add more supported format

This reverts commit 2ddbf1ba0d.

examples/demo_advanced.ipynb

@@ -4,9 +4,11 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Llama Parser <> LlamaIndex\n",
+    "# Advanced RAG with LlamaParse\n",
     "\n",
-    "This notebook is a complete walkthrough for using `LlamaParse` for RAG applications with `LlamaIndex`.\n",
+    "<a href=\"https://colab.research.google.com/github/run-llama/llama_parse/blob/main/examples/demo_advanced.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n",
+    "\n",
+    "This notebook shows you how to use LlamaParse with our advanced markdown ingestion and recursive retrieval algorithms to model tables/text within a document hierarchically. This lets you ask questions over both tables and text.\n",
     "\n",
     "Note for this example, we are using the `llama_index >=0.10.4` version"
    ]

examples/demo_basic.ipynb

@@ -182,9 +182,8 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
-  },
-  "orig_nbformat": 4
+   "version": "3.10.10"
+  }
  },
  "nbformat": 4,
  "nbformat_minor": 4

examples/demo_mongodb.ipynb

@@ -0,0 +1,434 @@
+{
+  "cells": [
+    {
+      "attachments": {},
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "W6SX9VAnximx"
+      },
+      "source": [
+        "# LlamaParse With MongoDB\n",
+        "\n",
+        "<a href=\"https://colab.research.google.com/github/run-llama/llama_parse/blob/main/examples/demo_mongodb.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n",
+        "\n",
+        "In this notebook, we provide a straightforward example of using LlamaParse with MongoDB Atlas VectorSearch.\n",
+        "\n",
+        "We illustrate the process of using llama-parse to parse a PDF document, then index the document with a MongoDB vector store, and subsequently perform basic queries against this store.\n",
+        "\n",
+        "This notebook is structured similarly to quick start guides, aiming to introduce users to utilizing llama-parse in conjunction with a MongoDB Atlas VectorSearch."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "rUJKhWDHxr_k"
+      },
+      "source": [
+        "### Installation"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "U6ZkIeBnxfRb"
+      },
+      "outputs": [],
+      "source": [
+        "!pip install llama-index llama-parse pip install llama-index-vector-stores-mongodb llama-index-llms-openai"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "wh1eeFJe1gkY"
+      },
+      "source": [
+        "### Setup API Keys"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 1,
+      "metadata": {
+        "id": "I5slpdnyxwIB"
+      },
+      "outputs": [],
+      "source": [
+        "import os\n",
+        "os.environ[\"LLAMA_CLOUD_API_KEY\"] = '' # Get it from https://cloud.llamaindex.ai/api-key\n",
+        "os.environ['OPENAI_API_KEY'] = '' # Get it from https://platform.openai.com/api-keys"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {
+        "id": "es2mz_OVyQw9"
+      },
+      "outputs": [],
+      "source": [
+        "# llama-parse is async-first, running the sync code in a notebook requires the use of nest_asyncio\n",
+        "import nest_asyncio\n",
+        "\n",
+        "nest_asyncio.apply()\n",
+        "\n",
+        "import requests\n",
+        "import pymongo\n",
+        "\n",
+        "from llama_index.vector_stores.mongodb import MongoDBAtlasVectorSearch\n",
+        "from llama_parse import LlamaParse\n",
+        "from llama_index.embeddings.openai import OpenAIEmbedding\n",
+        "from llama_index.core import VectorStoreIndex, StorageContext\n",
+        "from llama_index.core.node_parser import SimpleNodeParser"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Ou3bVdHQ10X5"
+      },
+      "source": [
+        "### Download Document\n",
+        "\n",
+        "We will use `Attention is all you need` paper."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 3,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "YO9lAk6bybV3",
+        "outputId": "5cee588a-bec5-482e-e8ef-fbb78e8a5967"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Download complete.\n"
+          ]
+        }
+      ],
+      "source": [
+        "# The URL of the file you want to download\n",
+        "url = \"https://arxiv.org/pdf/1706.03762.pdf\"\n",
+        "# The local path where you want to save the file\n",
+        "file_path = \"./attention.pdf\"\n",
+        "\n",
+        "# Perform the HTTP request\n",
+        "response = requests.get(url)\n",
+        "\n",
+        "# Check if the request was successful\n",
+        "if response.status_code == 200:\n",
+        "    # Open the file in binary write mode and save the content\n",
+        "    with open(file_path, \"wb\") as file:\n",
+        "        file.write(response.content)\n",
+        "    print(\"Download complete.\")\n",
+        "else:\n",
+        "    print(\"Error downloading the file.\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "1NtR7PGo13Hh"
+      },
+      "source": [
+        "### Parse the document using `LlamaParse`."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 4,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "reeJsblfyeSd",
+        "outputId": "bb569e9f-fe31-47b9-a059-d7da369b3f94"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Started parsing the file under job_id 09a49745-9f21-4190-9de8-27e4e1a4bdf5\n"
+          ]
+        }
+      ],
+      "source": [
+        "documents = LlamaParse(result_type=\"text\").load_data(file_path)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "-NIXtCBwyiPp",
+        "outputId": "ad4b3cec-2c23-4858-81f0-994ae2c96b8f"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "rmer - model architecture.\n",
+            "The Transformer follows this overall architecture using stacked self-attention and point-wise, fully\n",
+            "connected layers for both the encoder and decoder, shown in the left and right halves of Figure 1,\n",
+            "respectively.\n",
+            "3.1   Encoder and Decoder Stacks\n",
+            "Encoder:     The encoder is composed of a stack of N = 6 identical layers. Each layer has two\n",
+            "sub-layers. The first is a multi-head self-attention mechanism, and the second is a simple, position-\n",
+            "wise fully connected feed-forward network. We employ a residual connection [11] around each of\n",
+            "the two sub-layers, followed by layer normalization [1]. That is, the output of each sub-layer is\n",
+            "LayerNorm(x + Sublayer(x)), where Sublayer(x) is the function implemented by the sub-layer\n",
+            "itself. To facilitate these residual connections, all sub-layers in the model, as well as the embedding\n",
+            "layers, produce outputs of dimension dmodel = 512.\n",
+            "Decoder:    The decoder is also composed of a stack of N = 6 identical layers. In addition \n"
+          ]
+        }
+      ],
+      "source": [
+        "# Take a quick look at some of the parsed text from the document:\n",
+        "print(documents[0].get_content()[10000:11000])"
+      ]
+    },
+    {
+      "attachments": {},
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "wP9I5dhB1-w1"
+      },
+      "source": [
+        "### Create `MongoDBAtlasVectorSearch`."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 6,
+      "metadata": {
+        "id": "-4Ek0oK-yp3L"
+      },
+      "outputs": [],
+      "source": [
+        "mongo_uri = os.environ[\"MONGO_URI\"]\n",
+        "\n",
+        "mongodb_client = pymongo.MongoClient(mongo_uri)\n",
+        "mongodb_vector_store = MongoDBAtlasVectorSearch(mongodb_client)\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "GYiVwFok2DNf"
