Initial commit

.env

@@ -0,0 +1 @@
+OPENAI_API_KEY=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

.gitattributes

@@ -0,0 +1,39 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+docs/test_file.pdf filter=lfs diff=lfs merge=lfs -text
+stores/openai/* filter=lfs diff=lfs merge=lfs -text
+stores/openai/index.faiss filter=lfs diff=lfs merge=lfs -text
+stores/openai/index.pkl filter=lfs diff=lfs merge=lfs -text

.gitignore~

@@ -0,0 +1 @@
+.env

.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

.gradio/flagged/dataset1.csv

@@ -0,0 +1,2 @@
+query,output,timestamp
+,,2025-08-17 17:23:18.609639

README.md

@@ -0,0 +1,14 @@
+---
+title: AIGINEER LLM
+emoji: 👁
+colorFrom: green
+colorTo: pink
+sdk: gradio
+sdk_version: 5.42.0
+app_file: app.py
+pinned: false
+license: mit
+short_description: RAG LLM for the AIGINEER company documentation
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,188 @@
+import gradio as gr
+from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.vectorstores import FAISS
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.prompts import PromptTemplate
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_huggingface.llms import HuggingFacePipeline
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
+from langchain_openai import ChatOpenAI, OpenAIEmbeddings
+import os
+from dotenv import load_dotenv
+import tiktoken
+
+load_dotenv()
+#HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
+
+#embeddings_model_name = "cointegrated/rubert-tiny2"
+embeddings_model_name = "text-embedding-3-large"
+
+llm_model_name = "gpt-4o-mini"
+
+store_save_path = "stores/openai"
+
+# Step 1: Document Loading and Splitting
+def load_and_split_documents(pdf_path="docs/test_file.pdf"):
+    """
+    Loads a PDF document and splits it into smaller chunks.
+    """
+    loader = PyPDFLoader(pdf_path)
+    documents = loader.load()
+    
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=800,
+        chunk_overlap=200
+    )
+    docs = text_splitter.split_documents(documents)
+    return docs
+
+# Step 2: Embeddings and Vector Store
+def get_vector_store(docs, store_save_path=store_save_path):
+    """
+    Loads an existing vector store or creates a new one if it doesn't exist.
+    """
+    if os.path.exists(store_save_path):
+        print("Loading vector store from disk...")
+        embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
+        db = FAISS.load_local(store_save_path, embeddings, allow_dangerous_deserialization=True)
+    else:
+        print("Creating a new vector store...")
+        embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
+        db = FAISS.from_documents(docs, embeddings)
+        db.save_local(store_save_path)
+    return db
+
+# Step 3: Initialize the LLM
+def initialize_llm():
+    """
+    Initializes a Russian-specific LLM locally using transformers
+    """
+    #repo_id = "ai-forever/rugpt3large_based_on_gpt2"
+    #repo_id = "ai-forever/ruBert-base"
+    #repo_id = "ai-forever/ruGPT-3.5-13B"
+
+    '''
+    llm = HuggingFaceEndpoint(
+        repo_id=repo_id,
+        temperature=0.5,
+        #max_new_tokens=300,
+        task='text-generation'
+    )
+    '''
+
+    llm = ChatOpenAI(
+        model=llm_model_name,
+        temperature=0.7
+    )
+
+    return llm
+
+# Step 4: Create the LCEL RAG Chain
+def setup_rag_chain(pdf_path):
+    """
+    Sets up the complete Retrieval-Augmented Generation chain using LCEL.
+    """
+    docs = load_and_split_documents(pdf_path)
+    db = get_vector_store(docs)
+    retriever = db.as_retriever()
+    llm = initialize_llm()
+    
+    # Checking the vector store
+    #print(f"Number of vectors in FAISS index: {db.index.ntotal}")
+    
+    # Define the prompt template
+    template = """Используйте следующие фрагменты контекста, чтобы ответить на вопрос в конце. Если вы не знаете ответа, просто скажите, что не знаете, не пытайтесь что-то придумать. Всегда будьте вежливым.
+
+{context}
