estel/dify
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Model Runtime (#1858)

Browse Source 
Co-authored-by: StyleZhang <jasonapring2015@outlook.com>
Co-authored-by: Garfield Dai <dai.hai@foxmail.com>
Co-authored-by: chenhe <guchenhe@gmail.com>
Co-authored-by: jyong <jyong@dify.ai>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: Yeuoly <admin@srmxy.cn>
This commit is contained in:
takatost
2024-01-02 23:42:00 +08:00
committed by GitHub
parent e91dd28a76
commit d069c668f8
807 changed files with 171310 additions and 23806 deletions
Download Patch File Download Diff File Expand all files Collapse all files

0
api/tests/integration_tests/model_runtime/replicate/__init__.py Normal file
View File

119
api/tests/integration_tests/model_runtime/replicate/test_llm.py Normal file
View File

@@ -0,0 +1,119 @@
import os
from typing import Generator
import pytest
from core.model_runtime.entities.message_entities import SystemPromptMessage, UserPromptMessage, AssistantPromptMessage
from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, \
LLMResultChunkDelta
from core.model_runtime.errors.validate import CredentialsValidateFailedError
from core.model_runtime.model_providers.replicate.llm.llm import ReplicateLargeLanguageModel
def test_validate_credentials():
model = ReplicateLargeLanguageModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(
model='meta/llama-2-13b-chat',
credentials={
'replicate_api_token': 'invalid_key',
'model_version': 'f4e2de70d66816a838a89eeeb621910adffb0dd0baba3976c96980970978018d'
}
)
model.validate_credentials(
model='meta/llama-2-13b-chat',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'f4e2de70d66816a838a89eeeb621910adffb0dd0baba3976c96980970978018d'
}
)
def test_invoke_model():
model = ReplicateLargeLanguageModel()
response = model.invoke(
model='meta/llama-2-13b-chat',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'f4e2de70d66816a838a89eeeb621910adffb0dd0baba3976c96980970978018d'
},
prompt_messages=[
SystemPromptMessage(
content='You are a helpful AI assistant.',
),
UserPromptMessage(
content='Who are you?'
)
],
model_parameters={
'temperature': 1.0,
'top_k': 2,
'top_p': 0.5,
},
stop=['How'],
stream=False,
user="abc-123"
)
assert isinstance(response, LLMResult)
assert len(response.message.content) > 0
def test_invoke_stream_model():
model = ReplicateLargeLanguageModel()
response = model.invoke(
model='mistralai/mixtral-8x7b-instruct-v0.1',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': '2b56576fcfbe32fa0526897d8385dd3fb3d36ba6fd0dbe033c72886b81ade93e'
},
prompt_messages=[
SystemPromptMessage(
content='You are a helpful AI assistant.',
),
UserPromptMessage(
content='Who are you?'
)
],
model_parameters={
'temperature': 1.0,
'top_k': 2,
'top_p': 0.5,
},
stop=['How'],
stream=True,
user="abc-123"
)
assert isinstance(response, Generator)
for chunk in response:
assert isinstance(chunk, LLMResultChunk)
assert isinstance(chunk.delta, LLMResultChunkDelta)
assert isinstance(chunk.delta.message, AssistantPromptMessage)
def test_get_num_tokens():
model = ReplicateLargeLanguageModel()
num_tokens = model.get_num_tokens(
model='',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': '2b56576fcfbe32fa0526897d8385dd3fb3d36ba6fd0dbe033c72886b81ade93e'
},
prompt_messages=[
SystemPromptMessage(
content='You are a helpful AI assistant.',
),
UserPromptMessage(
content='Hello World!'
)
]
)
assert num_tokens == 14

151
api/tests/integration_tests/model_runtime/replicate/test_text_embedding.py Normal file
View File

@@ -0,0 +1,151 @@
import os
import pytest
from core.model_runtime.entities.text_embedding_entities import TextEmbeddingResult
from core.model_runtime.errors.validate import CredentialsValidateFailedError
from core.model_runtime.model_providers.replicate.text_embedding.text_embedding import ReplicateEmbeddingModel
def test_validate_credentials_one():
model = ReplicateEmbeddingModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(
model='replicate/all-mpnet-base-v2',
credentials={
'replicate_api_token': 'invalid_key',
'model_version': 'b6b7585c9640cd7a9572c6e129c9549d79c9c31f0d3fdce7baac7c67ca38f305'
}
)
model.validate_credentials(
model='replicate/all-mpnet-base-v2',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'b6b7585c9640cd7a9572c6e129c9549d79c9c31f0d3fdce7baac7c67ca38f305'
}
)
def test_validate_credentials_two():
model = ReplicateEmbeddingModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(
model='nateraw/bge-large-en-v1.5',
credentials={
'replicate_api_token': 'invalid_key',
'model_version': '9cf9f015a9cb9c61d1a2610659cdac4a4ca222f2d3707a68517b18c198a9add1'
}
)
model.validate_credentials(
model='nateraw/bge-large-en-v1.5',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': '9cf9f015a9cb9c61d1a2610659cdac4a4ca222f2d3707a68517b18c198a9add1'
}
)
def test_invoke_model_one():
model = ReplicateEmbeddingModel()
result = model.invoke(
model='nateraw/bge-large-en-v1.5',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': '9cf9f015a9cb9c61d1a2610659cdac4a4ca222f2d3707a68517b18c198a9add1'
},
texts=[
"hello",
"world"
],
user="abc-123"
)
assert isinstance(result, TextEmbeddingResult)
assert len(result.embeddings) == 2
assert result.usage.total_tokens == 2
def test_invoke_model_two():
model = ReplicateEmbeddingModel()
result = model.invoke(
model='andreasjansson/clip-features',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': '75b33f253f7714a281ad3e9b28f63e3232d583716ef6718f2e46641077ea040a'
},
texts=[
"hello",
"world"
],
user="abc-123"
)
assert isinstance(result, TextEmbeddingResult)
assert len(result.embeddings) == 2
assert result.usage.total_tokens == 2
def test_invoke_model_three():
model = ReplicateEmbeddingModel()
result = model.invoke(
model='replicate/all-mpnet-base-v2',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'b6b7585c9640cd7a9572c6e129c9549d79c9c31f0d3fdce7baac7c67ca38f305'
},
texts=[
"hello",
"world"
],
user="abc-123"
)
assert isinstance(result, TextEmbeddingResult)
assert len(result.embeddings) == 2
assert result.usage.total_tokens == 2
def test_invoke_model_four():
model = ReplicateEmbeddingModel()
result = model.invoke(
model='nateraw/jina-embeddings-v2-base-en',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'f8367a1c072ba2bc28af549d1faeacfe9b88b3f0e475add7a75091dac507f79e'
},
texts=[
"hello",
"world"
],
user="abc-123"
)
assert isinstance(result, TextEmbeddingResult)
assert len(result.embeddings) == 2
assert result.usage.total_tokens == 2
def test_get_num_tokens():
model = ReplicateEmbeddingModel()
num_tokens = model.get_num_tokens(
model='nateraw/jina-embeddings-v2-base-en',
credentials={
'replicate_api_token': os.environ.get('REPLICATE_API_KEY'),
'model_version': 'f8367a1c072ba2bc28af549d1faeacfe9b88b3f0e475add7a75091dac507f79e'
},
texts=[
"hello",
"world"
]
)
assert num_tokens == 2