openai/openai-python
Publicmirrored fromhttps://github.com/openai/openai-pythonAvailable
README.md
199lines · modecode
| 1 | # OpenAI Python Library |
| 2 | |
| 3 | The OpenAI Python library provides convenient access to the OpenAI API |
| 4 | from applications written in the Python language. It includes a |
| 5 | pre-defined set of classes for API resources that initialize |
| 6 | themselves dynamically from API responses which makes it compatible |
| 7 | with a wide range of versions of the OpenAI API. |
| 8 | |
| 9 | ## Documentation |
| 10 | |
| 11 | See the [OpenAI API docs](https://beta.openai.com/docs/api-reference?lang=python). |
| 12 | |
| 13 | ## Installation |
| 14 | |
| 15 | You don't need this source code unless you want to modify the package. If you just |
| 16 | want to use the package, just run: |
| 17 | |
| 18 | ```sh |
| 19 | pip install --upgrade openai |
| 20 | ``` |
| 21 | |
| 22 | Install from source with: |
| 23 | |
| 24 | ```sh |
| 25 | python setup.py install |
| 26 | ``` |
| 27 | |
| 28 | ## Usage |
| 29 | |
| 30 | The library needs to be configured with your account's secret key which is available on the [website](https://beta.openai.com/account/api-keys). Either set it as the `OPENAI_API_KEY` environment variable before using the library: |
| 31 | |
| 32 | ```bash |
| 33 | export OPENAI_API_KEY='sk-...' |
| 34 | ``` |
| 35 | |
| 36 | Or set `openai.api_key` to its value: |
| 37 | |
| 38 | ```python |
| 39 | import openai |
| 40 | openai.api_key = "sk-..." |
| 41 | |
| 42 | # list engines |
| 43 | engines = openai.Engine.list() |
| 44 | |
| 45 | # print the first engine's id |
| 46 | print(engines.data[0].id) |
| 47 | |
| 48 | # create a completion |
| 49 | completion = openai.Completion.create(engine="ada", prompt="Hello world") |
| 50 | |
| 51 | # print the completion |
| 52 | print(completion.choices[0].text) |
| 53 | ``` |
| 54 | |
| 55 | ### Microsoft Azure Endpoints |
| 56 | |
| 57 | In order to use the library with Microsoft Azure endpoints, you need to set the api_type, api_base and api_version in addition to the api_key. The api_type must be set to 'azure' and the others correspond to the properties of your endpoint. |
| 58 | In addition, the deployment name must be passed as the engine parameter. |
| 59 | |
| 60 | ```python |
| 61 | import openai |
| 62 | openai.api_type = "azure" |
| 63 | openai.api_key = "..." |
| 64 | openai.api_base = "https://example-endpoint.openai.azure.com" |
| 65 | openai.api_version = "2021-11-01-preview" |
| 66 | |
| 67 | # create a completion |
| 68 | completion = openai.Completion.create(engine="deployment-name", prompt="Hello world") |
| 69 | |
| 70 | # print the completion |
| 71 | print(completion.choices[0].text) |
| 72 | |
| 73 | # create a search and pass the deployment-name as the engine Id. |
| 74 | search = openai.Engine(id="deployment-name").search(documents=["White House", "hospital", "school"], query ="the president") |
| 75 | |
| 76 | # print the search |
| 77 | print(search) |
| 78 | ``` |
| 79 | |
| 80 | Please note that for the moment, the Microsoft Azure endpoints can only be used for completion, search and fine-tuning operations. |
| 81 | For a detailed example on how to use fine-tuning and other operations using Azure endpoints, please check out the following Jupyter notebooks: |
| 82 | * [Using Azure fine-tuning](https://github.com/openai/openai-cookbook/tree/main/examples/azure/finetuning.ipynb) |
| 83 | * [Using Azure embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/azure/embeddings.ipynb) |
| 84 | |
| 85 | ### Microsoft Azure Active Directory Authentication |
| 86 | |
| 87 | In order to use Microsoft Active Directory to authenticate to your Azure endpoint, you need to set the api_type to "azure_ad" and pass the acquired credential token to api_key. The rest of the parameters need to be set as specified in the previous section. |
| 88 | |
| 89 | |
| 90 | ```python |
| 91 | from azure.identity import DefaultAzureCredential |
| 92 | import openai |
| 93 | |
| 94 | # Request credential |
| 95 | default_credential = DefaultAzureCredential() |
| 96 | token = default_credential.get_token("https://cognitiveservices.azure.com") |
| 97 | |
| 98 | # Setup parameters |
| 99 | openai.api_type = "azure_ad" |
| 100 | openai.api_key = token.token |
| 101 | openai.api_base = "https://example-endpoint.openai.azure.com/" |
| 102 | openai.api_version = "2022-03-01-preview" |
| 103 | |
| 104 | # ... |
| 105 | ``` |
| 106 | ### Command-line interface |
| 107 | |
| 108 | This library additionally provides an `openai` command-line utility |
| 109 | which makes it easy to interact with the API from your terminal. Run |
| 110 | `openai api -h` for usage. |
| 111 | |
| 112 | ```sh |
| 113 | # list engines |
| 114 | openai api engines.list |
| 115 | |
| 116 | # create a completion |
| 117 | openai api completions.create -e ada -p "Hello world" |
| 118 | ``` |
| 119 | |
| 120 | ## Example code |
