novelai-python 0.5.0__tar.gz → 0.5.2__tar.gz

{novelai_python-0.5.0 → novelai_python-0.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: novelai-python
-Version: 0.5.0
+Version: 0.5.2
 Summary: NovelAI Python Binding With Pydantic
 Author-Email: sudoskys <coldlando@hotmail.com>
 License: Apache License 2.0
@@ -29,6 +29,7 @@ Requires-Dist: regex>=2023.12.25
 Requires-Dist: tokenizers>=0.15.2
 Requires-Dist: json-repair>=0.29.4
 Requires-Dist: robust-downloader>=0.0.2
+Requires-Dist: bchlib>=2.1.3
 Provides-Extra: testing
 Requires-Dist: pytest>=7.4.4; extra == "testing"
 Requires-Dist: pytest-asyncio>=0.23.4; extra == "testing"
@@ -75,7 +76,7 @@ The goal of this repository is to use Pydantic to build legitimate requests to a
 ### Usage 🖥️
 
 ```shell
-pip install novelai-python
+pip -U install novelai-python
 ```
 
 More examples can be found in the [playground](https://github.com/LlmKira/novelai-python/tree/main/playground)
@@ -122,7 +123,8 @@ from dotenv import load_dotenv
 from pydantic import SecretStr
 
 from novelai_python import APIError, LoginCredential
-from novelai_python.sdk.ai.generate import TextLLMModel, LLM, get_default_preset
+from novelai_python.sdk.ai.generate import TextLLMModel, LLM, get_default_preset, AdvanceLLMSetting
+from novelai_python.sdk.ai.generate._enum import get_model_preset
 
 load_dotenv()
 username = os.getenv("NOVELAI_USER", None)
@@ -141,8 +143,17 @@ async def chat(prompt: str):
         agent = LLM.build(
             prompt=prompt,
             model=model,
-            parameters=None  # Auto Select or get from preset
+            # parameters=None,  # Auto Select or get from preset
+            parameters=get_model_preset(TextLLMModel.ERATO).get_all_presets()[0].parameters,  # Select from enum preset
+            advanced_setting=AdvanceLLMSetting(
+                min_length=1,
+                max_length=None,  # Auto
+            )
         )
+        # NOTE:parameter > advanced_setting, which logic in generate/__init__.py
+        # If you not pass the parameter, it will use the default preset.
+        # So if you want to set the generation params, you should pass your own params.
+        # Only if you want to use some params not affect the generation, you can use advanced_setting.
         result = await agent.request(session=login_credential)
     except APIError as e:
         raise Exception(f"Error: {e.message}")
@@ -184,6 +195,14 @@ print(tokenizer_package.decode(tokenizer_package.encode(t_text)))
 
 ```
 
+## About Nsfw 🚫
+
+You might need some solutions for identifying NSFW content and adding a mosaic to prevent operational mishaps.
+
+https://dghs-imgutils.deepghs.org/main/api_doc/detect/nudenet.html
+
+https://dghs-imgutils.deepghs.org/main/api_doc/operate/censor.html
+
 ## Acknowledgements 🙏
 
 [BackEnd](https://api.novelai.net/docs)

{novelai_python-0.5.0 → novelai_python-0.5.2}/README.md

@@ -39,7 +39,7 @@ The goal of this repository is to use Pydantic to build legitimate requests to a
 ### Usage 🖥️
 
