camel-ai 0.2.18__py3-none-any.whl → 0.2.20__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of camel-ai might be problematic. Click here for more details.
- camel/__init__.py +1 -1
- camel/agents/chat_agent.py +29 -30
- camel/agents/knowledge_graph_agent.py +1 -5
- camel/agents/multi_hop_generator_agent.py +35 -3
- camel/agents/programmed_agent_instruction.py +73 -18
- camel/benchmarks/apibench.py +1 -5
- camel/benchmarks/nexus.py +1 -5
- camel/benchmarks/ragbench.py +2 -2
- camel/bots/telegram_bot.py +1 -5
- camel/configs/__init__.py +9 -0
- camel/configs/aiml_config.py +80 -0
- camel/configs/gemini_config.py +1 -1
- camel/configs/moonshot_config.py +63 -0
- camel/configs/sglang_config.py +4 -0
- camel/configs/siliconflow_config.py +91 -0
- camel/datagen/__init__.py +3 -1
- camel/datagen/self_improving_cot.py +821 -0
- camel/datagen/source2synth/__init__.py +31 -0
- camel/{synthetic_datagen → datagen}/source2synth/data_processor.py +194 -29
- camel/{synthetic_datagen → datagen}/source2synth/models.py +25 -0
- camel/{synthetic_datagen → datagen}/source2synth/user_data_processor_config.py +9 -8
- camel/datahubs/huggingface.py +3 -3
- camel/embeddings/__init__.py +2 -0
- camel/embeddings/jina_embedding.py +161 -0
- camel/messages/func_message.py +1 -1
- camel/models/__init__.py +4 -0
- camel/models/aiml_model.py +147 -0
- camel/models/deepseek_model.py +29 -11
- camel/models/groq_model.py +0 -2
- camel/models/model_factory.py +9 -0
- camel/models/moonshot_model.py +138 -0
- camel/models/openai_model.py +1 -9
- camel/models/siliconflow_model.py +142 -0
- camel/societies/workforce/role_playing_worker.py +2 -4
- camel/societies/workforce/single_agent_worker.py +1 -6
- camel/societies/workforce/workforce.py +3 -9
- camel/toolkits/__init__.py +4 -0
- camel/toolkits/reddit_toolkit.py +8 -38
- camel/toolkits/search_toolkit.py +17 -6
- camel/toolkits/semantic_scholar_toolkit.py +308 -0
- camel/toolkits/sympy_toolkit.py +778 -0
- camel/toolkits/whatsapp_toolkit.py +11 -32
- camel/types/enums.py +205 -16
- camel/types/unified_model_type.py +5 -0
- camel/utils/__init__.py +7 -2
- camel/utils/commons.py +198 -21
- camel/utils/deduplication.py +199 -0
- camel/utils/token_counting.py +1 -39
- {camel_ai-0.2.18.dist-info → camel_ai-0.2.20.dist-info}/METADATA +17 -12
- {camel_ai-0.2.18.dist-info → camel_ai-0.2.20.dist-info}/RECORD +53 -41
- /camel/datagen/{cotdatagen.py → cot_datagen.py} +0 -0
- {camel_ai-0.2.18.dist-info → camel_ai-0.2.20.dist-info}/LICENSE +0 -0
- {camel_ai-0.2.18.dist-info → camel_ai-0.2.20.dist-info}/WHEEL +0 -0
|
@@ -0,0 +1,63 @@
|
|
|
1
|
+
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
|
|
2
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
3
|
+
# you may not use this file except in compliance with the License.
|
|
4
|
+
# You may obtain a copy of the License at
|
|
5
|
+
#
|
|
6
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
|
7
|
+
#
|
|
8
|
+
# Unless required by applicable law or agreed to in writing, software
|
|
9
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
10
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
11
|
+
# See the License for the specific language governing permissions and
|
|
12
|
+
# limitations under the License.
|
|
13
|
+
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
|
|
14
|
+
|
|
15
|
+
from typing import List, Optional, Union
|
|
16
|
+
|
|
17
|
+
from camel.configs.base_config import BaseConfig
|
|
18
|
+
|
|
19
|
+
|
|
20
|
+
class MoonshotConfig(BaseConfig):
|
|
21
|
+
r"""Defines the parameters for generating chat completions using the
|
|
22
|
+
Moonshot API. You can refer to the following link for more details:
|
|
23
|
+
https://platform.moonshot.cn/docs/api-reference
|
|
24
|
+
|
|
25
|
+
Args:
|
|
26
|
+
temperature (float, optional): Controls randomness in the response.
