python client to interact with Aleph Alpha api endpoints
Project description
Aleph Alpha Client
Interact with the Aleph Alpha API via Python
Installation
The latest stable version is deployed to PyPi so you can install this package via pip.
pip install aleph-alpha-client
Usage
Completion Multimodal
from aleph_alpha_client import ImagePrompt, AlephAlphaModel, AlephAlphaClient, CompletionRequest, Prompt
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
model_name = "luminous-extended"
)
# You need to choose a model with multimodal capabilities for this example.
url = "https://cdn-images-1.medium.com/max/1200/1*HunNdlTmoPj8EKpl-jqvBA.png"
image = ImagePrompt.from_url(url)
prompt = Prompt([
image,
"Q: What does the picture show? A:",
])
request = CompletionRequest(prompt=prompt, maximum_tokens=20)
result = model.complete(request)
print(result.completions[0].completion)
Evaluation text prompt
from aleph_alpha_client import AlephAlphaClient, AlephAlphaModel, EvaluationRequest, Prompt
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
model_name = "luminous-extended"
)
request = EvaluationRequest(prompt=Prompt.from_text("The api works"), completion_expected=" well")
result = model.evaluate(request)
print(result)
Evaluation Multimodal
from aleph_alpha_client import ImagePrompt, AlephAlphaClient, AlephAlphaModel, EvaluationRequest, Prompt
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with multimodal capabilities for this example.
model_name = "luminous-extended"
)
url = "https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/2008-09-24_Blockbuster_in_Durham.jpg/330px-2008-09-24_Blockbuster_in_Durham.jpg"
image = ImagePrompt.from_url(url)
prompt = Prompt([
image,
"Q: What is the name of the store?\nA:",
])
request = EvaluationRequest(prompt=prompt, completion_expected=" Blockbuster Video")
result = model.evaluate(request)
print(result)
Embedding text prompt
from aleph_alpha_client import AlephAlphaModel, AlephAlphaClient, EmbeddingRequest, Prompt
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
model_name = "luminous-extended"
)
request = EmbeddingRequest(prompt=Prompt.from_text("This is an example."), layers=[-1], pooling=["mean"])
result = model.embed(request)
print(result)
Embedding multimodal prompt
from aleph_alpha_client import ImagePrompt, AlephAlphaClient, AlephAlphaModel, EmbeddingRequest, Prompt
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with multimodal capabilities for this example.
model_name = "luminous-extended"
)
url = "https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/2008-09-24_Blockbuster_in_Durham.jpg/330px-2008-09-24_Blockbuster_in_Durham.jpg"
image = ImagePrompt.from_url(url)
prompt = Prompt([
image,
"Q: What is the name of the store?\nA:",
])
request = EmbeddingRequest(prompt=prompt, layers=[-1], pooling=["mean"])
result = model.embed(request)
print(result)
Semantic embedding
Symmetric
from typing import Sequence
from aleph_alpha_client import ImagePrompt, AlephAlphaClient, AlephAlphaModel, SemanticEmbeddingRequest, SemanticRepresentation, Prompt
import math
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with multimodal capabilities for this example.
model_name = "luminous-base"
)
# Texts to compare
texts = [
"deep learning",
"artificial intelligence",
"deep diving",
"artificial snow",
]
embeddings = []
for text in texts:
request = SemanticEmbeddingRequest(prompt=Prompt.from_text(text), representation=SemanticRepresentation.Symmetric)
result = model.semantic_embed(request)
embeddings.append(result.embedding)
# Calculate cosine similarities. Can use numpy or scipy or another library to do this
def cosine_similarity(v1: Sequence[float], v2: Sequence[float]) -> float:
"compute cosine similarity of v1 to v2: (v1 dot v2)/{||v1||*||v2||)"
sumxx, sumxy, sumyy = 0, 0, 0
for i in range(len(v1)):
x = v1[i]; y = v2[i]
sumxx += x*x
sumyy += y*y
sumxy += x*y
return sumxy/math.sqrt(sumxx*sumyy)
# Cosine similarities are in [-1, 1]. Higher means more similar
print("Cosine similarity between \"%s\" and \"%s\" is: %.3f" % (texts[0], texts[1], cosine_similarity(embeddings[0], embeddings[1])))
print("Cosine similarity between \"%s\" and \"%s\" is: %.3f" % (texts[0], texts[2], cosine_similarity(embeddings[0], embeddings[2])))
print("Cosine similarity between \"%s\" and \"%s\" is: %.3f" % (texts[0], texts[3], cosine_similarity(embeddings[0], embeddings[3])))
Documents and Query
from typing import Sequence
from aleph_alpha_client import ImagePrompt, AlephAlphaClient, AlephAlphaModel, SemanticEmbeddingRequest, SemanticRepresentation, Prompt
import math
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with multimodal capabilities for this example.
