Generative ai projects added

GenerativeAI/Advance/Langchain based Chat Retrival Chatbot/README.md

@@ -0,0 +1,42 @@
+---
+title: ! 
+emoji: 🏠
+colorFrom: #FF5733
+colorTo: #FFBF00
+sdk: streamlit
+sdk_version: 1.25.0
+app_file: app.py
+pinned: false
+---
+
+Dataset Link : https://www.kaggle.com/datasets/sameep98/housing-prices-in-mumbai
+
+# Llama
+
+LlaMA (Large Language Model Meta AI) is a Generative AI model, specifically a group of foundational Large Language Models developed by Meta AI, a company owned by Meta(Formerly Facebook). Meta announced Llama in Feb of 2023. Meta released Llama in different sizes(based on parameters), i.e., 7,13,33, and 65 billion parameters with a context length of 2k tokens. The model is with the intent to help researchers advance their knowledge in the field of AI. The small 7B models allow researchers with low computation power to study these models.
+
+# Llama 2
+
+LlaMA 2 surpasses the previous version, LlaMA version 1, which Meta released in July of 2023. It came out in three sizes: 7B, 13B, and 70B parameter models. Upon its release, LlaMA 2 achieved the highest score on Hugging Face. Even across all segments (7B, 13B, and 70B), the top-performing model on Hugging Face originates from LlaMA 2, having been fine-tuned or retrained.
+
+This release includes model weights and starting code for pretrained and fine-tuned Llama language models — ranging from 7B to 70B parameters.
+
+## Download
+
+The source code for Llama 2 is available on GitHub. If you want to work with the original weights, these are also available, but for this, you need to provide your name and email to the Meta AIs website. So go to the Meta AI by [clicking here](https://ai.meta.com/resources/models-and-libraries/llama-downloads/), then enter your name, email address, and organization(student if you are not working). Then scroll down and click on accept and continue. Now you will get a mail stating that you can download the model weights
+
+### Access on Hugging Face
+
+We are also providing downloads on [Hugging Face](https://huggingface.co/TheBloke/Llama-2-7B-GGML/tree/main). You must first request a download from the Meta website using the same email address as your Hugging Face account. After doing so, you can request access to any of the models on Hugging Face and within 1-2 days your account will be granted access to all versions.
+
+
+## References
+
+1. [Research Paper](https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/)
+2. [Llama 2 technical overview](https://ai.meta.com/resources/models-and-libraries/llama)
+3. [Hugging Face](https://huggingface.co/meta-llama)
+
+## Original LLaMA
+
+The repo for the original llama release is in the [`llama_v1`](https://github.com/facebookresearch/llama/tree/llama_v1) branch.
+

