gradio interface with haystack

Data Flow Diagram.excalidraw

@@ -0,0 +1,1338 @@
+{
+  "type": "excalidraw",
+  "version": 2,
+  "source": "https://excalidraw.com",
+  "elements": [
+    {
+      "type": "rectangle",
+      "version": 353,
+      "versionNonce": 412064510,
+      "isDeleted": false,
+      "id": "b7pxkDczcchwIbszJLfD4",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 367.7243303571429,
+      "y": 428.81696428571433,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 245,
+      "height": 246,
+      "seed": 1822494526,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "RPXwlWir1Xn59OZgYVQ3H"
+        },
+        {
+          "id": "1VyagIK3JnMhHshHeoj6Y",
+          "type": "arrow"
+        },
+        {
+          "id": "EK8UJFziH75SqI7uiNjuw",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 379,
+      "versionNonce": 772414434,
+      "isDeleted": false,
+      "id": "RPXwlWir1Xn59OZgYVQ3H",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 414.4423555646624,
+      "y": 446.81696428571433,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 151.56394958496094,
+      "height": 210,
+      "seed": 1135617790,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "fontSize": 28,
+      "fontFamily": 1,
+      "text": "User Data\n\nPDFs\nDOCX\nPPTX\nTEX",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "b7pxkDczcchwIbszJLfD4",
+      "originalText": "User Data\n\nPDFs\nDOCX\nPPTX\nTEX",
+      "lineHeight": 1.25,
+      "baseline": 200
+    },
+    {
+      "type": "rectangle",
+      "version": 309,
+      "versionNonce": 606047038,
+      "isDeleted": false,
+      "id": "3JMJqMfpTKLeo3zoOKNFW",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 920.8481522817463,
+      "y": 421.70424107142844,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 355,
+      "height": 248,
+      "seed": 1065192190,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "XCJgLVM6VIUBYBhM5nHVM"
+        },
+        {
+          "id": "1VyagIK3JnMhHshHeoj6Y",
+          "type": "arrow"
+        },
+        {
+          "id": "-JpT91ZVqlWSE3rCOtneJ",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 396,
+      "versionNonce": 35536351,
+      "isDeleted": false,
+      "id": "XCJgLVM6VIUBYBhM5nHVM",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 952.916206480965,
+      "y": 458.20424107142844,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 290.8638916015625,
+      "height": 175,
+      "seed": 1375887358,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686152296772,
+      "link": null,
+      "locked": false,
+      "fontSize": 28,
+      "fontFamily": 1,
+      "text": "Preprocessor\nExtracts metadata \nfrom documents\n\nApache Tika",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "3JMJqMfpTKLeo3zoOKNFW",
+      "originalText": "Preprocessor\nExtracts metadata from documents\n\nApache Tika",
+      "lineHeight": 1.25,
+      "baseline": 165
+    },
+    {
+      "type": "arrow",
+      "version": 665,
+      "versionNonce": 684217278,
+      "isDeleted": false,
+      "id": "1VyagIK3JnMhHshHeoj6Y",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 621.7727131931823,
+      "y": 552.6077510793499,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 289.5850660920373,
+      "height": 1.4030689601490849,
+      "seed": 184780478,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "b7pxkDczcchwIbszJLfD4",
+        "focus": 0.0015636720616594639,
+        "gap": 9.048382836039423
+      },
+      "endBinding": {
+        "elementId": "3JMJqMfpTKLeo3zoOKNFW",
+        "focus": -0.07378316395119473,
+        "gap": 9.490372996526617
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          289.5850660920373,
+          1.4030689601490849
+        ]
+      ]
+    },
+    {
+      "type": "rectangle",
+      "version": 304,
+      "versionNonce": 1701203746,
+      "isDeleted": false,
+      "id": "wt4iNAGGyjNR3CuIkVH9N",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 824.4450644841272,
+      "y": 340.5541294642855,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 1140.2583705357147,
+      "height": 848.6277901785713,
+      "seed": 144761826,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "rectangle",
+      "version": 575,
+      "versionNonce": 332667618,
+      "isDeleted": false,
+      "id": "J22jlqkSc0rEm9yDMgCuT",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 1488.056183345734,
+      "y": 780.3056466238836,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 420.7855902777774,
+      "height": 346.9097222222222,