 ```shell
-pip install novelai-python
+pip -U install novelai-python
 ```
 
 More examples can be found in the [playground](https://github.com/LlmKira/novelai-python/tree/main/playground)
@@ -86,7 +86,8 @@ from dotenv import load_dotenv
 from pydantic import SecretStr
 
 from novelai_python import APIError, LoginCredential
-from novelai_python.sdk.ai.generate import TextLLMModel, LLM, get_default_preset
+from novelai_python.sdk.ai.generate import TextLLMModel, LLM, get_default_preset, AdvanceLLMSetting
+from novelai_python.sdk.ai.generate._enum import get_model_preset
 
 load_dotenv()
 username = os.getenv("NOVELAI_USER", None)
@@ -105,8 +106,17 @@ async def chat(prompt: str):
         agent = LLM.build(
             prompt=prompt,
             model=model,
-            parameters=None  # Auto Select or get from preset
+            # parameters=None,  # Auto Select or get from preset
+            parameters=get_model_preset(TextLLMModel.ERATO).get_all_presets()[0].parameters,  # Select from enum preset
+            advanced_setting=AdvanceLLMSetting(
+                min_length=1,
+                max_length=None,  # Auto
+            )
         )
+        # NOTE:parameter > advanced_setting, which logic in generate/__init__.py
+        # If you not pass the parameter, it will use the default preset.
+        # So if you want to set the generation params, you should pass your own params.
+        # Only if you want to use some params not affect the generation, you can use advanced_setting.
         result = await agent.request(session=login_credential)
     except APIError as e:
         raise Exception(f"Error: {e.message}")
@@ -148,6 +158,14 @@ print(tokenizer_package.decode(tokenizer_package.encode(t_text)))
 
 ```
 
+## About Nsfw 🚫
+
+You might need some solutions for identifying NSFW content and adding a mosaic to prevent operational mishaps.
+
+https://dghs-imgutils.deepghs.org/main/api_doc/detect/nudenet.html
+
+https://dghs-imgutils.deepghs.org/main/api_doc/operate/censor.html
+
 ## Acknowledgements 🙏
 
 [BackEnd](https://api.novelai.net/docs)

{novelai_python-0.5.0 → novelai_python-0.5.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "novelai-python"
-version = "0.5.0"
+version = "0.5.2"
 description = "NovelAI Python Binding With Pydantic"
 authors = [
     { name = "sudoskys", email = "coldlando@hotmail.com" },
@@ -28,6 +28,7 @@ dependencies = [
     "tokenizers>=0.15.2",
     "json-repair>=0.29.4",
     "robust-downloader>=0.0.2",
+    "bchlib>=2.1.3",
 ]
 requires-python = ">=3.9"
 readme = "README.md"

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/sdk/ai/_cost.py

@@ -1,11 +1,11 @@
 import math
 import random
-from typing import List, Optional, Union
+from typing import List, Optional
 
 from pydantic import BaseModel
 
 from novelai_python.sdk.ai._const import map, initialN, initial_n, step, newN
-from novelai_python.sdk.ai._enum import Sampler, Model, ModelGroups, get_model_group, ModelTypeAlias
+from novelai_python.sdk.ai._enum import Sampler, ModelGroups, get_model_group, ModelTypeAlias
 
 
 class Args(BaseModel):

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/sdk/ai/generate/__init__.py

@@ -83,9 +83,16 @@ class LLM(ApiBaseModel):
 
     @model_validator(mode="after")
     def normalize_model(self):
+        if self.model in [
+            TextLLMModel.NEO_2B, TextLLMModel.J_6B, TextLLMModel.J_6B_V3, TextLLMModel.J_6B_V4,
+            TextLLMModel.GENJI_JP_6B, TextLLMModel.GENJI_JP_6B_V2, TextLLMModel.GENJI_PYTHON_6B,
+            TextLLMModel.EUTERPE_V0, TextLLMModel.EUTERPE_V2, TextLLMModel.KRAKE_V1, TextLLMModel.KRAKE_V2,
+            TextLLMModel.CASSANDRA, TextLLMModel.COMMENT_BOT, TextLLMModel.INFILL, TextLLMModel.CLIO
+        ]:
+            self.endpoint = "https://api.novelai.net"
         tokenizer = NaiTokenizer(get_tokenizer_model(self.model))
         model_group = get_llm_group(self.model)
-        total_tokens = len(tokenizer.encode(self.input))
+        total_tokens = tokenizer.total_tokens()
         if isinstance(self.input, str):
             prompt = tokenizer.encode(self.input)
             dtype = "uint32" if self.model in [TextLLMModel.ERATO] else "uint16"
@@ -238,10 +245,10 @@ class LLM(ApiBaseModel):
 