GenerativeAI/Advance/Langchain based Chat Retrival Chatbot/app.py

@@ -0,0 +1,90 @@
+import streamlit as st
+from langchain.chains import ConversationalRetrievalChain
+from langchain.document_loaders import CSVLoader
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.llms import CTransformers
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import FAISS
+from langchain.memory import ConversationBufferMemory
+from langchain.document_loaders.csv_loader import CSVLoader
+
+# Set Streamlit page configuration
+st.set_page_config(
+    page_title="ApnaGhar 🏠 - Mumbai Property Insights",
+    page_icon="",  # You can set an icon if needed
+    layout="wide",
+    initial_sidebar_state="expanded",
+)
+
+# Load data and set up language chain components
+loader = CSVLoader(file_path='Mumbai1.csv')
+documents = loader.load()
+
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+text_chunks = text_splitter.split_documents(documents)
+
+embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2",
+                                   model_kwargs={'device': "cpu"})
+
+vector_store = FAISS.from_documents(text_chunks, embeddings)
+
+llm = CTransformers(model="link llma model", model_type="llama",
+                    config={'max_new_tokens': 128, 'temperature': 0.01})
+
+memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
+
+chain = ConversationalRetrievalChain.from_llm(llm=llm, chain_type='stuff',
+                                              retriever=vector_store.as_retriever(search_kwargs={"k": 2}),
+                                              memory=memory)
+
+st.sidebar.title("ApnaGhar 🏠 - Mumbai Property Insights")
+st.sidebar.info("Welcome to ApnaGhar! Explore Mumbai's real estate and neighborhood information.")
+github_link = "[GitHub]()"
+st.sidebar.info("To contribute and Sponsor - " + github_link)
+
+if 'query_stack' not in st.session_state:
+    st.session_state['query_stack'] = []
+
+st.sidebar.subheader("Property Queries:")
+for i, past_query in enumerate(reversed(st.session_state['query_stack'])):
+    st.sidebar.write(f"{i + 1}. {past_query}")
+
+st.title("ApnaGhar - Mumbai Property Insights")
+
+if 'history' not in st.session_state:
+    st.session_state['history'] = []
+
+if 'generated' not in st.session_state:
+    st.session_state['generated'] = ["Hello! I'm ApnaGhar, here to help you with Mumbai property insights."]
+
+if 'past' not in st.session_state:
+    st.session_state['past'] = ["Hello!"]
+
+reply_container = st.container()
+container = st.container()
+
+with container:
+    with st.form(key='user_form', clear_on_submit=True):
+        user_input = st.text_input("Your Query:", placeholder="Ask anything about Mumbai properties or neighborhoods", key='input_user')
+        submit_button_user = st.form_submit_button(label='Send')
+
+    try:
+        if submit_button_user and user_input:
+            output = chain({"question": user_input, "chat_history": st.session_state['history']})["answer"]
+            st.session_state['past'].append(user_input)
+            st.session_state['generated'].append(output)
+
+            # Push user input to the stack
+            st.session_state['query_stack'].append(user_input)
+    except Exception as e:
+        st.error(f"An error occurred: {str(e)}")
+
+if st.session_state['generated']:
+    with reply_container:
+        for i in range(len(st.session_state['generated'])):
+            user_message = st.session_state["past"][i]
+            generated_message = st.session_state["generated"][i]
+
+            st.text(f"User: {user_message}")
+
+            st.text(f"ApnaGhar: {generated_message}")

GenerativeAI/Advance/Langchain based Chat Retrival Chatbot/requirements.txt

@@ -0,0 +1,14 @@
+langchain
+torch
+accelerate
+#bitsandbytes
+transformers
+sentence_transformers
+streamlit
+streamlit_chat
+faiss-cpu
+altair
+tiktoken
+huggingface-hub
+pypdf
+ctransformers

