File size: 3,454 Bytes
a8b3f00 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
import os
from collections.abc import Generator
import pytest
from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
from core.model_runtime.entities.message_entities import AssistantPromptMessage, SystemPromptMessage, UserPromptMessage
from core.model_runtime.errors.validate import CredentialsValidateFailedError
from core.model_runtime.model_providers.bedrock.llm.llm import BedrockLargeLanguageModel
def test_validate_credentials():
model = BedrockLargeLanguageModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(model="meta.llama2-13b-chat-v1", credentials={"anthropic_api_key": "invalid_key"})
model.validate_credentials(
model="meta.llama2-13b-chat-v1",
credentials={
"aws_region": os.getenv("AWS_REGION"),
"aws_access_key": os.getenv("AWS_ACCESS_KEY"),
"aws_secret_access_key": os.getenv("AWS_SECRET_ACCESS_KEY"),
},
)
def test_invoke_model():
model = BedrockLargeLanguageModel()
response = model.invoke(
model="meta.llama2-13b-chat-v1",
credentials={
"aws_region": os.getenv("AWS_REGION"),
"aws_access_key": os.getenv("AWS_ACCESS_KEY"),
"aws_secret_access_key": os.getenv("AWS_SECRET_ACCESS_KEY"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
model_parameters={"temperature": 0.0, "top_p": 1.0, "max_tokens_to_sample": 10},
stop=["How"],
stream=False,
user="abc-123",
)
assert isinstance(response, LLMResult)
assert len(response.message.content) > 0
def test_invoke_stream_model():
model = BedrockLargeLanguageModel()
response = model.invoke(
model="meta.llama2-13b-chat-v1",
credentials={
"aws_region": os.getenv("AWS_REGION"),
"aws_access_key": os.getenv("AWS_ACCESS_KEY"),
"aws_secret_access_key": os.getenv("AWS_SECRET_ACCESS_KEY"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
model_parameters={"temperature": 0.0, "max_tokens_to_sample": 100},
stream=True,
user="abc-123",
)
assert isinstance(response, Generator)
for chunk in response:
print(chunk)
assert isinstance(chunk, LLMResultChunk)
assert isinstance(chunk.delta, LLMResultChunkDelta)
assert isinstance(chunk.delta.message, AssistantPromptMessage)
assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
def test_get_num_tokens():
model = BedrockLargeLanguageModel()
num_tokens = model.get_num_tokens(
model="meta.llama2-13b-chat-v1",
credentials={
"aws_region": os.getenv("AWS_REGION"),
"aws_access_key": os.getenv("AWS_ACCESS_KEY"),
"aws_secret_access_key": os.getenv("AWS_SECRET_ACCESS_KEY"),
},
messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
)
assert num_tokens == 18
|