         valid_sequences = []
         for cell in logit_bias_group_exp:
-            if not any(token < 0 or token >= total_tokens for token in cell.sequence):
+            if any(token < 0 or token >= total_tokens for token in cell.sequence):
                 # 超出范围的 Token
-                logger.debug(
-                    f"Bias {cell} contains tokens that are out of range and will be ignored."
+                logger.trace(
+                    f"Bias [{cell}] contains tokens that are out of range and will be ignored."
                 )
             else:
                 # 将有效偏置组添加到列表
@@ -258,8 +265,8 @@ class LLM(ApiBaseModel):
         valid_bad_words_ids = []
         self.parameters.bad_words_ids = bad_words_ids
         for ban_word in self.parameters.bad_words_ids:
-            if not any(token < 0 or token >= total_tokens for token in ban_word):
-                logger.warning(
+            if any(token < 0 or token >= total_tokens for token in ban_word):
+                logger.trace(
                     f"Bad word {ban_word} contains tokens that are out of range and will be ignored."
                 )
             else:
@@ -277,6 +284,10 @@ class LLM(ApiBaseModel):
         self.advanced_setting.num_logprobs = self.logprobs_count
         if not self.advanced_setting.max_length:
             self.advanced_setting.max_length = 40
+        if self.parameters.repetition_penalty_range == 0:
+            self.parameters.repetition_penalty_range = None
+        if self.parameters.repetition_penalty_slope == 0:
+            self.parameters.repetition_penalty_slope = None
         return self
 
     @classmethod
@@ -294,7 +305,15 @@ class LLM(ApiBaseModel):
               **kwargs
               ) -> "LLM":
         """
-        Generate Prompt
+        Generate
+        :param prompt: prompt, exp "Hello, World!"
+        :param model: TextLLMModelTypeAlias
+        :param advanced_setting: AdvanceLLMSetting, default None means auto-select
+        :param parameters: LLMGenerationParams, default None means auto-select, use get_model_preset
+        :param default_bias: bool, default True
+        :param logprobs_count: int, default None
+        :param bias_dinkus_asterism: bool, default False
+        :param prefix: str, default "vanilla"
         """
         assert isinstance(prompt, str), "Prompt must be a string"
         created = cls(
@@ -358,7 +377,7 @@ class LLM(ApiBaseModel):
                 message = response_data.get("error") or response_data.get(
                     "message") or f"Server error with status_code {response.status_code}"
                 status_code = response_data.get("statusCode", response.status_code)
-                if status_code == 200:
+                if status_code == 200 or status_code == 201:
                     output = response_data.get("output", None)
                     if not output:
                         raise APIError(

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/sdk/ai/generate/_enum.py

@@ -148,30 +148,12 @@ def get_repetition_penalty_whitelist(group: TextTokenizerGroup) -> List[str]:
     return []
 
 
-def get_default_preset(model: TextLLMModel):
-    if model == TextLLMModel.J_6B_V4:
-        return J6bv4.get_all_presets()[0]
-    elif model == TextLLMModel.NEO_2B:
-        return Neo2b.get_all_presets()[0]
-    elif model == TextLLMModel.GENJI_PYTHON_6B:
-        return Genjipython6b.get_all_presets()[0]
-    elif model == TextLLMModel.GENJI_JP_6B_V2:
-        return Genjijp6bv2.get_all_presets()[0]
-    elif model == TextLLMModel.EUTERPE_V2:
-        return Euterpev2.get_all_presets()[0]
-    elif model == TextLLMModel.KRAKE_V1:
-        return Krakev1.get_all_presets()[0]
-    elif model == TextLLMModel.KRAKE_V2:
-        return Krakev2.get_all_presets()[0]
-    elif model in COLORS_LLM:
-        return Colorful.get_all_presets()[0]
-    elif model == TextLLMModel.CLIO:
-        return Clio.get_all_presets()[0]
-    elif model == TextLLMModel.KAYRA:
-        return Kayra.get_all_presets()[0]
-    elif model == TextLLMModel.ERATO:
-        return Erato.get_all_presets()[0]
-    return Default.get_all_presets()[0]
+def get_model_preset(model: TextLLMModel) -> "ModelPresetManager":
+    return PresetMapping.get(model, Default)
+
+
+def get_default_preset(model: TextLLMModel) -> Preset:
+    return get_model_preset(model).get_all_presets()[0]
 