| 121 | |
| 122 | Examples of how to use this Python library to accomplish various tasks can be found in the [OpenAI Cookbook](https://github.com/openai/openai-cookbook/). It contains code examples for: |
| 123 | |
| 124 | * Classification using fine-tuning |
| 125 | * Clustering |
| 126 | * Code search |
| 127 | * Customizing embeddings |
| 128 | * Question answering from a corpus of documents |
| 129 | * Recommendations |
| 130 | * Visualization of embeddings |
| 131 | * And more |
| 132 | |
| 133 | Prior to July 2022, this OpenAI Python library hosted code examples in its examples folder, but since then all examples have been migrated to the [OpenAI Cookbook](https://github.com/openai/openai-cookbook/). |
| 134 | |
| 135 | ### Embeddings |
| 136 | |
| 137 | In the OpenAI Python library, an embedding represents a text string as a fixed-length vector of floating point numbers. Embeddings are designed to measure the similarity or relevance between text strings. |
| 138 | |
| 139 | To get an embedding for a text string, you can use the embeddings method as follows in Python: |
| 140 | |
| 141 | ```python |
| 142 | import openai |
| 143 | openai.api_key = "sk-..." # supply your API key however you choose |
| 144 | |
| 145 | # choose text to embed |
| 146 | text_string = "sample text" |
| 147 | |
| 148 | # choose an embedding |
| 149 | model_id = "text-similarity-davinci-001" |
| 150 | |
| 151 | # compute the embedding of the text |
| 152 | embedding = openai.Embedding.create(input=text_string, engine=model_id)['data'][0]['embedding'] |
| 153 | ``` |
| 154 | |
| 155 | An example of how to call the embeddings method is shown in this [get embeddings notebook](https://github.com/openai/openai-cookbook/blob/main/examples/Get_embeddings.ipynb). |
| 156 | |
| 157 | Examples of how to use embeddings are shared in the following Jupyter notebooks: |
| 158 | |
| 159 | - [Classification using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Classification_using_embeddings.ipynb) |
| 160 | - [Clustering using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Clustering.ipynb) |
| 161 | - [Code search using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Code_search.ipynb) |
| 162 | - [Semantic text search using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Semantic_text_search_using_embeddings.ipynb) |
| 163 | - [User and product embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/User_and_product_embeddings.ipynb) |
| 164 | - [Zero-shot classification using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Zero-shot_classification_with_embeddings.ipynb) |
| 165 | - [Recommendation using embeddings](https://github.com/openai/openai-cookbook/blob/main/examples/Recommendation_using_embeddings.ipynb) |
| 166 | |
| 167 | For more information on embeddings and the types of embeddings OpenAI offers, read the [embeddings guide](https://beta.openai.com/docs/guides/embeddings) in the OpenAI documentation. |
| 168 | |
| 169 | ### Fine tuning |
| 170 | |
| 171 | Fine tuning a model on training data can both improve the results (by giving the model more examples to learn from) and reduce the cost & latency of API calls (by reducing the need to include training examples in prompts). |
| 172 | |
| 173 | Examples of fine tuning are shared in the following Jupyter notebooks: |
| 174 | |
| 175 | - [Classification with fine tuning](https://github.com/openai/openai-cookbook/blob/main/examples/Fine-tuned_classification.ipynb) (a simple notebook that shows the steps required for fine tuning) |
| 176 | - Fine tuning a model that answers questions about the 2020 Olympics |
| 177 | - [Step 1: Collecting data](https://github.com/openai/openai-cookbook/blob/main/examples/fine-tuned_qa/olympics-1-collect-data.ipynb) |
| 178 | - [Step 2: Creating a synthetic Q&A dataset](https://github.com/openai/openai-cookbook/blob/main/examples/fine-tuned_qa/olympics-2-create-qa.ipynb) |
| 179 | - [Step 3: Train a fine-tuning model specialized for Q&A](https://github.com/openai/openai-cookbook/blob/main/examples/fine-tuned_qa/olympics-3-train-qa.ipynb) |
| 180 | |
| 181 | Sync your fine-tunes to [Weights & Biases](https://wandb.me/openai-docs) to track experiments, models, and datasets in your central dashboard with: |
| 182 | |
| 183 | ```bash |
| 184 | openai wandb sync |
| 185 | ``` |
| 186 | |
| 187 | For more information on fine tuning, read the [fine-tuning guide](https://beta.openai.com/docs/guides/fine-tuning) in the OpenAI documentation. |
| 188 | |
| 189 | ## Requirements |
| 190 | |
| 191 | - Python 3.7.1+ |
| 192 | |
| 193 | In general we want to support the versions of Python that our |
| 194 | customers are using, so if you run into issues with any version |
| 195 | issues, please let us know at support@openai.com. |
| 196 | |
| 197 | ## Credit |
| 198 | |
| 199 | This library is forked from the [Stripe Python Library](https://github.com/stripe/stripe-python). |
| 200 | |