+      "seed": 765672318,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "2hwXZXG8bFrKNwxnVUQP4"
+        },
+        {
+          "id": "-0MJWgPI0_rWhzy-t8Ns6",
+          "type": "arrow"
+        },
+        {
+          "id": "1aqouuarBWLBwIpKAQfLf",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 614,
+      "versionNonce": 1463352382,
+      "isDeleted": false,
+      "id": "2hwXZXG8bFrKNwxnVUQP4",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 1578.6230126032751,
+      "y": 866.2605077349947,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 239.6519317626953,
+      "height": 175,
+      "seed": 1609044158,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "fontSize": 28,
+      "fontFamily": 1,
+      "text": "Vector Database\n\nMilvus\nPgvector\nElasticsearch",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "J22jlqkSc0rEm9yDMgCuT",
+      "originalText": "Vector Database\n\nMilvus\nPgvector\nElasticsearch",
+      "lineHeight": 1.25,
+      "baseline": 165
+    },
+    {
+      "type": "rectangle",
+      "version": 325,
+      "versionNonce": 617630306,
+      "isDeleted": false,
+      "id": "6yls6E8tQxE7macw6vhxQ",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 888.1119869171628,
+      "y": 810.1126902746778,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 420,
+      "height": 334,
+      "seed": 151127038,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "9em--BMfFgH2xeGATfyo4"
+        },
+        {
+          "id": "zewGru4dg7-hvs0WH2K1T",
+          "type": "arrow"
+        },
+        {
+          "id": "-0MJWgPI0_rWhzy-t8Ns6",
+          "type": "arrow"
+        },
+        {
+          "id": "1aqouuarBWLBwIpKAQfLf",
+          "type": "arrow"
+        },
+        {
+          "id": "-JpT91ZVqlWSE3rCOtneJ",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 481,
+      "versionNonce": 117015742,
+      "isDeleted": false,
+      "id": "9em--BMfFgH2xeGATfyo4",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 897.8560665070065,
+      "y": 837.1126902746778,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 400.5118408203125,
+      "height": 280,
+      "seed": 538166434,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "fontSize": 28,
+      "fontFamily": 1,
+      "text": "Encoders (Creates Index for\nthe Database)\n\nSBERT (semantic tf-er)\nFAISS (in mem)\nHNSW\nANN (Annoy, in mem)\nBM25",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "6yls6E8tQxE7macw6vhxQ",
+      "originalText": "Encoders (Creates Index for the Database)\n\nSBERT (semantic tf-er)\nFAISS (in mem)\nHNSW\nANN (Annoy, in mem)\nBM25",
+      "lineHeight": 1.25,
+      "baseline": 270
+    },
+    {
+      "type": "text",
+      "version": 192,
+      "versionNonce": 824613346,
+      "isDeleted": false,
+      "id": "kqrjouCXz6SGF9fSBYqBg",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 872.2247101314483,
+      "y": 358.0536624968997,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 160.23594665527344,
+      "height": 45,
+      "seed": 711131170,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "fontSize": 36,
+      "fontFamily": 1,
+      "text": "Haystack",
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "containerId": null,
+      "originalText": "Haystack",
+      "lineHeight": 1.25,
+      "baseline": 32
+    },
+    {
+      "type": "arrow",
+      "version": 994,
+      "versionNonce": 1301409086,
+      "isDeleted": false,
+      "id": "-JpT91ZVqlWSE3rCOtneJ",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 1085.9694477489766,
+      "y": 681.8462648742615,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 1.1966632701874005,
+      "height": 115.56747980878049,
+      "seed": 582495870,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "3JMJqMfpTKLeo3zoOKNFW",
+        "focus": 0.07712329987309789,
+        "gap": 12.142023802833023
+      },
+      "endBinding": {
+        "elementId": "6yls6E8tQxE7macw6vhxQ",
+        "focus": -0.042909296806605254,
+        "gap": 12.698945591635834
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          1.1966632701874005,
+          115.56747980878049
+        ]
+      ]
+    },
+    {
+      "type": "rectangle",
+      "version": 279,
+      "versionNonce": 2126319010,
+      "isDeleted": false,
+      "id": "iyXbiLaMXYBy-HtmWBvOo",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 402.479422433036,
+      "y": 945.8490028986851,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 169,
+      "height": 115,
+      "seed": 1955233698,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "OYaABKgvwa_k0LiljW5Yb"
+        },
