dtype=float32)>

test_age_col = train_features['Age']
test_age_layer = get_category_encoding_layer(name='Age',
                                             dataset=train_ds,
                                             dtype='int64',
                                             max_tokens=5)
test_age_layer(test_age_col)

<tf.Tensor: shape=(5, 5), dtype=float32, numpy=
array([[0., 0., 1., 0., 0.],
       [1., 0., 0., 0., 0.],
       [0., 0., 1., 0., 0.],
       [1., 0., 0., 0., 0.],
       [0., 0., 0., 1., 0.]], dtype=float32)>

batch_size = 256
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)

/tmpfs/tmp/ipykernel_775031/3479182957.py:4: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}
/tmpfs/tmp/ipykernel_775031/3479182957.py:4: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}
/tmpfs/tmp/ipykernel_775031/3479182957.py:4: FutureWarning: Support for multi-dimensional indexing (e.g. `obj[:, None]`) is deprecated and will be removed in a future version.  Convert to a numpy array before indexing instead.
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}

all_inputs = []
encoded_features = []

# Numerical features.
for header in ['PhotoAmt', 'Fee']:
  numeric_col = tf.keras.Input(shape=(1,), name=header)
  normalization_layer = get_normalization_layer(header, train_ds)
  encoded_numeric_col = normalization_layer(numeric_col)
  all_inputs.append(numeric_col)
  encoded_features.append(encoded_numeric_col)

age_col = tf.keras.Input(shape=(1,), name='Age', dtype='int64')

encoding_layer = get_category_encoding_layer(name='Age',
                                             dataset=train_ds,
                                             dtype='int64',
                                             max_tokens=5)
encoded_age_col = encoding_layer(age_col)
all_inputs.append(age_col)
encoded_features.append(encoded_age_col)

categorical_cols = ['Type', 'Color1', 'Color2', 'Gender', 'MaturitySize',
                    'FurLength', 'Vaccinated', 'Sterilized', 'Health', 'Breed1']

for header in categorical_cols:
  categorical_col = tf.keras.Input(shape=(1,), name=header, dtype='string')
  encoding_layer = get_category_encoding_layer(name=header,
                                               dataset=train_ds,
                                               dtype='string',
                                               max_tokens=5)
  encoded_categorical_col = encoding_layer(categorical_col)
  all_inputs.append(categorical_col)
  encoded_features.append(encoded_categorical_col)

all_features = tf.keras.layers.concatenate(encoded_features)
x = tf.keras.layers.Dense(32, activation="relu")(all_features)
x = tf.keras.layers.Dropout(0.5)(x)
output = tf.keras.layers.Dense(1)(x)

model = tf.keras.Model(all_inputs, output)

model.compile(optimizer='adam',
              loss=tf.keras.losses.BinaryCrossentropy(from_logits=True),
              metrics=["accuracy"])

# Use `rankdir='LR'` to make the graph horizontal.
tf.keras.utils.plot_model(model, show_shapes=True, rankdir="LR")

model.fit(train_ds, epochs=10, validation_data=val_ds)

Epoch 1/10
/tmpfs/src/tf_docs_env/lib/python3.9/site-packages/keras/engine/functional.py:638: UserWarning: Input dict contained keys ['target'] which did not match any model input. They will be ignored by the model.
  inputs = self._flatten_to_reference_inputs(inputs)
37/37 [==============================] - 3s 21ms/step - loss: 0.7197 - accuracy: 0.3228 - val_loss: 0.6172 - val_accuracy: 0.5078
Epoch 2/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5970 - accuracy: 0.5875 - val_loss: 0.5807 - val_accuracy: 0.6941
Epoch 3/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5761 - accuracy: 0.6721 - val_loss: 0.5660 - val_accuracy: 0.7218
Epoch 4/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5645 - accuracy: 0.6873 - val_loss: 0.5552 - val_accuracy: 0.7496
Epoch 5/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5512 - accuracy: 0.7009 - val_loss: 0.5462 - val_accuracy: 0.7530
Epoch 6/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5446 - accuracy: 0.7143 - val_loss: 0.5399 - val_accuracy: 0.7530
Epoch 7/10
37/37 [==============================] - 0s 7ms/step - loss: 0.5445 - accuracy: 0.7054 - val_loss: 0.5354 - val_accuracy: 0.7556
Epoch 8/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5330 - accuracy: 0.7201 - val_loss: 0.5317 - val_accuracy: 0.7556
Epoch 9/10
37/37 [==============================] - 0s 8ms/step - loss: 0.5350 - accuracy: 0.7199 - val_loss: 0.5289 - val_accuracy: 0.7582
Epoch 10/10
37/37 [==============================] - 0s 7ms/step - loss: 0.5264 - accuracy: 0.7300 - val_loss: 0.5274 - val_accuracy: 0.7608
<keras.callbacks.History at 0x7fe2f00c3fa0>

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

5/5 [==============================] - 0s 7ms/step - loss: 0.5233 - accuracy: 0.7444
Accuracy 0.7443674206733704

model.save('my_pet_classifier')
reloaded_model = tf.keras.models.load_model('my_pet_classifier')

WARNING:absl:Function `_wrapped_model` contains input name(s) PhotoAmt, Fee, Age, Type, Color1, Color2, Gender, MaturitySize, FurLength, Vaccinated, Sterilized, Health, Breed1 with unsupported characters which will be renamed to photoamt, fee, age, type, color1, color2, gender, maturitysize, furlength, vaccinated, sterilized, health, breed1 in the SavedModel.
INFO:tensorflow:Assets written to: my_pet_classifier/assets
INFO:tensorflow:Assets written to: my_pet_classifier/assets

sample = {
    'Type': 'Cat',
    'Age': 3,
    'Breed1': 'Tabby',
    'Gender': 'Male',
    'Color1': 'Black',
    'Color2': 'White',
    'MaturitySize': 'Small',
    'FurLength': 'Short',
    'Vaccinated': 'No',
    'Sterilized': 'No',
    'Health': 'Healthy',
    'Fee': 100,
    'PhotoAmt': 2,
}

input_dict = {name: tf.convert_to_tensor([value]) for name, value in sample.items()}
predictions = reloaded_model.predict(input_dict)
prob = tf.nn.sigmoid(predictions[0])

print(
    "This particular pet had a %.1f percent probability "
    "of getting adopted." % (100 * prob)
)

1/1 [==============================] - 0s 469ms/step
This particular pet had a 79.3 percent probability of getting adopted.