Add files using upload-large-folder tool

.gitattributes

@@ -41,3 +41,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 02[[:space:]]-[[:space:]]Java[[:space:]]Basics/002[[:space:]]JDK[[:space:]]Installation.mp4 filter=lfs diff=lfs merge=lfs -text
 02[[:space:]]-[[:space:]]Java[[:space:]]Basics/006[[:space:]]Comments[[:space:]]in[[:space:]]Java[[:space:]]source[[:space:]]code.mp4 filter=lfs diff=lfs merge=lfs -text
 02[[:space:]]-[[:space:]]Java[[:space:]]Basics/005[[:space:]]Integrated[[:space:]]Development[[:space:]]Environment,[[:space:]]Eclipse[[:space:]]Overview.mp4 filter=lfs diff=lfs merge=lfs -text
+03[[:space:]]-[[:space:]]Primitive[[:space:]]Data[[:space:]]Types,[[:space:]]Variables[[:space:]]and[[:space:]]Arrays/002[[:space:]]Number[[:space:]]Systems.mp4 filter=lfs diff=lfs merge=lfs -text
+03[[:space:]]-[[:space:]]Primitive[[:space:]]Data[[:space:]]Types,[[:space:]]Variables[[:space:]]and[[:space:]]Arrays/001[[:space:]]Primitive[[:space:]]types[[:space:]]and[[:space:]]variables.mp4 filter=lfs diff=lfs merge=lfs -text
+04[[:space:]]-[[:space:]]Eclipse[[:space:]]Tips[[:space:]]and[[:space:]]Tricks/001[[:space:]]Packages[[:space:]]creation[[:space:]]and[[:space:]]package[[:space:]]presentation[[:space:]]in[[:space:]]eclipse.mp4 filter=lfs diff=lfs merge=lfs -text
+04[[:space:]]-[[:space:]]Eclipse[[:space:]]Tips[[:space:]]and[[:space:]]Tricks/002[[:space:]]Plugins[[:space:]]how[[:space:]]to[[:space:]]install[[:space:]]free[[:space:]]plugins,[[:space:]]eclipse[[:space:]]marketplace,[[:space:]]workspace[[:space:]]styles.mp4 filter=lfs diff=lfs merge=lfs -text
+03[[:space:]]-[[:space:]]Primitive[[:space:]]Data[[:space:]]Types,[[:space:]]Variables[[:space:]]and[[:space:]]Arrays/003[[:space:]]Arrays.mp4 filter=lfs diff=lfs merge=lfs -text
+04[[:space:]]-[[:space:]]Eclipse[[:space:]]Tips[[:space:]]and[[:space:]]Tricks/003[[:space:]]Code[[:space:]]Refactoring[[:space:]]in[[:space:]]Eclipse.mp4 filter=lfs diff=lfs merge=lfs -text

03 - Primitive Data Types, Variables and Arrays/001 Primitive types and variables.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:561a6b594ddeaa00fe253d34a8a186fdf63b624a23e20eba0106920185101c30
+size 239598856

03 - Primitive Data Types, Variables and Arrays/002 Number Systems.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48d80e1dc858892612a7e404de526c9f27f66b64e7449bc53b3be51a679e8078
+size 79001924

03 - Primitive Data Types, Variables and Arrays/003 Arrays.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b77919792ce45c953609f44ad7e2fb3edd63a74592a170b8e3ce7f6ea7bb078
+size 232461248

04 - Eclipse Tips and Tricks/001 Packages creation and package presentation in eclipse.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f938bf3c404ffdcdcb67392d77fb15a9c06b46d1ec1ea328a339d384ed455f6c
+size 9925338

04 - Eclipse Tips and Tricks/002 Plugins how to install free plugins, eclipse marketplace, workspace styles.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff9cf3072fb19331601a1e9e218a75fc917ae312ed20e490261faf4a1771015b
+size 15216101

04 - Eclipse Tips and Tricks/003 Code Refactoring in Eclipse.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:738dd29139e1e5c7bbe8eb718c996b9ac7a1920ac4eb0a57bbb86ecd0ac9dc30
+size 89152283