model_name = "luminous-base"
)
# Documents to search in
documents = [
# AI wikipedia article
"Artificial intelligence (AI) is intelligence demonstrated by machines, as opposed to the natural intelligence displayed by animals including humans. AI research has been defined as the field of study of intelligent agents, which refers to any system that perceives its environment and takes actions that maximize its chance of achieving its goals.",
# Deep Learning Wikipedia article
"Deep learning (also known as deep structured learning) is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised or unsupervised.",
# Deep Diving Wikipedia article
"Deep diving is underwater diving to a depth beyond the norm accepted by the associated community. In some cases this is a prescribed limit established by an authority, while in others it is associated with a level of certification or training, and it may vary depending on whether the diving is recreational, technical or commercial. Nitrogen narcosis becomes a hazard below 30 metres (98 ft) and hypoxic breathing gas is required below 60 metres (200 ft) to lessen the risk of oxygen toxicity.",
]
# Keyword to search documents with
query = "artificial intelligence"
# Embed Query
request = SemanticEmbeddingRequest(prompt=Prompt.from_text(query), representation=SemanticRepresentation.Query)
result = model.semantic_embed(request)
query_embedding = result.embedding
# Embed documents
document_embeddings = []
for document in documents:
request = SemanticEmbeddingRequest(prompt=Prompt.from_text(document), representation=SemanticRepresentation.Document)
result = model.semantic_embed(request)
document_embeddings.append(result.embedding)
# Calculate cosine similarities. Can use numpy or scipy or another library to do this
def cosine_similarity(v1: Sequence[float], v2: Sequence[float]) -> float:
"compute cosine similarity of v1 to v2: (v1 dot v2)/{||v1||*||v2||)"
sumxx, sumxy, sumyy = 0, 0, 0
for i in range(len(v1)):
x = v1[i]; y = v2[i]
sumxx += x*x
sumyy += y*y
sumxy += x*y
return sumxy/math.sqrt(sumxx*sumyy)
# Cosine similarities are in [-1, 1]. Higher means more similar
print("Cosine similarity between \"%s\" and \"%s...\" is: %.3f" % (query, documents[0][:10], cosine_similarity(query_embedding, document_embeddings[0])))
print("Cosine similarity between \"%s\" and \"%s...\" is: %.3f" % (query, documents[1][:10], cosine_similarity(query_embedding, document_embeddings[1])))
print("Cosine similarity between \"%s\" and \"%s...\" is: %.3f" % (query, documents[2][:10], cosine_similarity(query_embedding, document_embeddings[2])))
Q&A with a Docx Document
from aleph_alpha_client import Document, AlephAlphaClient, AlephAlphaModel, QaRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
docx_file = "./tests/sample.docx"
document = Document.from_docx_file(docx_file)
request = QaRequest(
query = "What is a computer program?",
documents = [document]
)
result = model.qa(request)
print(result)
Q&A with a Text
from aleph_alpha_client import AlephAlphaClient, AlephAlphaModel, QaRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
prompt = "In imperative programming, a computer program is a sequence of instructions in a programming language that a computer can execute or interpret."
document = Document.from_text(prompt)
request = QaRequest(
query = "What is a computer program?",
documents = [document],
)
result = model.qa(request)
print(result)
Q&A with a multimodal prompt
from aleph_alpha_client import Document, ImagePrompt, AlephAlphaClient, AlephAlphaModel, QaRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
url = "https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/2008-09-24_Blockbuster_in_Durham.jpg/330px-2008-09-24_Blockbuster_in_Durham.jpg"
image = ImagePrompt.from_url(url)
prompt = [image]
document = Document.from_prompt(prompt)
request = QaRequest (
query = "What is the name of the store?",
documents = [document]
)
result = model.qa(request)
print(result)
Summary with a Docx Document
from aleph_alpha_client import Document, AlephAlphaClient, AlephAlphaModel, SummarizationRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
docx_file = "./tests/sample.docx"
document = Document.from_docx_file(docx_file)
request = SummarizationRequest(document)
result = model.summarize(request)
print(result.summary)
Summary with a Text
from aleph_alpha_client import AlephAlphaClient, AlephAlphaModel, SummarizationRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
# https://en.wikipedia.org/wiki/Imperative_programming
prompt = """In computer science, imperative programming is a programming paradigm of software that uses statements that change a program's state. In much the same way that the imperative mood in natural languages expresses commands, an imperative program consists of commands for the computer to perform. Imperative programming focuses on describing how a program operates step by step, rather than on high-level descriptions of its expected results.