|
|
27
|
+
Lower values make the output more focused and deterministic.
|
|
28
|
+
(default: :obj:`0.3`)
|
|
29
|
+
max_tokens (int, optional): The maximum number of tokens to generate.
|
|
30
|
+
(default: :obj:`None`)
|
|
31
|
+
stream (bool, optional): Whether to stream the response.
|
|
32
|
+
(default: :obj:`False`)
|
|
33
|
+
tools (list, optional): List of tools that the model can use for
|
|
34
|
+
function calling. Each tool should be a dictionary containing
|
|
35
|
+
type, function name, description, and parameters.
|
|
36
|
+
(default: :obj:`None`)
|
|
37
|
+
top_p (float, optional): Controls diversity via nucleus sampling.
|
|
38
|
+
(default: :obj:`1.0`)
|
|
39
|
+
n (int, optional): How many chat completion choices to generate for
|
|
40
|
+
each input message. (default: :obj:`1`)
|
|
41
|
+
presence_penalty (float, optional): Penalty for new tokens based on
|
|
42
|
+
whether they appear in the text so far.
|
|
43
|
+
(default: :obj:`0.0`)
|
|
44
|
+
frequency_penalty (float, optional): Penalty for new tokens based on
|
|
45
|
+
their frequency in the text so far.
|
|
46
|
+
(default: :obj:`0.0`)
|
|
47
|
+
stop (Optional[Union[str, List[str]]], optional): Up to 4 sequences
|
|
48
|
+
where the API will stop generating further tokens.
|
|
49
|
+
(default: :obj:`None`)
|
|
50
|
+
"""
|
|
51
|
+
|
|
52
|
+
temperature: float = 0.3
|
|
53
|
+
max_tokens: Optional[int] = None
|
|
54
|
+
stream: bool = False
|
|
55
|
+
tools: Optional[list] = None
|
|
56
|
+
top_p: float = 1.0
|
|
57
|
+
n: int = 1
|
|
58
|
+
presence_penalty: float = 0.0
|
|
59
|
+
frequency_penalty: float = 0.0
|
|
60
|
+
stop: Optional[Union[str, List[str]]] = None
|
|
61
|
+
|
|
62
|
+
|
|
63
|
+
MOONSHOT_API_PARAMS = {param for param in MoonshotConfig.model_fields.keys()}
|
camel/configs/sglang_config.py
CHANGED
|
@@ -56,6 +56,10 @@ class SGLangConfig(BaseConfig):
|
|
|
56
56
|
in the chat completion. The total length of input tokens and
|
|
57
57
|
generated tokens is limited by the model's context length.
|
|
58
58
|
(default: :obj:`None`)
|
|
59
|
+
tools (list[FunctionTool], optional): A list of tools the model may
|
|
60
|
+
call. Currently, only functions are supported as a tool. Use this
|
|
61
|
+
to provide a list of functions the model may generate JSON inputs
|
|
62
|
+
for. A max of 128 functions are supported.
|
|
59
63
|
"""
|
|
60
64
|
|
|
61
65
|
stop: Union[str, Sequence[str], NotGiven] = NOT_GIVEN
|
|
@@ -0,0 +1,91 @@
|
|
|
1
|
+
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
|
|
2
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
3
|
+
# you may not use this file except in compliance with the License.
|
|
4
|
+
# You may obtain a copy of the License at
|
|
5
|
+
#
|
|
6
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
|
7
|
+
#
|
|
8
|
+
# Unless required by applicable law or agreed to in writing, software
|
|
9
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
10
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
11
|
+
# See the License for the specific language governing permissions and
|
|
12
|
+
# limitations under the License.
|
|
13
|
+
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
|
|
14
|
+
from __future__ import annotations
|
|
15
|
+
|
|
16
|
+
from typing import Any, Sequence, Type, Union
|
|
17
|
+
|
|
18
|
+
from pydantic import BaseModel
|
|
19
|
+
|
|
20
|
+
from camel.configs.base_config import BaseConfig
|
|
21
|
+
from camel.types import NOT_GIVEN, NotGiven
|
|
22
|
+
|
|
23
|
+
|
|
24
|
+
class SiliconFlowConfig(BaseConfig):
|
|
25
|
+
r"""Defines the parameters for generating chat completions using the
|
|
26
|
+
SiliconFlow API.
