Accessing DataFrame with [('features', 'vector'), ('label', 'double')] schema

accessing_dataframe_with_vector_double_schema.py

from pyspark.mllib.linalg import Vectors

# Sample dataset 
data = sc.parallelize([
  (0.0, [0.0, 1.0, 2.0]),
  (1.0, [1.0, 2.0, 3.0]),
  (3.0, [2.0, 3.0, 4.0]),
  (2.0, [3.0, 4.0, 5.0])
])

# Load each word and create row object
parts = data.map(lambda t: Row(label=t[0], features=Vectors.dense(t[1])))

# Infer schema (using reflection)
df = parts.toDF()

# Run selectExpr
df.selectExpr("max(label) as max_value","min(label) as min_value").show()