mirror of https://github.com/microsoft/semantic-kernel.git synced 2026-10-12 02:19:06 +00:00

Files

T

Jordan Henkel 407fb36e59 Python: cleanup/sync/fixes (for examples, docs, tests, notebooks) (#317 )

### Motivation and Context

The `python-preview` branch is under active development and many
docs/tests/examples are getting a bit out of sync. There are also a few
small cleanup chores/refactors to make _usage_ easier (e.g., to make
examples cleaner). I've grouped these small fixes (plus a few bug fixes)
into this "cleanup" PR.

### Description

This PR does the following:

1. Updates the out-of-date example in `README.md`
2. Fixes all 5 of the notebooks we support right now (1,2,3,4, and 6;
notebook 5 is planner, I've blanked out most of the code there so we
don't confuse people, especially as the planner API is changing)
3. Fixes a bug w/ `stop_sequences` in `(Azure)OpenAITextCompletion`
4. Fixes a bug introduced in our upgrades to the embeddings cosine-sim
check (we should add more thorough tests here)
5. Cleans up tests
6. Ensures end-to-end tests are all runnable 
7. Fixes up the `kernel_extensions` to work more naturally (so that end
users can use them directly on the `kernel` object and get good
type-hinting)

2023-04-13 16:12:36 -07:00

4.8 KiB

Raw Blame History

Basic Loading of the Kernel

To run the notebooks we recommend using Poetry and starting a shell with a virtual environment prepared to use SK.

See DEV_SETUP.md for more information.

In [ ]:

!python -m pip install ../../../python

In [ ]:

import semantic_kernel as sk
import semantic_kernel.ai.open_ai as sk_oai

You can instantiate the kernel in a few ways, depending on your use case.

In [ ]:

# Simple instance
kernel_1 = sk.create_kernel()

In [ ]:

# Instance with a custom logger
my_logger = sk.NullLogger()
kernel_2 = (
    sk.kernel_builder()
    .with_logger(my_logger)
    .build()
)

In [ ]:

# Instance with a custom configuration
my_config = sk.KernelConfig()
kernel_2 = (
    sk.kernel_builder()
    .with_configuration(my_config)
    .build()
)

When using the kernel for AI requests, the kernel needs some settings like URL and credentials to the AI models.

The SDK currently supports OpenAI and Azure OpenAI, other services will be added over time.

If you need an Azure OpenAI key, go here.

In [ ]:

kernel = sk.create_kernel()

kernel.config.add_text_backend(               # We are adding a text backend
    "Azure_curie",                            # The alias we can use in prompt templates' config.json
    sk_oai.AzureTextCompletion(
        "my-finetuned-Curie",                 # Azure OpenAI *Deployment ID*
        "https://contoso.openai.azure.com/",  # Azure OpenAI *Endpoint*
        "...your Azure OpenAI Key..."         # Azure OpenAI *Key*
    )
)

kernel.config.add_text_backend(               # We are adding a text backend
    "OpenAI_davinci",                         # The alias we can use in prompt templates' config.json
    sk_oai.OpenAITextCompletion(
        "text-davinci-003",                   # OpenAI Model Name
        "...your OpenAI API Key...",          # OpenAI API key
        "...your OpenAI Org ID..."            # *optional* OpenAI Organization ID
    )
)

When working with multiple backends and multiple models, the first backend defined is also the "default" used in these scenarios:

a prompt configuration doesn't specify which AI backend to use
a prompt configuration requires a backend unknown to the kernel

The default can be set and changed programmatically:

In [ ]:

kernel.config.set_default_text_backend("Azure_curie")

Great, now that you're familiar with setting up the Semantic Kernel, let's see how we can use it to run prompts.

4.8 KiB Raw Blame History

Basic Loading of the Kernel

4.8 KiB

Raw Blame History