New to tensor databases

[tachochain]

how to create a tensor DBMS

[AElnamaki]

import numpy as np
import sqlite3

class TensorDB:
def init(self, db_file):
self.conn = sqlite3.connect(db_file)
self.cursor = self.conn.cursor()
self._create_tables()

def _create_tables(self):
    self.cursor.execute("""
        CREATE TABLE Tensors (
            id INTEGER PRIMARY KEY,
            name TEXT,
            shape TEXT,
            data BLOB
        )
    """)
    self.conn.commit()

def store_tensor(self, name, tensor):
    shape = tensor.shape
    data = tensor.tobytes()
    self.cursor.execute("""
        INSERT INTO Tensors (name, shape, data) VALUES (?, ?, ?)
    """, (name, str(shape), data))
    self.conn.commit()

def retrieve_tensor(self, name):
    self.cursor.execute("""
        SELECT shape, data FROM Tensors WHERE name=?
    """, (name,))
    row = self.cursor.fetchone()
    if row:
        shape = tuple(map(int, row[0].split(',')))
        data = np.frombuffer(row[1], dtype=np.float32).reshape(shape)
        return data
    else:
        return None

def close(self):
    self.conn.close()

Example usage:

tensor_db = TensorDB("my_tensor_db.sqlite")

Store a tensor

tensor_name = "example_tensor"
tensor_data = np.random.randn(3, 3)
tensor_db.store_tensor(tensor_name, tensor_data)

Retrieve a tensor

retrieved_tensor = tensor_db.retrieve_tensor(tensor_name)
print("Retrieved tensor:")
print(retrieved_tensor)

tensor_db.close()

[AElnamaki]

provides a basic implementation of a tensor DBMS in Python using SQLite. Depending on your requirements

[Shanibhai302t2]

def _create_tables(self):
self.cursor.execute("""
CREATE TABLE Tensors (
id INTEGER PRIMARY KEY,
name TEXT,
shape TEXT,
data BLOB
)
""")
self.conn.commit()

def store_tensor(self, name, tensor):
shape = tensor.shape
data = tensor.tobytes()
self.cursor.execute("""
INSERT INTO Tensors (name, shape, data) VALUES (?, ?, ?)
""", (name, str(shape), data))
self.conn.commit()

def retrieve_tensor(self, name):
self.cursor.execute("""
SELECT shape, data FROM Tensors WHERE name=?
""", (name,))
row = self.cursor.fetchone()
if row:
shape = tuple(map(int, row[0].split(',')))
data = np.frombuffer(row[1], dtype=np.float32).reshape(shape)
return data
else:
return None

def close(self):
self.conn.close()

[AElnamaki]

nice

[chris6090]

how to create a tensor DBMS

Kriseng83@gmail.com, Julielove31624@gmail.com, seanoxcoltd@gmail.com, callemchurch@gmail.com, changyoson57@gmail.com