+        {
+          "id": "zewGru4dg7-hvs0WH2K1T",
+          "type": "arrow"
+        },
+        {
+          "id": "4PQgww3CeUMS8WpVB9aYk",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 321,
+      "versionNonce": 563975550,
+      "isDeleted": false,
+      "id": "OYaABKgvwa_k0LiljW5Yb",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 438.03744397844616,
+      "y": 980.8490028986851,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 97.88395690917969,
+      "height": 45,
+      "seed": 495139170,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883822,
+      "link": null,
+      "locked": false,
+      "fontSize": 36,
+      "fontFamily": 1,
+      "text": "Query",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "iyXbiLaMXYBy-HtmWBvOo",
+      "originalText": "Query",
+      "lineHeight": 1.25,
+      "baseline": 32
+    },
+    {
+      "type": "arrow",
+      "version": 892,
+      "versionNonce": 2122986942,
+      "isDeleted": false,
+      "id": "zewGru4dg7-hvs0WH2K1T",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 580.6782222057445,
+      "y": 994.3966234246425,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 300.2850219382026,
+      "height": 1.8820099431195558,
+      "seed": 1899198114,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "iyXbiLaMXYBy-HtmWBvOo",
+        "gap": 9.198799772708469,
+        "focus": -0.16438005876920447
+      },
+      "endBinding": {
+        "elementId": "6yls6E8tQxE7macw6vhxQ",
+        "gap": 7.148742773215697,
+        "focus": -0.12195444661637078
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          300.2850219382026,
+          1.8820099431195558
+        ]
+      ]
+    },
+    {
+      "type": "arrow",
+      "version": 1069,
+      "versionNonce": 782994722,
+      "isDeleted": false,
+      "id": "-0MJWgPI0_rWhzy-t8Ns6",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 1319.5066044569376,
+      "y": 939.000448305401,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 157.3723657110604,
+      "height": 2.984683418077111,
+      "seed": 1876722146,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "6yls6E8tQxE7macw6vhxQ",
+        "focus": -0.2527547454942667,
+        "gap": 11.3946175397748
+      },
+      "endBinding": {
+        "elementId": "J22jlqkSc0rEm9yDMgCuT",
+        "focus": 0.04267875575963368,
+        "gap": 11.177213177736007
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          157.3723657110604,
+          2.984683418077111
+        ]
+      ]
+    },
+    {
+      "type": "arrow",
+      "version": 1078,
+      "versionNonce": 226020862,
+      "isDeleted": false,
+      "id": "1aqouuarBWLBwIpKAQfLf",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 1472.1963245060938,
+      "y": 958.3225976216577,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 151.57340552498817,
+      "height": 1.2806783171292864,
+      "seed": 1918084286,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "J22jlqkSc0rEm9yDMgCuT",
+        "focus": -0.03292613251715128,
+        "gap": 15.859858839640196
+      },
+      "endBinding": {
+        "elementId": "6yls6E8tQxE7macw6vhxQ",
+        "focus": -0.13006015555653785,
+        "gap": 12.510932063942846
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          -151.57340552498817,
+          -1.2806783171292864
+        ]
+      ]
+    },
+    {
+      "type": "text",
+      "version": 76,
+      "versionNonce": 1036097762,
+      "isDeleted": false,
+      "id": "MPkXu4Tmso68ovoKBHl46",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 214.6854538690477,
+      "y": 872.1880095951138,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 584.3878173828125,
+      "height": 45,
+      "seed": 205870526,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "fontSize": 36,
+      "fontFamily": 1,
+      "text": "Retrieval (QnA, Semantic Search)",
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "containerId": null,
+      "originalText": "Retrieval (QnA, Semantic Search)",
+      "lineHeight": 1.25,
+      "baseline": 32
+    },
+    {
+      "type": "text",
+      "version": 199,
+      "versionNonce": 728391230,
+      "isDeleted": false,
+      "id": "Qgdqs9vNKTunwwTouWhb0",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 367.7602306547619,
+      "y": 357.70884292844715,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 271.2239074707031,
+      "height": 45,
+      "seed": 1449124542,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "fontSize": 36,
+      "fontFamily": 1,
+      "text": "Data Ingesting",