|
|
27
|
+
|
|
28
|
+
Args:
|
|
29
|
+
temperature (float, optional): Determines the degree of randomness
|
|
30
|
+
in the response. (default: :obj:`0.7`)
|
|
31
|
+
top_p (float, optional): The top_p (nucleus) parameter is used to
|
|
32
|
+
dynamically adjust the number of choices for each predicted token
|
|
33
|
+
based on the cumulative probabilities. (default: :obj:`0.7`)
|
|
34
|
+
n (int, optional): Number of generations to return. (default::obj:`1`)
|
|
35
|
+
response_format (object, optional): An object specifying the format
|
|
36
|
+
that the model must output.
|
|
37
|
+
stream (bool, optional): If set, tokens are returned as Server-Sent
|
|
38
|
+
Events as they are made available. (default: :obj:`False`)
|
|
39
|
+
stop (str or list, optional): Up to :obj:`4` sequences where the API
|
|
40
|
+
will stop generating further tokens. (default: :obj:`None`)
|
|
41
|
+
max_tokens (int, optional): The maximum number of tokens to generate.
|
|
42
|
+
(default: :obj:`None`)
|
|
43
|
+
frequency_penalty (float, optional): Number between :obj:`-2.0` and
|
|
44
|
+
:obj:`2.0`. Positive values penalize new tokens based on their
|
|
45
|
+
existing frequency in the text so far, decreasing the model's
|
|
46
|
+
likelihood to repeat the same line verbatim. See more information
|
|
47
|
+
about frequency and presence penalties. (default: :obj:`0.0`)
|
|
48
|
+
tools (list[FunctionTool], optional): A list of tools the model may
|
|
49
|
+
call. Currently, only functions are supported as a tool. Use this
|
|
50
|
+
to provide a list of functions the model may generate JSON inputs
|
|
51
|
+
for. A max of 128 functions are supported.
|
|
52
|
+
"""
|
|
53
|
+
|
|
54
|
+
temperature: float = 0.7
|
|
55
|
+
top_p: float = 0.7
|
|
56
|
+
n: int = 1
|
|
57
|
+
stream: bool = False
|
|
58
|
+
stop: Union[str, Sequence[str], NotGiven] = NOT_GIVEN
|
|
59
|
+
max_tokens: Union[int, NotGiven] = NOT_GIVEN
|
|
60
|
+
response_format: Union[Type[BaseModel], dict, NotGiven] = NOT_GIVEN
|
|
61
|
+
frequency_penalty: float = 0.0
|
|
62
|
+
|
|
63
|
+
def as_dict(self) -> dict[str, Any]:
|
|
64
|
+
r"""Convert the current configuration to a dictionary.
|
|
65
|
+
|
|
66
|
+
This method converts the current configuration object to a dictionary
|
|
67
|
+
representation, which can be used for serialization or other purposes.
|
|
68
|
+
|
|
69
|
+
Returns:
|
|
70
|
+
dict[str, Any]: A dictionary representation of the current
|
|
71
|
+
configuration.
|
|
72
|
+
"""
|
|
73
|
+
config_dict = self.model_dump()
|
|
74
|
+
if self.tools:
|
|
75
|
+
from camel.toolkits import FunctionTool
|
|
76
|
+
|
|
77
|
+
tools_schema = []
|
|
78
|
+
for tool in self.tools:
|
|
79
|
+
if not isinstance(tool, FunctionTool):
|
|
80
|
+
raise ValueError(
|
|
81
|
+
f"The tool {tool} should "
|
|
82
|
+
"be an instance of `FunctionTool`."
|
|
83
|
+
)
|
|
84
|
+
tools_schema.append(tool.get_openai_tool_schema())
|
|
85
|
+
config_dict["tools"] = NOT_GIVEN
|
|
86
|
+
return config_dict
|
|
87
|
+
|
|
88
|
+
|
|
89
|
+
SILICONFLOW_API_PARAMS = {
|
|
90
|
+
param for param in SiliconFlowConfig.model_fields.keys()
|
|
91
|
+
}
|
camel/datagen/__init__.py
CHANGED
|
@@ -12,10 +12,12 @@
|
|
|
12
12
|
# limitations under the License.
|
|
13
13
|
# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
|
|
14
14
|
|
|
15
|
-
from .
|
|
15
|
+
from .cot_datagen import CoTDataGenerator
|
|
16
|
+
from .self_improving_cot import SelfImprovingCoTPipeline
|
|
16
17
|
from .self_instruct import SelfInstructPipeline
|
|
17
18
|
|
|
18
19
|
__all__ = [
|
|
19
20
|
"CoTDataGenerator",
|
|
20
21
|
"SelfInstructPipeline",
|
|
22
|
+
"SelfImprovingCoTPipeline",
|
|
21
23
|
]
|