Tell about the transaction isolation levels in PostgreSQL and why they are needed.

In PostgreSQL, there are four main transaction isolation levels that determine how data changes are visible across transactions and how anomalies are prevented during concurrent access:

1. Read Uncommitted — the lowest isolation level. Allows seeing even uncommitted changes of other transactions (dirty read). PostgreSQL effectively behaves like Read Committed.

2. Read Committed (default level) — a transaction sees only data that was committed at the start of each individual query. This prevents dirty reads but allows non-repeatable reads and phantom reads.

3. Repeatable Read — a transaction sees data as it was at the start of the transaction. This prevents non-repeatable reads but phantom reads are still possible.

4. Serializable — the strictest level, providing full transaction isolation as if transactions were executed sequentially. It prevents all anomalies, including phantom reads, but may lead to conflicts and transaction rollbacks.

Reasons for using isolation levels:

• To balance between performance and data consistency.
• To control which read anomalies are permissible in a specific application.
• To prevent conflicts during concurrent data access.

Example of setting the isolation level in PostgreSQL:

BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE;
-- operations
COMMIT;

In Python, using psycopg2, you can set the isolation level as follows:

import psycopg2
conn = psycopg2.connect(...)
conn.set_isolation_level(psycopg2.extensions.ISOLATION_LEVEL_SERIALIZABLE)