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "containerId": null,
+      "originalText": "Data Ingesting",
+      "lineHeight": 1.25,
+      "baseline": 32
+    },
+    {
+      "type": "rectangle",
+      "version": 160,
+      "versionNonce": 1570778274,
+      "isDeleted": false,
+      "id": "k2CnsPZWHyeDkXgJrsBlr",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": -270.83038039434496,
+      "y": 394.2065294053817,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 435.6138392857142,
+      "height": 699.4363839285713,
+      "seed": 1309730942,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "Uj-mc_MUovWhwyFHNYovn"
+        },
+        {
+          "id": "EK8UJFziH75SqI7uiNjuw",
+          "type": "arrow"
+        },
+        {
+          "id": "4PQgww3CeUMS8WpVB9aYk",
+          "type": "arrow"
+        }
+      ],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false
+    },
+    {
+      "type": "text",
+      "version": 118,
+      "versionNonce": 911240895,
+      "isDeleted": false,
+      "id": "Uj-mc_MUovWhwyFHNYovn",
+      "fillStyle": "hachure",
+      "strokeWidth": 1,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": -215.16741216750347,
+      "y": 608.9247213696674,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 324.28790283203125,
+      "height": 270,
+      "seed": 648440126,
+      "groupIds": [],
+      "roundness": null,
+      "boundElements": [],
+      "updated": 1686152362705,
+      "link": null,
+      "locked": false,
+      "fontSize": 36,
+      "fontFamily": 1,
+      "text": "Desktop/Web\nApplication\n\nCan be any cross \nplatform or web \nframework",
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "containerId": "k2CnsPZWHyeDkXgJrsBlr",
+      "originalText": "Desktop/Web\nApplication\n\nCan be any cross platform or web framework",
+      "lineHeight": 1.25,
+      "baseline": 257
+    },
+    {
+      "type": "arrow",
+      "version": 327,
+      "versionNonce": 1945520226,
+      "isDeleted": false,
+      "id": "EK8UJFziH75SqI7uiNjuw",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 179.0803338913695,
+      "y": 561.2444942991496,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 178.55561755952363,
+      "height": 2.644754493833375,
+      "seed": 199703422,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "k2CnsPZWHyeDkXgJrsBlr",
+        "focus": -0.507623074443353,
+        "gap": 14.296875000000227
+      },
+      "endBinding": {
+        "elementId": "b7pxkDczcchwIbszJLfD4",
+        "focus": -0.03860838532454731,
+        "gap": 10.088378906249773
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          178.55561755952363,
+          -2.644754493833375
+        ]
+      ]
+    },
+    {
+      "type": "arrow",
+      "version": 62,
+      "versionNonce": 599418558,
+      "isDeleted": false,
+      "id": "4PQgww3CeUMS8WpVB9aYk",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "angle": 0,
+      "x": 175.28569103422657,
+      "y": 1002.6973746892165,
+      "strokeColor": "#000000",
+      "backgroundColor": "transparent",
+      "width": 218.84858630952377,
+      "height": 2.2294232810793346,
+      "seed": 776191394,
+      "groupIds": [],
+      "roundness": {
+        "type": 2
+      },
+      "boundElements": [],
+      "updated": 1686123883823,
+      "link": null,
+      "locked": false,
+      "startBinding": {
+        "elementId": "k2CnsPZWHyeDkXgJrsBlr",
+        "focus": 0.72867463147048,
+        "gap": 10.502232142857338
+      },
+      "endBinding": {
+        "elementId": "iyXbiLaMXYBy-HtmWBvOo",
+        "focus": -0.04324158834028842,
+        "gap": 8.34514508928578
+      },
+      "lastCommittedPoint": null,
+      "startArrowhead": null,
+      "endArrowhead": "arrow",
+      "points": [
+        [
+          0,
+          0
+        ],
+        [
+          218.84858630952377,
+          2.2294232810793346
+        ]
+      ]
+    },
+    {
+      "id": "XLD_CHZjm1VGui922OvT3",
+      "type": "text",
+      "x": -240.00684175037122,
+      "y": -135.90442669580824,
+      "width": 880.9190418243402,
+      "height": 100.96728752682807,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "#ffc9c9",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 74490257,
+      "version": 206,
+      "versionNonce": 1379886431,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152291537,
+      "link": null,
+      "locked": false,
+      "text": "Possible Architectures",
+      "fontSize": 80.77383002146237,
+      "fontFamily": 1,
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "baseline": 71.00000000000011,
