Getting Started#

Install EvaDB#

To install EvaDB, we recommend using the pip package manager.

1. Create a new virtual environment called evadb-venv.

python -m venv evadb-venv

Now, activate the virtual environment:

source evadb-venv/bin/activate

  1. Once inside the virtual environment, run the command below to mitigate the dependency issues.

pip install --upgrade pip setuptools wheel

  1. Install EvaDB

pip install evadb

  1. Verify EvaDB installation

pip freeze

You should see a list of installed packages including but not limited to the following:

Package           Version
----------------- -------
aenum             3.1.15
decorator         5.1.1
diskcache         5.6.3
evadb             0.3.3
greenlet          2.0.2
lark              1.1.7
numpy             1.25.2
pandas            2.1.0
...

  1. Run EvaDB

Copy the following Python program to a file called run_evadb.py.

The program runs a SQL query for listing all the built-in functions in EvaDB. It consists of importing and connecting to EvaDB, and then running the query. The query’s result is returned as a Dataframe.

# Import the EvaDB package
import evadb

# Connect to EvaDB and get a database cursor for running queries
cursor = evadb.connect().cursor()

# List all the built-in functions in EvaDB
print(cursor.query("SHOW FUNCTIONS;").df())

Now, run the Python program:

python -m run_evadb.py

You should see a list of built-in functions including but not limited to the following:

        name                                             inputs  ...                                               impl metadata
0  ArrayCount   [Input_Array NDARRAY ANYTYPE (), Search_Key ANY]  ...  /home/jarulraj3/evadb/evadb/functions/ndarray/array...       []
1        Crop  [Frame_Array NDARRAY UINT8 (3, None, None), bb...  ...   /home/jarulraj3/evadb/evadb/functions/ndarray/crop.py       []
2     ChatGPT  [query NDARRAY STR (1,), content NDARRAY STR (...  ...        /home/jarulraj3/evadb/evadb/functions/chatgpt.py       []

[3 rows x 6 columns]

Note

Go over the Python API to learn more about connect() and cursor.

Note

EvaDB supports additional installation options for extending its functionality. Go over the Installation Options for all the available options.

Illustrative AI Query#

Here is an illustrative MNIST image classification AI query in EvaDB.

--- This AI query retrieves images in the loaded MNIST video with label 4
--- We constrain the query to only search through the first 100 frames
--- We limit the query to only return the first five frames with label 4
SELECT data, id, MnistImageClassifier(data)
FROM MnistVideo
WHERE MnistImageClassifier(data) = '4' AND id < 100
LIMIT 5;

The complete MNIST notebook is available on Colab. Try out EvaDB by experimenting with this introductory notebook.