hjxdl 0.1.6__py3-none-any.whl → 0.1.8__py3-none-any.whl

hdl/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.1.6'
-__version_tuple__ = version_tuple = (0, 1, 6)
+__version__ = version = '0.1.8'
+__version_tuple__ = version_tuple = (0, 1, 8)

hdl/utils/llm/chat.py

@@ -1,10 +1,85 @@
 import typing as t
 
 from llama_cpp import Llama
-from jupyfuncs.llm.openapi import (
-    chat_oai_invoke,
-    chat_oai_stream
-)
+from openai import OpenAI
+# import traceback
+
+
+def chat_oai_stream(
+    base_url="http://127.0.0.1:8000/v1",
+    api_key="dummy_key",
+    model="/data/models/Qwen-7B-Chat-Int4",
+    prompt="Who are you?",
+    *args,
+    **kwargs
+):
+    """Chat with OpenAI's GPT-3 model using the specified parameters.
+    
+    Args:
+        base_url (str): The base URL for the OpenAI API. Default is "http://127.0.0.1:8000/v1".
+        api_key (str): The API key for accessing the OpenAI API. Default is "dummy_key".
+        model (str): The model ID to use for the chat. Default is "/data/models/Qwen-7B-Chat-Int4".
+        prompt (str): The initial prompt for the chat conversation.
+    
+    Yields:
+        str: The generated content from the chat conversation.
+    
+    """
+    client = OpenAI(
+        base_url=base_url,
+        api_key=api_key,
+    )
+    response = client.chat.completions.create(
+        model=model,
+        messages=[{
+            "role": "user",
+            "content": prompt
+        }],
+        stream=True,
+        *args,
+        **kwargs
+    )
+    
+    for chunk in response:
+        content = chunk.choices[0].delta.content
+        yield content
+
+
+def chat_oai_invoke(
+    base_url="http://127.0.0.1:8000/v1",
+    api_key="dummy_key",
+    model="/data/models/Qwen-7B-Chat-Int4",
+    prompt="Who are you?",
+    *args,
+    **kwargs
+):
+    """Invoke OpenAI chat API to generate a response based on the given prompt.
+    
+    Args:
+        base_url (str): The base URL of the OpenAI API. Default is "http://127.0.0.1:8000/v1".
+        api_key (str): The API key for accessing the OpenAI API. Default is "dummy_key".
+        model (str): The model to use for generating the response. Default is "/data/models/Qwen-7B-Chat-Int4".
+        prompt (str): The prompt message to start the conversation. Default is "Who are you?".
+    
+    Returns:
+        str: The response generated by the OpenAI chat API based on the prompt.
+    """
+    client = OpenAI(
+        base_url=base_url,
+        api_key=api_key,
+    )
+    response = client.chat.completions.create(
+        model=model,
+        messages=[{
+            "role": "user",
+            "content": prompt
+        }],
+        stream=False,
+        *args,
+        **kwargs
+    )
+    
+    return response.choices[0].message.content
 
 
 class GGUF_M(Llama):

hdl/utils/llm/embs.py

@@ -1,4 +1,4 @@
-from sentence_transformers import SentenceTransformer
+import re
 
 
 class BEEmbedder():
@@ -118,6 +118,8 @@ class HFEmbedder():
         Returns:
             None
         """
+
+        from sentence_transformers import SentenceTransformer
         
         self.device = device
         self.emb_dir = emb_dir
@@ -185,4 +187,28 @@ class HFEmbedder():
         output_1 = self.encode(sentences_1, *args, **kwargs)
         output_2 = self.encode(sentences_2, *args, **kwargs)
         similarity = output_1 @ output_2.T
-        return similarity
+        return similarity
+
+
+def get_n_tokens(
+    paragraph,
+    model: str = ""
+):
+    """Get the number of tokens in a paragraph using a specified model.
+    
+    Args:
+        paragraph (str): The input paragraph to tokenize.
+        model (str): The name of the model to use for tokenization. If None, a default CJK tokenization will be used.
+    
+    Returns:
+        int: The number of tokens in the paragraph based on the specified model or default CJK tokenization.
+    """
+    if model == "":
+        cjk_regex = re.compile(u'[\u1100-\uFFFDh]+?')
+        trimed_cjk = cjk_regex.sub( ' a ', paragraph, 0)
+        return len(trimed_cjk.split())
+    else:
+        import tiktoken
+        encoding = tiktoken.encoding_for_model(model)
+        num_tokens = len(encoding.encode(paragraph))
+        return num_tokens