+      "containerId": null,
+      "originalText": "Possible Architectures",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "cknJXduAcHHVgiokCQJl9",
+      "type": "text",
+      "x": -200.62402925037122,
+      "y": 186.5179365854417,
+      "width": 840.3793682734175,
+      "height": 73.11718750000001,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "#ffc9c9",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 417436191,
+      "version": 124,
+      "versionNonce": 347582833,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152322739,
+      "link": null,
+      "locked": false,
+      "text": "Serverless API, cloud service",
+      "fontSize": 58.49375000000004,
+      "fontFamily": 1,
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "baseline": 50.99999999999996,
+      "containerId": null,
+      "originalText": "Serverless API, cloud service",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "p6fDcBnze9bghfsVNikUp",
+      "type": "text",
+      "x": -155.9755917503711,
+      "y": 1301.1263350229415,
+      "width": 616.8109197616585,
+      "height": 67.6328125000001,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "#ffc9c9",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 2031929649,
+      "version": 71,
+      "versionNonce": 377734911,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152340018,
+      "link": null,
+      "locked": false,
+      "text": "Fully Native Application",
+      "fontSize": 54.10625000000007,
+      "fontFamily": 1,
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "baseline": 47.000000000000014,
+      "containerId": null,
+      "originalText": "Fully Native Application",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "9nD5Kmsc66eWPTCtuItxf",
+      "type": "rectangle",
+      "x": -193.686529250371,
+      "y": 1410.2996748666915,
+      "width": 1415.1093749999995,
+      "height": 1010.7343750000001,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "seed": 920662609,
+      "version": 93,
+      "versionNonce": 1891433041,
+      "isDeleted": false,
+      "boundElements": [],
+      "updated": 1686152577220,
+      "link": null,
+      "locked": false
+    },
+    {
+      "id": "aMcwXbbSlBsMXY7WDb1WP",
+      "type": "rectangle",
+      "x": -140.0693417503711,
+      "y": 1540.8465498666912,
+      "width": 479,
+      "height": 332,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "seed": 239065919,
+      "version": 92,
+      "versionNonce": 874992721,
+      "isDeleted": false,
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "H94xIN-iTX0TKY4Fdqqr7"
+        }
+      ],
+      "updated": 1686152548582,
+      "link": null,
+      "locked": false
+    },
+    {
+      "id": "H94xIN-iTX0TKY4Fdqqr7",
+      "type": "text",
+      "x": -117.10928926013673,
+      "y": 1661.8465498666912,
+      "width": 433.07989501953125,
+      "height": 90,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 1975898687,
+      "version": 93,
+      "versionNonce": 65246527,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152548583,
+      "link": null,
+      "locked": false,
+      "text": "MAUI for Native Cross \nPlatform",
+      "fontSize": 36,
+      "fontFamily": 1,
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "baseline": 77,
+      "containerId": "aMcwXbbSlBsMXY7WDb1WP",
+      "originalText": "MAUI for Native Cross Platform",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "M5wxK_BMQQ4Xi63hZpgBT",
+      "type": "text",
+      "x": -74.88965425037111,
+      "y": 1460.0340498666915,
+      "width": 281.77191162109375,
+      "height": 45,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 177592977,
+      "version": 15,
+      "versionNonce": 1726060465,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152399643,
+      "link": null,
+      "locked": false,
+      "text": ".NET Framework",
+      "fontSize": 36,
+      "fontFamily": 1,
+      "textAlign": "left",
+      "verticalAlign": "top",
+      "baseline": 32,
+      "containerId": null,
+      "originalText": ".NET Framework",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "ueSFRdb3iaD6B8oxX-_m5",
+      "type": "rectangle",
+      "x": 539.8525332496286,
+      "y": 1945.3699873666908,
+      "width": 545,
+      "height": 357,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "seed": 1468510751,
+      "version": 189,
+      "versionNonce": 1669326705,
+      "isDeleted": false,
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "Wjjh4s9oU_xhIrWtWIFxF"
+        }
+      ],
+      "updated": 1686153281266,
+      "link": null,
+      "locked": false
