[CodeGemma 1]Common Use Cases
Copyright 2024 Google LLC.
CodeGemma - Common use cases
This notebook demonstrates the basic task that Gemma can solve by using the right prompting.
Setup
Select the Colab runtime
To complete this tutorial, you'll need to have a Colab runtime with sufficient resources to run the Gemma model. In this case, you can use a T4 GPU:
- In the upper-right of the Colab window, select ▾ (Additional connection options).
- Select Change runtime type.
- Under Hardware accelerator, select T4 GPU.
Gemma setup
To complete this tutorial, you'll first need to complete the setup instructions at Gemma setup. The Gemma setup instructions show you how to do the following:
- Get access to Gemma on kaggle.com.
- Select a Colab runtime with sufficient resources to run the Gemma 2B model.
- Generate and configure a Kaggle username and an API key as Colab secrets.
After you've completed the Gemma setup, move on to the next section, where you'll set environment variables for your Colab environment.
Configure your credentials
Add your Kaggle credentials to the Colab Secrets manager to securely store it.
- Open your Google Colab notebook and click on the 🔑 Secrets tab in the left panel.

- Create new secrets:
KAGGLE_USERNAMEandKAGGLE_KEY - Copy/paste your username into
KAGGLE_USERNAME - Copy/paste your key into
KAGGLE_KEY - Toggle the buttons on the left to allow notebook access to the secrets.
Install dependencies
Run the cell below to install all the required dependencies.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 9.3 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 570.5/570.5 kB 15.3 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 311.2/311.2 kB 3.7 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 950.8/950.8 kB 14.5 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.2/5.2 MB 17.2 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 589.8/589.8 MB 2.5 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.3/5.3 MB 81.6 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.2/2.2 MB 95.6 MB/s eta 0:00:00 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.5/5.5 MB 85.5 MB/s eta 0:00:00 ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. tf-keras 2.15.1 requires tensorflow<2.16,>=2.15, but you have tensorflow 2.16.1 which is incompatible.
Exploring prompting capabilities
CodeGemma
CodeGemma models are text-to-text and text-to-code decoder-only models that specializes in code completion and code generation tasks. The CodeGemma 2B and 7B variants are specially tuned for code infilling tasks.
This example uses CodeGemma's fill-in-the-middle (FIM) capability to complete code based on the surrounding context. This is particularly useful in code editor applications for inserting code where the text cursor is based on the code around it (before and after the cursor).
CodeGemma lets you use 4 user-defined tokens - 3 (<|fim_prefix|>, <|fim_suffix|>, <|fim_middle|>) for FIM and a <|file_separator|> token for multi-file context support. Use these to define constants. More about that in the next cells.
Downloading from https://www.kaggle.com/api/v1/models/keras/codegemma/keras/code_gemma_1.1_2b_en/1/download/metadata.json...
100%|██████████| 143/143 [00:00<00:00, 126kB/s]
Downloading from https://www.kaggle.com/api/v1/models/keras/codegemma/keras/code_gemma_1.1_2b_en/1/download/config.json...
100%|██████████| 501/501 [00:00<00:00, 428kB/s]
Downloading from https://www.kaggle.com/api/v1/models/keras/codegemma/keras/code_gemma_1.1_2b_en/1/download/model.weights.h5...
100%|██████████| 4.67G/4.67G [02:25<00:00, 34.4MB/s]
Downloading from https://www.kaggle.com/api/v1/models/keras/codegemma/keras/code_gemma_1.1_2b_en/1/download/tokenizer.json...
100%|██████████| 315/315 [00:00<00:00, 413kB/s]
Downloading from https://www.kaggle.com/api/v1/models/keras/codegemma/keras/code_gemma_1.1_2b_en/1/download/assets/tokenizer/vocabulary.spm...
100%|██████████| 4.04M/4.04M [00:00<00:00, 7.16MB/s]
Prompting example: Code infilling
--- RAW Response ---
<|fim_prefix|>def calculate_area_of_rectangle(a: int, b: int) -> int:<|fim_suffix|>
return area<|fim_middle|>
"""
Calculate the area of a rectangle.
Args:
a (int): The length of the rectangle.
b (int): The width of the rectangle.
Returns:
int: The area of the rectangle.
"""
area = a * b<|file_separator|>
--- The generated (FIM) piece of code: ---
"""
Calculate the area of a rectangle.
Args:
a (int): The length of the rectangle.
b (int): The width of the rectangle.
Returns:
int: The area of the rectangle.
"""
area = a * b
--- The whole function: ---
def calculate_area_of_rectangle(a: int, b: int) -> int:
"""
Calculate the area of a rectangle.
Args:
a (int): The length of the rectangle.
b (int): The width of the rectangle.
Returns:
int: The area of the rectangle.
"""
area = a * b
return area
Prompting example: Code Generation
Note: While the 2B version of CodeGemma was primarily designed for code completion, it can also handle basic code generation tasks. However, for optimal code generation results, we recommend using the instruction-tuned 7B model.
Write a one-liner in Python that check if a year is a leap year.
Examples:
>>> is_a_leap_year(2016)
True
>>> is_a_leap_year(2001)
False
>>> is_a_leap_year(2052)
True
def is_a_leap_year(year: int) -> bool:
return year % 4 == 0 and (year % 100 != 0 or year % 400 == 0)
<|file_separator|>
{'status': 'ok', 'restart': True}
Run in Google Colab