eileen-code4fun

from dgl.nn.tensorflow import GraphConv

class GCN(tf.keras.Model):
    def __init__(self, g, feat_dim, hidden_dim, class_num):
        super(GCN, self).__init__()
        self.g = g
        self.h1 = GraphConv(feat_dim, hidden_dim, activation=tf.nn.relu, allow_zero_in_degree=True)
        self.dropout = tf.keras.layers.Dropout(0.5)
        self.h2 = GraphConv(hidden_dim, class_num), allow_zero_in_degree=True
    

eileen-code4fun

model = tf.keras.models.Sequential()
model.add(tf.keras.layers.Input(shape=(1433,)))
model.add(tf.keras.layers.Dense(32, activation='relu'))
model.add(tf.keras.layers.Dropout(0.5))
model.add(tf.keras.layers.Dense(7))

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

eileen-code4fun

import dgl
import tensorflow as tf

dataset = dgl.data.CoraGraphDataset()

# A DGL dataset may contain multiple graphs.
# In the case of Cora, there is only one graph.
g = dataset[0]

# g.ndata is a dictionary of nodes related data.

eileen-code4fun

import tensorflow_data_validation as tfdv
from tensorflow_metadata.proto.v0 import anomalies_pb2

stats = tfdv.load_stats_binary(GCS_PATH_FOR_STATS_PB)
tfdv.visualize_statistics(stats)
# Display output in screenshot later.

schema = tfdv.load_schema_text(GCS_PATH_FOR_SCHEMA_PBTXT)
tfdv.display_schema(schema)
# Display output in screenshot later.

eileen-code4fun

import tensorflow as tf

def normalize(image, label):
  return tf.cast(image, dtype=tf.float32) / 255.0, label

train = ds['train'].map(normalize).batch(32)
test = ds['test'].map(normalize).batch(32)

encoding_model = tf.keras.Model(inputs=model.input, outputs=model.get_layer('flatten').output)

eileen-code4fun

import tensorflow as tf

# Load the saved custom model.
model = tf.keras.models.load_model(GCS_PATH_FOR_SAVED_MODEL)
model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
              metrics=['accuracy'])

model.summary()
'''

eileen-code4fun

import numpy as np
from sklearn import decomposition

# Reuse the same ds from Tensorflow Dataset
train_data = list(ds['train'].map(lambda x, y: tf.reshape(x, [-1])))
test_data = list(ds['test'].map(lambda x, y: tf.reshape(x, [-1])))
X = tf.concat([train_data, test_data], 0).numpy()
print(np.shape(X))
# Output: (60000, 3072)

eileen-code4fun

SELECT
  # Display all metrics.
  *
FROM
ML.EVALUATE(
  MODEL `bqml.e2e-tutorial`,
  (
    SELECT
      int64_field_0 as label,
      * EXCEPT (int64_field_0)

eileen-code4fun

CREATE MODEL
  # Name of the model.
  `bqml.e2e-tutorial`
OPTIONS
  # BigQuery will detect that it's for multiclass.
  ( MODEL_TYPE='LOGISTIC_REG') AS
SELECT
  # Use the first column as the label.
  int64_field_0 as label,
  # The rest are features.

eileen-code4fun

SELECT * FROM `bqml.train` LIMIT 10;