+    },
+    {
+      "id": "Wjjh4s9oU_xhIrWtWIFxF",
+      "type": "text",
+      "x": 602.886590866816,
+      "y": 2056.369987366691,
+      "width": 418.931884765625,
+      "height": 135,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 12769375,
+      "version": 304,
+      "versionNonce": 433114641,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686153429461,
+      "link": null,
+      "locked": false,
+      "text": "FAISSSharp/FAISSMask\nHNSW\nSBert ONNX model",
+      "fontSize": 36,
+      "fontFamily": 1,
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "baseline": 122,
+      "containerId": "ueSFRdb3iaD6B8oxX-_m5",
+      "originalText": "FAISSSharp/FAISSMask\nHNSW\nSBert ONNX model",
+      "lineHeight": 1.25
+    },
+    {
+      "id": "oVpSnp_-z_Q3QbkS6_y_G",
+      "type": "rectangle",
+      "x": 519.0556582496286,
+      "y": 1522.8465498666915,
+      "width": 613.3984374999998,
+      "height": 335.9453125,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": {
+        "type": 3
+      },
+      "seed": 987409713,
+      "version": 61,
+      "versionNonce": 36928945,
+      "isDeleted": false,
+      "boundElements": [
+        {
+          "type": "text",
+          "id": "xxaEPY9T0qU1Y0Tj-SH5c"
+        }
+      ],
+      "updated": 1686152551280,
+      "link": null,
+      "locked": false
+    },
+    {
+      "id": "xxaEPY9T0qU1Y0Tj-SH5c",
+      "type": "text",
+      "x": 564.7369326636909,
+      "y": 1600.8192061166915,
+      "width": 522.035888671875,
+      "height": 180,
+      "angle": 0,
+      "strokeColor": "#1e1e1e",
+      "backgroundColor": "transparent",
+      "fillStyle": "hachure",
+      "strokeWidth": 4,
+      "strokeStyle": "solid",
+      "roughness": 1,
+      "opacity": 100,
+      "groupIds": [],
+      "roundness": null,
+      "seed": 1960999999,
+      "version": 123,
+      "versionNonce": 1358775871,
+      "isDeleted": false,
+      "boundElements": null,
+      "updated": 1686152605639,
+      "link": null,
+      "locked": false,
+      "text": "Tikaondotnet\n\nTika binding for .NET using\nIKVM (Java binding for .NET)",
+      "fontSize": 36,
+      "fontFamily": 1,
+      "textAlign": "center",
+      "verticalAlign": "middle",
+      "baseline": 167,
+      "containerId": "oVpSnp_-z_Q3QbkS6_y_G",
+      "originalText": "Tikaondotnet\n\nTika binding for .NET using\nIKVM (Java binding for .NET)",
+      "lineHeight": 1.25
+    }
+  ],
+  "appState": {
+    "gridSize": null,
+    "viewBackgroundColor": "#ffffff"
+  },
+  "files": {}
+}

main.py

@@ -0,0 +1,55 @@
+from haystack.nodes import PreProcessor, PDFToTextConverter, EmbeddingRetriever
+from haystack.document_stores import InMemoryDocumentStore
+from haystack.pipelines import DocumentSearchPipeline
+import gradio as gr
+
+preprocessor = PreProcessor(
+    clean_empty_lines=True,
+    clean_whitespace=True,
+    clean_header_footer=True,
+    split_by="word",
+    split_length=100,
+    split_respect_sentence_boundary=True,
+    split_overlap=3
+)
+document_store = InMemoryDocumentStore()
+retriever = EmbeddingRetriever(document_store=document_store, embedding_model="sentence-transformers/all-MiniLM-L6-v2")
+pipeline = DocumentSearchPipeline(retriever)
+
+def print_answers(results):
+    fields = ["answer", "score"] # "context"
+    answers = results["answers"]
+    filtered_answers = []
+    for ans in answers:
+        filtered_ans = {
+            field: getattr(ans, field) for field in fields if getattr(ans, field) is not None
+        }
+        filtered_answers.append(filtered_ans)
+    return filtered_answers
+
+def write_pdf(pdf_file):
+    converter = PDFToTextConverter(remove_numeric_tables=True, valid_languages=["en"])
+    document = converter.convert(file_path=pdf_file.name, meta=None)[0]
+    preprocessed_docs = preprocessor.process(document)
+    document_store.write_documents(preprocessed_docs)
+
+def predict(question, pdf_file):
+    print("Start processing pdf")
+    write_pdf(pdf_file)
+    print("Processing done.")
+    result = pipeline.run(query=question, params={"Retriever": { "top_k": 2 }})
+    answers = print_answers(result)
+    return answers
+
+title = "Search"
+interface = gr.Interface(
+    fn=predict,
+    inputs=[gr.components.Textbox(lines = 3, label="Ask an open question!"),gr.components.File(file_count="single", type="file", label="Upload a pdf")],
+    outputs="text",
+    title=title, 
+    flagging_options=["top", "medium", "bad"],
+    interpretation="default",
+    theme="default" # “default", “huggingface", “dark-grass", “peach"
+)
+
+interface.launch()