+
+Вопрос: {question}
+
+Полезный ответ:"""
+    
+    prompt = PromptTemplate.from_template(template)
+    
+    # Corrected RAG chain construction
+    rag_chain = (
+        {"context": retriever, "question": RunnablePassthrough()}
+        | prompt
+        | llm
+        | StrOutputParser()
+    )
+    
+    return rag_chain
+
+# Initialize the chain
+document_name = "docs/test_file.pdf"
+qa_chain = setup_rag_chain(pdf_path=document_name)
+
+# Gradio Interface
+def chat_with_doc(query):
+    """
+    Function to handle the user query and return a response.
+    """
+    try:
+        # Pass the query directly, not as a dictionary
+        result = qa_chain.invoke(query)
+        return result
+    except Exception as e:
+        return f"Произошла ошибка: {type(e).__name__} - {e!r}"
+
+def count_tokens(text, model_name):
+    encoding = tiktoken.encoding_for_model(model_name)
+    num_tokens = len(encoding.encode(text))
+    return num_tokens
+
+iface = gr.Interface(
+    fn=chat_with_doc,
+    inputs=gr.Textbox(lines=5, placeholder="Спросите что-нибудь о документе..."),
+    outputs="text",
+    title="RAG LLM модель для AIGINEER",
+    description="Задайте вопрос о содержании документации",
+)
+
+css_code = """
+#submit-button {
+    background-color: #4CAF50 !important;
+    color: white !important;
+}
+
+#centered-text {
+    text-align: center;
+    //justify-content: center;
+}
+
+#fixed-height-textarea textarea {
+    overflow-y: auto !important;
+}
+"""
+
+heading_text = "# AIGINEER-ИИ Модель"
+subheading_text = 'Узнайте любую информацию о нормативно-технической документации (НТД) со 100% точностью при помощи ИИ модели AIGINEER'
+
+with gr.Blocks(css=css_code) as demo:
+    gr.Markdown(heading_text, elem_id='centered-text')
+    gr.Markdown(subheading_text, elem_id='centered-text')
+    with gr.Row(scale=1):
+        with gr.Column():
+            query_input = gr.Textbox(interactive=True, label='Вопрос', lines=5, placeholder="Спросите что-нибудь о документе...")
+            with gr.Row():
+                clear_button = gr.ClearButton(components=[query_input], variant='secondary', value='Очистить')
+                submit_button = gr.Button(variant='primary', value='Отправить')
+            #with gr.Column():
+            #    count_tokens_output = gr.TextArea(interactive=False, label='Стоимость запроса в токенах')
+            #    count_tokens_button = gr.Button(variant='secondary', value='Посчитать стоимость в токенах')
+        response_output = gr.TextArea(interactive=True, label='Ответ', lines=8, placeholder='Тут будет отображаться ответ.')
+    
+    submit_button.click(fn=chat_with_doc, inputs=query_input, outputs=response_output)
+    #count_tokens_button.click(fn=lambda text_input: count_tokens(text_input, llm_model_name), inputs=[query_input], outputs=[count_tokens_output])
+
+# Launch the Gradio app
+if __name__ == "__main__":
+    # Uncomment to run as CLI
+    #query = input(f"Спросите что нибудь о документе {document_name}: ")
+    #result = chat_with_doc(query)
+    #print(result)
+    
+    # Run Gradio app
+    demo.launch()

docs/test_file.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d09c4619a621b34f7245d18b681fa78a2c33904e7b5a9eb9e2d7c6381c5573dd
+size 1248822

requirements.txt

@@ -0,0 +1,9 @@
+gradio
+langchain-community
+langchain-core
+langchain-huggingface
+faiss-cpu
+pypdf
+sentence-transformers
+huggingface-hub
+langchain_openai

stores/openai/index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3ab998c61db77f7d7cb6170c9b56bdf1ad71add86ba86274a5b6171966c5d3
+size 1222701

stores/openai/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87e22ec5ccf7497c865e0993b818d49e27a6140e94d2ab4210bc547fbc196fdc
+size 284848

test.py

@@ -0,0 +1,5 @@
+from transformers import pipeline
+
+generator = pipeline("text-generation", model="ai-forever/rugpt3large_based_on_gpt2")
+result = generator("Сколько будет 2+2?", max_length=30)
+print(result[0]["generated_text"])