 
 class ModelPresetManager:
@@ -2279,3 +2261,24 @@ Erato.register_preset(
         description="Designed for writing in Japanese. Works fine for English too."
     )
 )
+
+PresetMapping = {
+    TextLLMModel.J_6B_V4: J6bv4,
+    TextLLMModel.NEO_2B: Neo2b,
+    TextLLMModel.GENJI_PYTHON_6B: Genjipython6b,
+    TextLLMModel.GENJI_JP_6B_V2: Genjijp6bv2,
+    TextLLMModel.EUTERPE_V2: Euterpev2,
+    TextLLMModel.KRAKE_V1: Krakev1,
+    TextLLMModel.KRAKE_V2: Krakev2,
+    TextLLMModel.BLUE: Colorful,
+    TextLLMModel.RED: Colorful,
+    TextLLMModel.GREEN: Colorful,
+    TextLLMModel.PURPLE: Colorful,
+    TextLLMModel.PINK: Colorful,
+    TextLLMModel.YELLOW: Colorful,
+    TextLLMModel.WHITE: Colorful,
+    TextLLMModel.BLACK: Colorful,
+    TextLLMModel.CLIO: Clio,
+    TextLLMModel.KAYRA: Kayra,
+    TextLLMModel.ERATO: Erato,
+}

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/sdk/ai/generate/_schema.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from enum import Enum
 from typing import List, Optional, Tuple, Union
 
-from pydantic import BaseModel, Field, model_validator, ConfigDict
+from pydantic import BaseModel, Field, ConfigDict
 
 """
 class StorySettings(BaseModel):
@@ -77,11 +77,6 @@ class LogitBiasGroup(BaseModel):
     ensure_sequence_finish: bool
     generate_once: bool
 
-    @model_validator(mode="before")
-    def validate_sequence(cls, value):
-        print(value)
-        return value
-
 
 class AdvanceLLMSetting(BaseModel):
     """
@@ -89,7 +84,7 @@ class AdvanceLLMSetting(BaseModel):
     """
     min_length: Optional[int] = 1
     max_length: Optional[int] = None
-    repetition_penalty: Optional[float] = None
+    repetition_penalty: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
     generate_until_sentence: Optional[bool] = True
     use_cache: Optional[bool] = False
     use_string: Optional[bool] = False
@@ -107,36 +102,36 @@ class LLMGenerationParams(BaseModel):
     LLM Generation Settings
     """
     textGenerationSettingsVersion: Optional[int] = None
-    temperature: Optional[float] = None
+    temperature: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
     max_length: Optional[int] = Field(default=None, ge=1, le=100)  # Max 150(vip3),100(vip2)
     min_length: Optional[int] = Field(default=None, ge=1, le=100)  # Max 150(vip3),100(vip2)
-    top_k: Optional[float] = Field(default=None, ge=0)
-    top_p: Optional[float] = Field(default=None, gt=0, le=1)
-    top_a: Optional[float] = Field(default=None, ge=0)
-    typical_p: Optional[float] = Field(default=None, ge=0, le=1)
-    tail_free_sampling: Optional[float] = Field(default=None, ge=0)
-    repetition_penalty: Optional[float] = Field(default=None, gt=0)
+    top_k: Optional[int] = Field(default=None, ge=0)
+    top_p: Optional[Union[float, int]] = Field(default=None, gt=0, le=1, allow_inf_nan=False)
+    top_a: Optional[Union[float, int]] = Field(default=None, ge=0, allow_inf_nan=False)
+    typical_p: Optional[Union[float, int]] = Field(default=None, ge=0, le=1, allow_inf_nan=False)
+    tail_free_sampling: Optional[Union[float, int]] = Field(default=None, ge=0, allow_inf_nan=False)
+    repetition_penalty: Optional[Union[float, int]] = Field(default=None, gt=0, allow_inf_nan=False)
     repetition_penalty_range: Optional[int] = Field(default=None, ge=0)
-    repetition_penalty_slope: Optional[float] = Field(default=None, ge=0)
+    repetition_penalty_slope: Optional[Union[float, int]] = Field(default=None, ge=0, allow_inf_nan=False)
 