GenerativeAI/Basic/ImageToTextGenerator/ImageToTextGenerator.ipynb

@@ -0,0 +1,210 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "provenance": [],
+      "gpuType": "T4"
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "accelerator": "GPU",
+    "gpuClass": "standard"
+  },
+  "cells": [
+    {
+      "cell_type": "code",
+      "execution_count": 16,
+      "metadata": {
+        "id": "SeT-a9Byby1n"
+      },
+      "outputs": [],
+      "source": [
+        "\n",
+        "from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer\n",
+        "import torch\n",
+        "from PIL import Image"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "\n",
+        "model = VisionEncoderDecoderModel.from_pretrained(\"nlpconnect/vit-gpt2-image-captioning\")\n",
+        "feature_extractor = ViTImageProcessor.from_pretrained(\"nlpconnect/vit-gpt2-image-captioning\")\n",
+        "tokenizer = AutoTokenizer.from_pretrained(\"nlpconnect/vit-gpt2-image-captioning\")\n",
+        "\n",
+        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
+        "model.to(device)"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "Gh2jscQnot8g",
+        "outputId": "fe64ca40-7f91-4cd4-8967-5c00bb0ce857"
+      },
+      "execution_count": 17,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "VisionEncoderDecoderModel(\n",
+              "  (encoder): ViTModel(\n",
+              "    (embeddings): ViTEmbeddings(\n",
+              "      (patch_embeddings): ViTPatchEmbeddings(\n",
+              "        (projection): Conv2d(3, 768, kernel_size=(16, 16), stride=(16, 16))\n",
+              "      )\n",
+              "      (dropout): Dropout(p=0.0, inplace=False)\n",
+              "    )\n",
+              "    (encoder): ViTEncoder(\n",
+              "      (layer): ModuleList(\n",
+              "        (0-11): 12 x ViTLayer(\n",
+              "          (attention): ViTAttention(\n",
+              "            (attention): ViTSelfAttention(\n",
+              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
+              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
+              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
+              "              (dropout): Dropout(p=0.0, inplace=False)\n",
+              "            )\n",
+              "            (output): ViTSelfOutput(\n",
+              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
+              "              (dropout): Dropout(p=0.0, inplace=False)\n",
+              "            )\n",
+              "          )\n",
+              "          (intermediate): ViTIntermediate(\n",
+              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
+              "            (intermediate_act_fn): GELUActivation()\n",
+              "          )\n",
+              "          (output): ViTOutput(\n",
+              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
+              "            (dropout): Dropout(p=0.0, inplace=False)\n",
+              "          )\n",
+              "          (layernorm_before): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
+              "          (layernorm_after): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
+              "        )\n",
+              "      )\n",
+              "    )\n",
+              "    (layernorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
+              "    (pooler): ViTPooler(\n",
+              "      (dense): Linear(in_features=768, out_features=768, bias=True)\n",
+              "      (activation): Tanh()\n",
+              "    )\n",
+              "  )\n",
+              "  (decoder): GPT2LMHeadModel(\n",
+              "    (transformer): GPT2Model(\n",
+              "      (wte): Embedding(50257, 768)\n",
+              "      (wpe): Embedding(1024, 768)\n",
+              "      (drop): Dropout(p=0.1, inplace=False)\n",
+              "      (h): ModuleList(\n",
+              "        (0-11): 12 x GPT2Block(\n",
+              "          (ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
+              "          (attn): GPT2Attention(\n",
+              "            (c_attn): Conv1D()\n",
+              "            (c_proj): Conv1D()\n",
+              "            (attn_dropout): Dropout(p=0.1, inplace=False)\n",
+              "            (resid_dropout): Dropout(p=0.1, inplace=False)\n",
+              "          )\n",
+              "          (ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
+              "          (crossattention): GPT2Attention(\n",
+              "            (c_attn): Conv1D()\n",
+              "            (q_attn): Conv1D()\n",
+              "            (c_proj): Conv1D()\n",
+              "            (attn_dropout): Dropout(p=0.1, inplace=False)\n",
+              "            (resid_dropout): Dropout(p=0.1, inplace=False)\n",
+              "          )\n",
+              "          (ln_cross_attn): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
+              "          (mlp): GPT2MLP(\n",
+              "            (c_fc): Conv1D()\n",
+              "            (c_proj): Conv1D()\n",
+              "            (act): NewGELUActivation()\n",
+              "            (dropout): Dropout(p=0.1, inplace=False)\n",
+              "          )\n",
+              "        )\n",
+              "      )\n",
+              "      (ln_f): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
+              "    )\n",
+              "    (lm_head): Linear(in_features=768, out_features=50257, bias=False)\n",
+              "  )\n",
+              ")"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 17
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "max_length = 16\n",
+        "num_beams = 4\n",
+        "gen_kwargs = {\"max_length\": max_length, \"num_beams\": num_beams}"
+      ],
+      "metadata": {
+        "id": "hm6EtiPoot27"
+      },
+      "execution_count": 18,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "def predict_step(image_paths):\n",
+        "  images = []\n",
+        "  for image_path in image_paths:\n",
+        "    i_image = Image.open(image_path)\n",
+        "    if i_image.mode != \"RGB\":\n",
+        "      i_image = i_image.convert(mode=\"RGB\")\n",
+        "\n",
+        "    images.append(i_image)\n",
+        "\n",
+        "  pixel_values = feature_extractor(images=images, return_tensors=\"pt\").pixel_values\n",
+        "  pixel_values = pixel_values.to(device)\n",
+        "\n",
+        "  output_ids = model.generate(pixel_values, **gen_kwargs)\n",
+        "\n",
+        "  preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)\n",
+        "  preds = [pred.strip() for pred in preds]\n",
+        "  return preds"
+      ],
+      "metadata": {
+        "id": "FQ298E4gotu-"
+      },
+      "execution_count": 19,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "predict_step(['/content/drive/MyDrive/images/Plane-flying-on-earth-atmosphere.jpg']) "
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "u5ajgeTho6G_",
+        "outputId": "e6f62aa7-b0b7-4eff-947f-85a3a84e87ce"
+      },
+      "execution_count": 22,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "['a large jetliner flying through a blue sky']"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 22
+        }
+      ]
+    }
+  ]
+}