Upload 8 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* 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
+data/Encyclopedia[[:space:]]of[[:space:]]Bodybuilding_[[:space:]]The[[:space:]]Complete[[:space:]]A-Z[[:space:]]Book[[:space:]]on[[:space:]]Muscle[[:space:]]Building[[:space:]]([[:space:]]PDFDrive[[:space:]])_compressed.pdf filter=lfs diff=lfs merge=lfs -text
+gym_vector_db/index.faiss filter=lfs diff=lfs merge=lfs -text

data/Encyclopedia of Bodybuilding_ The Complete A-Z Book on Muscle Building ( PDFDrive )_compressed.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0501fc057d1216c4ef74ccba202e925238d422fd39af91c073e8422ecf101271
+size 35789889

gemini_gym_app.py

@@ -0,0 +1,124 @@
+from langchain_community.vectorstores import FAISS
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.prompts import PromptTemplate
+import os
+from langchain.memory import ConversationBufferWindowMemory
+from langchain.chains import ConversationalRetrievalChain
+import time
+from langchain_google_genai import ChatGoogleGenerativeAI
+import streamlit as st
+
+st.set_page_config(page_title="GymGPT")
+col1, col2, col3 = st.columns([1,2,1])
+with col2:
+    st.title("GymGPT 🦾")
+
+
+st.sidebar.title("Welcome to GymGPT")
+st.sidebar.image(r"C:\Users\Sagaya Abinesh\Documents\gymgpt\WE_GO_JIM.png", width=275)
+st.sidebar.title("Shoot your gym-related questions")
+st.markdown(
+    """
+    <style>
+div.stButton > button:first-child {
+    background-color: #ffd0d0;
+}
+
+div.stButton > button:active {
+    background-color: #ff6262;
+}
+
+.st-emotion-cache-6qob1r {
+    position: relative;
+    height: 100%;
+    width: 100%;
+    background-color: black;
+    overflow: overlay;
+}
+
+   div[data-testid="stStatusWidget"] div button {
+        display: none;
+        }
+    
+    .reportview-container {
+            margin-top: -2em;
+        }
+        #MainMenu {visibility: hidden;}
+        .stDeployButton {display:none;}
+        footer {visibility: hidden;}
+        #stDecoration {display:none;}
+    button[title="View fullscreen"]{
+    visibility: hidden;}
+        </style>
+""",
+    unsafe_allow_html=True,
+)
+
+def reset_conversation():
+    st.session_state.messages = []
+    st.session_state.memory.clear()
+
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
+if "memory" not in st.session_state:
+    st.session_state.memory = ConversationBufferWindowMemory(k=2, memory_key="chat_history",return_messages=True) 
+
+embeddings = HuggingFaceEmbeddings(model_name="nomic-ai/nomic-embed-text-v1-ablated", model_kwargs={"trust_remote_code": True})
+db = FAISS.load_local(r"C:\Users\Sagaya Abinesh\Documents\gymgpt\gym_vector_db", embeddings)
+db_retriever = db.as_retriever(search_type="similarity",search_kwargs={"k": 4})
+
+prompt_template = """<s>[INST]This is a chat template and you are the gym trainer, your primary objective is to provide accurate and concise information related to gym, workout, bodybuilding based on the user's questions. Do not generate your own questions and answers. You will adhere strictly to the instructions provided, offering relevant context from the knowledge base while avoiding unnecessary details. Your responses will be brief, to the point, and in compliance with the established format. If a question falls outside the given context, rely on your own knowledge base to generate an appropriate response. You will prioritize the user's query and refrain from posing additional questions and do not repeat the prompt template and the things that you have said already.Reply Only in English.give a detailed answer.
+QUESTION: {question}
+CONTEXT: {context}
+CHAT HISTORY: {chat_history}[/INST]
+ASSISTANT:
+</s>
+"""
+
+prompt = PromptTemplate(template=prompt_template,
+                        input_variables=['question', 'context', 'chat_history'])
+
+import os
+os.environ['GOOGLE_API_KEY'] = "AIzaSyCIXtU3S3iGxsvPBbVhUUt3g23iL3hJuVs"
+# google_api_key="AIzaSyCIXtU3S3iGxsvPBbVhUUt3g23iL3hJuVs"
+# if not google_api_key:
+#     raise ValueError("GEMINI_API_KEY environment variable not set.")
+
+llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash")
+
+# Create a conversational chain using only your database retriever
+qa = ConversationalRetrievalChain.from_llm(
+    llm=llm,
+    memory=st.session_state.memory,
+    retriever=db_retriever,
+    combine_docs_chain_kwargs={'prompt': prompt}
+)
+
+for message in st.session_state.messages:
+    with st.chat_message(message.get("role")):
+        st.write(message.get("content"))
+
+input_prompt = st.chat_input("Say something")
+
+if input_prompt:
+    with st.chat_message("user"):
+        st.write(input_prompt)
+
+    st.session_state.messages.append({"role":"user","content":input_prompt})
+
+    with st.chat_message("assistant"):
+        with st.status("Lifting data, one bit at a time 💡🦾...",expanded=True):
+            result = qa.invoke(input=input_prompt)
+
+            message_placeholder = st.empty()
+
+            full_response = "⚠️ **_Note: Information provided may be inaccurate._** \n\n\n"
+        for chunk in result["answer"]:
+            full_response+=chunk
+            time.sleep(0.02)
+            
+            message_placeholder.markdown(full_response+" ▌")
+        st.button('Reset All Chat 🗑️', on_click=reset_conversation)
+
+    st.session_state.messages.append({"role":"assistant","content":result["answer"]})

gym_vector_db/index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40dc2703384c7b5c4023d86992e2a5b2e79bd09dd188e80f851e25b5ce5df24b
+size 7253037

gym_vector_db/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16a2e580cab7ba89ce6727d80f7f48fb6532c58ffcdbbf021a5db9630dab750b
+size 2117373

ingest.py

@@ -0,0 +1,31 @@
+from dotenv import load_dotenv,find_dotenv
+from transformers import pipeline
+import os
+
+
+
+load_dotenv(find_dotenv())
+
+
+from langchain_community.document_loaders import PyPDFLoader, DirectoryLoader
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+import faiss
+from langchain_community.embeddings import HuggingFaceInferenceAPIEmbeddings
+
+directory_path = "data"  # Path to the directory containing your PDF files
+loader = DirectoryLoader(directory_path, glob="./*.pdf", loader_cls=PyPDFLoader)
+documents = loader.load()
+
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=200)
+texts = text_splitter.split_documents(documents)
+
+# Create a SentenceTransformer object
+embeddings = HuggingFaceEmbeddings(model_name="nomic-ai/nomic-embed-text-v1-ablated",model_kwargs={"trust_remote_code":True})
+
+
+db = FAISS.from_documents(texts, embeddings)
+
+
+db.save_local("gym_vector_db")

requirements.txt

@@ -0,0 +1,8 @@
+einops
+langchain-groq
+sentence_transformers
+faiss-cpu
+langchain-community
+langchain
+pypdf
+huggingface_hub