+      },
+      "source": [
+        "### Create nodes."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 7,
+      "metadata": {
+        "id": "aqdF6ZonytHF"
+      },
+      "outputs": [],
+      "source": [
+        "node_parser = SimpleNodeParser()\n",
+        "\n",
+        "nodes = node_parser.get_nodes_from_documents(documents)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "U5fMoGrA2GSH"
+      },
+      "source": [
+        "### Create Index and Query Engine."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 8,
+      "metadata": {
+        "id": "gQUieIrAywSC"
+      },
+      "outputs": [],
+      "source": [
+        "storage_context = StorageContext.from_defaults(vector_store=mongodb_vector_store)\n",
+        "\n",
+        "index = VectorStoreIndex(\n",
+        "    nodes=nodes,\n",
+        "    storage_context=storage_context,\n",
+        "    embed_model=OpenAIEmbedding(),\n",
+        ")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {
+        "id": "snkZZss-zKDb"
+      },
+      "outputs": [],
+      "source": [
+        "query_engine = index.as_query_engine(similarity_top_k=2)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "rTKT34XO2LYk"
+      },
+      "source": [
+        "### Test Query"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 10,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "r66ciuPkzNv1",
+        "outputId": "919218e3-0884-4992-802c-ab1c4622ec4b"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\n",
+            "***********New LlamaParse+ Basic Query Engine***********\n",
+            "The BLEU score on the WMT 2014 English-to-German translation task is 28.4.\n"
+          ]
+        }
+      ],
+      "source": [
+        "query = \"What is BLEU score on the WMT 2014 English-to-German translation task?\"\n",
+        "\n",
+        "response = query_engine.query(query)\n",
+        "print(\"\\n***********New LlamaParse+ Basic Query Engine***********\")\n",
+        "print(response)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 11,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "K7RsivpwzQBo",
+        "outputId": "9bcbf62e-250c-46db-f247-e1f293c09bbe"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "We varied the learning\n",
+            "rate over the course of training, according to the formula:\n",
+            "               lrate = d−0.5                                                                          (3)\n",
+            "                         model · min(step_num−0.5, step_num · warmup_steps−1.5)\n",
+            "This corresponds to increasing the learning rate linearly for the first warmup_steps training steps,\n",
+            "and decreasing it thereafter proportionally to the inverse square root of the step number. We used\n",
+            "warmup_steps = 4000.\n",
+            "5.4   Regularization\n",
+            "We employ three types of regularization during training:\n",
+            "                                                    7\n",
+            "---\n",
+            "Table 2: The Transformer achieves better BLEU scores than previous state-of-the-art models on the\n",
+            "English-to-German and English-to-French newstest2014 tests at a fraction of the training cost.\n",
+            "        Model                                          BLEU               Training Cost (FLOPs)\n",
+            "                                                 EN-DE      EN-FR          EN-DE         EN-FR\n",
+            "        ByteNet [18]                              23.75\n",
+            "        Deep-Att + PosUnk [39]                               39.2                       1.0 · 1020\n",
+            "        GNMT + RL [38]                            24.6       39.92        2.3 · 1019    1.4 · 1020\n",
+            "        ConvS2S [9]                               25.16      40.46        9.6 · 1018    1.5 · 1020\n",
+            "        MoE [32]                                  26.03      40.56        2.0 · 1019    1.2 · 1020\n",
+            "        Deep-Att + PosUnk Ensemble [39]                      40.4                       8.0 · 1020\n",
+            "        GNMT + RL Ensemble [38]                   26.30      41.16        1.8 · 1020    1.1 · 1021\n",
+            "        ConvS2S Ensemble [9]                      26.36      41.29        7.7 · 1019    1.2 · 1021\n",
+            "        Transformer (base model)                  27.3       38.1               3.3 · 1018\n",
+            "        Transformer (big)                         28.4       41.8                2.3 · 1019\n",
+            "Residual Dropout       We apply dropout [33] to the output of each sub-layer, before it is added to the\n",
+            "sub-layer input and normalized. In addition, we apply dropout to the sums of the embeddings and the\n",
+            "positional encodings in both the encoder and decoder stacks. For the base model, we use a rate of\n",
+            "Pdrop = 0.1.\n",
+            "Label Smoothing        During training, we employed label smoothing of value ϵls = 0.1 [36]. This\n",
+            "hurts perplexity, as the model learns to be more unsure, but improves accuracy and BLEU score.\n",
+            "6    Results\n",
+            "6.1   Machine Translation\n",
+            "On the WMT 2014 English-to-German translation task, the big transformer model (Transformer (big)\n",
+            "in Table 2) outperforms the best previously reported models (including ensembles) by more than 2.0\n",
+            "BLEU, establishing a new state-of-the-art BLEU score of 28.4. The configuration of this model is\n",
+            "listed in the bottom line of Table 3. Training took 3.5 days on 8 P100 GPUs. Even our base model\n",
+            "surpasses all previously published models and ensembles, at a fraction of the training cost of any of\n",
+            "the competitive models.\n",
+            "On the WMT 2014 English-to-French translation task, our big model achieves a BLEU score of 41.0,\n",
+            "outperforming all of the previously published single models, at less than 1/4 the training cost of the\n",
+            "previous state-of-the-art model. The Transformer (big) model trained for English-to-French used\n",
+            "dropout rate Pdrop = 0.1, instead of 0.3.\n",
+            "For the base models, we used a single model obtained by averaging the last 5 checkpoints, which\n",
+            "were written at 10-minute intervals. For the big models, we averaged the last 20 checkpoints. We\n",
+            "used beam search with a beam size of 4 and length penalty α = 0.6 [38]. These hyperparameters\n",
+            "were chosen after experimentation on the development set. We set the maximum output length during\n",
+            "inference to input length + 50, but terminate early when possible [38].\n",
+            "Table 2 summarizes our results and compares our translation quality and training costs to other model\n",
+            "architectures from the literature.\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Take a look at one of the source nodes from the response\n",
+        "print(response.source_nodes[0].get_content())"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    }
+  ],
+  "metadata": {
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "display_name": "anthropic_env",
+      "language": "python",
+      "name": "anthropic_env"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.11.3"
+    },
+    "vscode": {
+      "interpreter": {
+        "hash": "b0fa6594d8f4cbf19f97940f81e996739fb7646882a419484c72d19e05852a7e"
+      }
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}

