def row_feature_hasher(row):
hash_counts = collections.defaultdict(int)
for column in self.columns:
hashed_value = simple_hash(row[column], self.num_features)
hash_counts[hashed_value] = hash_counts[hashed_value] + 1
return {
f"hash_{joined_columns}_{i}": hash_counts[i]
for i in range(self.num_features)
}
def test_simple_hash():
# Tests simple_hash determinism.
assert simple_hash(1, 100) == 83
assert simple_hash("a", 100) == 52
assert simple_hash("banana", 100) == 16
assert simple_hash([1, 2, "apple"], 100) == 37
def hash_count(tokens: List[str]) -> Counter:
hashed_tokens = [simple_hash(token, self.num_features) for token in tokens]
return Counter(hashed_tokens)