{hjxdl-0.1.6.dist-info → hjxdl-0.1.8.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: hjxdl
-Version: 0.1.6
+Version: 0.1.8
 Summary: A collection of functions for Jupyter notebooks
 Home-page: https://github.com/huluxiaohuowa/hdl
 Author: Jianxing Hu

{hjxdl-0.1.6.dist-info → hjxdl-0.1.8.dist-info}/RECORD

@@ -1,5 +1,5 @@
 hdl/__init__.py,sha256=5sZZNySv08wwfzJcSDssGTqUn9wlmDsR6R4XB8J8mFM,70
-hdl/_version.py,sha256=L9DFp_i-1xztMRdin6ZHqLFMlDKUn5bsVgheqXFwTTc,411
+hdl/_version.py,sha256=PdJ7dZoz_SyEgX0MdrMfQYBFlGcwpemv6ibF8NKALBY,411
 hdl/args/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/args/loss_args.py,sha256=s7YzSdd7IjD24rZvvOrxLLFqMZQb9YylxKeyelSdrTk,70
 hdl/controllers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -84,12 +84,12 @@ hdl/utils/database_tools/connect.py,sha256=KUnVG-8raifEJ_N0b3c8LkTTIfn9NIyw8LX6q
 hdl/utils/general/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/utils/general/glob.py,sha256=8-RCnt6L297wMIfn34ZAMCsGCZUjHG3MGglGZI1cX0g,491
 hdl/utils/llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-hdl/utils/llm/chat.py,sha256=H2c8assJlSdZQKIfPkYrVZHqv66TsdsxtaLXv0kNe1w,11565
-hdl/utils/llm/embs.py,sha256=sC8tga7HgDwPI2m7TDWKp9kkxEIMxEyMtgmEhfRi4vI,6362
+hdl/utils/llm/chat.py,sha256=gsbqWh8fTcJUENU6ZuMClZAuSOLFnD5VP8kXOxGh3Zw,13776
+hdl/utils/llm/embs.py,sha256=Tf0FOYrOFZp7qQpEPiSCXzlgyHH0X9HVTUtsup74a9E,7174
 hdl/utils/llm/extract.py,sha256=2sK_WJzmYIc8iuWaM9DA6Nw3_6q1O4lJ5pKpcZo-bBA,6512
 hdl/utils/schedulers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/utils/schedulers/norm_lr.py,sha256=bDwCmdEK-WkgxQMFBiMuchv8Mm7C0-GZJ6usm-PQk14,4461
-hjxdl-0.1.6.dist-info/METADATA,sha256=CN_P3ubA3O0BR-KbIh1hclbj2N8bS_BN-ZaBR2jJCXc,542
-hjxdl-0.1.6.dist-info/WHEEL,sha256=-oYQCr74JF3a37z2nRlQays_SX2MqOANoqVjBBAP2yE,91
-hjxdl-0.1.6.dist-info/top_level.txt,sha256=-kxwTM5JPhylp06z3zAVO3w6_h7wtBfBo2zgM6YZoTk,4
-hjxdl-0.1.6.dist-info/RECORD,,
+hjxdl-0.1.8.dist-info/METADATA,sha256=a9BaE0EGy5G9EM3Tbsi4LMmIrCMFJUuDjFnmmu_nBW4,542
+hjxdl-0.1.8.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
+hjxdl-0.1.8.dist-info/top_level.txt,sha256=-kxwTM5JPhylp06z3zAVO3w6_h7wtBfBo2zgM6YZoTk,4
+hjxdl-0.1.8.dist-info/RECORD,,

{hjxdl-0.1.6.dist-info → hjxdl-0.1.8.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (71.0.3)
+Generator: setuptools (72.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any