llama_parse/base.py

@@ -110,17 +110,53 @@ class Language(str, Enum):
 
 SUPPORTED_FILE_TYPES = [
     ".pdf",
-    ".xml"
+    ".602",
+    ".abw",
+    ".cgm",
+    ".cwk",
     ".doc",
     ".docx",
-    ".pptx",
-    ".rtf",
-    ".pages",
+    ".docm",
+    ".dot",
+    ".dotm",
+    ".hwp",
     ".key",
-    ".epub"
+    ".lwp",
+    ".mw",
+    ".mcw",
+    ".pages",
+    ".pbd",
+    ".ppt",
+    ".pptm",
+    ".pptx",
+    ".pot",
+    ".potm",
+    ".potx",
+    ".rtf",
+    ".sda",
+    ".sdd",
+    ".sdp",
+    ".sdw",
+    ".sgl",
+    ".sti",
+    ".sxi",
+    ".sxw",
+    ".stw",
+    ".sxg",
+    ".txt",
+    ".uof",
+    ".uop",
+    ".uot",
+    ".vor",
+    ".wpd", 
+    ".wps",
+    ".xml",
+    ".zabw",
+    ".epub",
+    ".htm",
+    ".html"
 ]
 
-
 class LlamaParse(BasePydanticReader):
     """A smart-parser for files."""