     eos_token_id: int = None
     bad_words_ids: List[List[int]] = None
     logit_bias_groups: Optional[List[LogitBiasGroup]] = []
 
-    repetition_penalty_frequency: Optional[float] = None
-    repetition_penalty_presence: Optional[float] = None
+    repetition_penalty_frequency: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    repetition_penalty_presence: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
     repetition_penalty_whitelist: Optional[List[int]] = None
     repetition_penalty_default_whitelist: Optional[bool] = None
-    cfg_scale: Optional[float] = None
+    cfg_scale: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
     cfg_uc: Optional[str] = None
     phrase_rep_pen: PenStyle = PenStyle.Off
-    top_g: Optional[float] = None
-    mirostat_tau: Optional[float] = None
-    mirostat_lr: Optional[float] = None
-    math1_temp: Optional[float] = None
-    math1_quad: Optional[float] = None
-    math1_quad_entropy_scale: Optional[float] = None
-    min_p: Optional[float] = None
+    top_g: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    mirostat_tau: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    mirostat_lr: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    math1_temp: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    math1_quad: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    math1_quad_entropy_scale: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
+    min_p: Optional[Union[float, int]] = Field(default=None, allow_inf_nan=False)
 
     order: Union[List[int], List[KeyOrderEntry]] = [
         KeyOrderEntry(id=Key.Cfg, enabled=False),

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/sdk/ai/generate_stream.py

@@ -134,7 +134,7 @@ class LLMStream(LLM):
                             request=request_data, code=status_code, response=message
                         )
                 else:
-                    if response.status_code != 200:
+                    if response.status_code not in [200, 201]:
                         raise APIError(
                             f"Server error with status code {response.status_code}",
                             request=request_data, code=response.status_code, response=response.content

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/tokenizer/__init__.py

@@ -187,6 +187,13 @@ class NaiTokenizer:
             return self.tokenizer.encode(text).tokens
         raise NotImplementedError("Tokenizer does not support token encoding")
 
+    def total_tokens(self) -> int:
+        if isinstance(self.tokenizer, Tokenizer):
+            return len(self.tokenizer.get_vocab())
+        if isinstance(self.tokenizer, SimpleTokenizer):
+            return len(self.tokenizer.get_vocab())
+        raise NotImplementedError("Tokenizer does not support token encoding")
+
     def encode(self, sentence: str) -> List[int]:
         if isinstance(self.tokenizer, SimpleTokenizer):
             return self.tokenizer.encode(sentence).ids

{novelai_python-0.5.0 → novelai_python-0.5.2}/src/novelai_python/tokenizer/clip_simple_tokenizer.py

@@ -144,3 +144,11 @@ class SimpleTokenizer(object):
         text = ''.join([self.decoder[token] for token in tokens])
         text = bytearray([self.byte_decoder[c] for c in text]).decode('utf-8', errors="replace").replace('</w>', ' ')
         return text
+
+    def get_vocab(self):
+        return self.encoder
+
+
+if __name__ == '__main__':
+    tokenizer = SimpleTokenizer()
+    print(len(tokenizer.get_vocab()))