The term is often used in contrast to declarative programming, which focuses on what the program should accomplish without specifying all the details of how the program should achieve the result.[1]
Imperative and procedural programming
Procedural programming is a type of imperative programming in which the program is built from one or more procedures (also termed subroutines or functions). The terms are often used as synonyms, but the use of procedures has a dramatic effect on how imperative programs appear and how they are constructed. Heavy procedural programming, in which state changes are localized to procedures or restricted to explicit arguments and returns from procedures, is a form of structured programming. From the 1960s onwards, structured programming and modular programming in general have been promoted as techniques to improve the maintainability and overall quality of imperative programs. The concepts behind object-oriented programming attempt to extend this approach.
Procedural programming could be considered a step toward declarative programming. A programmer can often tell, simply by looking at the names, arguments, and return types of procedures (and related comments), what a particular procedure is supposed to do, without necessarily looking at the details of how it achieves its result. At the same time, a complete program is still imperative since it fixes the statements to be executed and their order of execution to a large extent."""
document = Document.from_text(prompt)
request = SummarizationRequest(document)
result = model.summarize(request)
print(result.summary)
Summary with a multimodal prompt
from aleph_alpha_client import Document, ImagePrompt, AlephAlphaClient, AlephAlphaModel, SummarizationRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
# You need to choose a model with qa support for this example.
model_name = "luminous-extended"
)
url = "https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/2008-09-24_Blockbuster_in_Durham.jpg/330px-2008-09-24_Blockbuster_in_Durham.jpg"
image = ImagePrompt.from_url(url)
prompt = [image]
document = Document.from_prompt(prompt)
request = SummarizationRequest(document)
result = model.summarize(request)
print(result.summary)
Tokenize a text prompt
from aleph_alpha_client import AlephAlphaClient, AlephAlphaModel, TokenizationRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
model_name = "luminous-extended"
)
# You need to choose a model with qa support and multimodal capabilities for this example.
request = TokenizationRequest(prompt="This is an example.", tokens=True, token_ids=True)
response = model.tokenize(request)
print(response)
Detokenize a token IDs into text prompt
from aleph_alpha_client import AlephAlphaClient, AlephAlphaModel, DetokenizationRequest
import os
model = AlephAlphaModel(
AlephAlphaClient(host="https://api.aleph-alpha.com", token=os.getenv("AA_TOKEN")),
model_name = "luminous-extended"
)
# You need to choose a model with qa support and multimodal capabilities for this example.
request = DetokenizationRequest(token_ids=[1730, 387, 300, 4377, 17])
response = model.detokenize(request)
print(response)
Testing
Tests use pytests with (optional) coverage plugin. Install the locally cloned repo in editable mode with:
pip install -e .[test]
Tests make api calls that reduce your quota!
Run tests
Tests can be run using pytest. Make sure to create a .env
file with the following content:
# test settings
TEST_API_URL=https://test.api.aleph-alpha.com
TEST_MODEL=luminous-base
TEST_TOKEN=your_token
Instead of a token username and password can be used.
# test settings
TEST_API_URL=https://api.aleph-alpha.com
TEST_MODEL=luminous-base
TEST_USERNAME=your_username
TEST_PASSWORD=your_password
- A coverage report can be created using the optional arguments --cov-report and --cov (see pytest documentation)
- A subset of tests can be selected by pointing to the module within tests
# run all tests, output coverage report of aleph_alpha_client module in terminal
pytest --cov-report term --cov=aleph_alpha_client tests
pytest tests -v # start verbose
If an html coverage report has been created a simple http server can be run to serve static files.
python -m http.server --directory htmlcov 8000
Update README
Do not change the README.md directly as it is generated from readme.ipynb
To update the readme, do the following:
-
pip install -e .[dev]
-
Edit the notebook in your favorite jupyter editor and run all python cells to verify that the code examples still work.
-
To generate a new README.md first remove all output cells from the Jupyter notebook and then execute the command:
jupyter nbconvert --to markdown readme.ipynb --output README.md
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for aleph_alpha_client-2.4.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0215a6bf4172de9cd9b40a774d97a9ba5ebe35e28fa0a7773e975d1094040fc3 |
|
MD5 | 22fab43bc06f158201b8e4b9329dc7cb |
|
BLAKE2b-256 | d27ff1dc73565df2b31e1a82f70bd0bb08b0d1eb209bb8b0f96e69c17b3eb6cc |