10 - Methods in Java/004 Variable Length Arguments_en.srt

@@ -0,0 +1,176 @@
+1
+00:00:00,000 --> 00:00:00,000
+Hello team.
+
+2
+00:00:00,000 --> 00:00:04,000
+Today we will discuss what are variable length arguments in Java.
+
+3
+00:00:04,000 --> 00:00:06,000
+Let's start from the problem statement.
+
+4
+00:00:06,000 --> 00:00:11,000
+You want to create method, but you don't know how much arguments will be passed to that method.
+
+5
+00:00:11,000 --> 00:00:16,000
+There can be different business cases, but let's simplify and pretend that you need to create method
+
+6
+00:00:16,000 --> 00:00:22,000
+which will some numbers, but you don't know how much numbers will be passed to this method.
+
+7
+00:00:22,000 --> 00:00:27,000
+To not create method which returns sum of two numbers, and another method which returns sum of three
+
+8
+00:00:27,000 --> 00:00:31,000
+numbers, and so on, you create one method with variable length arguments.
+
+9
+00:00:31,000 --> 00:00:33,000
+The syntax is simple.
+
+10
+00:00:33,000 --> 00:00:36,000
+You need to specify type of arguments, right?
+
+11
+00:00:36,000 --> 00:00:38,000
+Ellipses and variable name.
+
+12
+00:00:38,000 --> 00:00:41,000
+Inside the method you can treat this variable as array.
+
+13
+00:00:42,000 --> 00:00:46,000
+This syntax will allow you to pass different amount of arguments to the method.
+
+14
+00:00:46,000 --> 00:00:51,000
+Here you can see that I pass five integers and here I pass just one argument.
+
+15
+00:00:51,000 --> 00:00:55,000
+In both cases this method is called and result is returned.
+
+16
+00:00:55,000 --> 00:01:00,000
+You can even change the way how you declare a main method instead of specifying that the main method
+
+17
+00:01:00,000 --> 00:01:06,000
+takes an array of strings, we can say that the main method can work with variable length arguments
+
+18
+00:01:06,000 --> 00:01:10,000
+of type string declaration like this is also possible.
+
+19
+00:01:10,000 --> 00:01:15,000
+Important thing to flag here is that variable length arguments should be declared as last parameter
+
+20
+00:01:15,000 --> 00:01:16,000
+in the method.
+
+21
+00:01:17,000 --> 00:01:22,000
+For example, this method declaration will produce compilation error because the variable argument type
+
+22
+00:01:22,000 --> 00:01:25,000
+int of the method must be the last parameter.
+
+23
+00:01:25,000 --> 00:01:31,000
+Because during runtime, JVM should know when int arguments are finished and strings are started, and
+
+24
+00:01:31,000 --> 00:01:36,000
+when we adjusted the method declaration, everything is fine and there is no compilation error.
+
+25
+00:01:37,000 --> 00:01:41,000
+Variable length argument always should be declared as the last parameter in the method.
+
+26
+00:01:41,000 --> 00:01:47,000
+You already saw real life example of variable length argument, but probably didn't pay enough attention.
+
+27
+00:01:47,000 --> 00:01:50,000
+Do you remember lesson about string formatting?
+
+28
+00:01:50,000 --> 00:01:56,000
+Method format takes variable length arguments as a parameter, because we never know how much format
+
+29
+00:01:56,000 --> 00:01:59,000
+specifiers will be present in string, which we need to format.
+
+30
+00:02:00,000 --> 00:02:04,000
+That's all what I wanted to share with you regarding variable length arguments.
+
+31
+00:02:04,000 --> 00:02:08,000
+Now let's recap what you have learned about methods in this section.
+
+32
+00:02:08,000 --> 00:02:10,000
+You learned how to declare and call methods.
+
+33
+00:02:10,000 --> 00:02:14,000
+You understand what method signature is and how we can overload methods.
+
+34
+00:02:14,000 --> 00:02:20,000
+Now you understand that both reference and primitive types of data in Java are passed by values.
+
+35
+00:02:20,000 --> 00:02:26,000
+During the method invocation, we wrote recursive methods and learn what variable length arguments are.
+
+36
+00:02:26,000 --> 00:02:28,000
+Let's take a look at your homework now.
+
+37
+00:02:29,000 --> 00:02:32,000
+Your homework primarily consists of coding exercises.
+
+38
+00:02:32,000 --> 00:02:38,000
+Now when you know how to write custom methods, how to use loops, arrays, and primitive types of data,
+
+39
+00:02:38,000 --> 00:02:41,000
+our tasks might become a little more complicated.
+
+40
+00:02:41,000 --> 00:02:45,000
+I tried to describe homework with as much details as I could.
+
+41
+00:02:45,000 --> 00:02:47,000
+Try to solve these tasks.
+
+42
+00:02:47,000 --> 00:02:52,000
+I will share with you source code of the solution for these tasks, but try to solve them by yourself
+
+43
+00:02:52,000 --> 00:02:53,000
+first.
+
+44
+00:02:53,000 --> 00:02:56,000
+Thanks a lot for your attention and see you in the next lesson.
+

10 - Methods in Java/005 Find max int in array.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "005 Find max int in array", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program starts and asks user to input integer numbers separated by space.</p></li><li><p>Program creates array object with entered numbers.</p></li><li><p>Program calls specific method which takes int[] as an parameter and returns max value in this array.<br><br>Method should look like this:<br><strong><em>public static int findMaxIntInArray(int[] intArray) {</em></strong></p></li></ol></li></ol><p><strong><em>&lt;write your code here&gt;<br>}</em></strong><br><br></p><ol><li><ol><li><p>Program prints max value from the array to the console.</p></li></ol></li></ol>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\n\nimport org.junit.Test;\n\npublic class Evaluate {\n\t\n\t@Test\n\tpublic void shouldFindMaxIntInArray() {\n\t\tint[] arr = {1, 100, 2, 5, 8 -100};\n\t\tassertEquals(100, FindMaxInt.findMaxIntInArray(arr));\n\t}\n\n}\n"}], "solutions": [{"file_name": "FindMaxInt.java", "content": "import java.util.Arrays;\r\nimport java.util.Scanner;\r\n\r\npublic class FindMaxInt {\r\n\t\r\n\tpublic static void main(String[] args) {\r\n\t\tScanner sc = new Scanner(System.in);\r\n\t\tSystem.out.print(\"Please, enter integer numbers separated by space: \");\r\n\t\tString numbers = sc.nextLine();\r\n\t\tint[] intArray = convertStringArrayToIntArray(numbers.split(\"\\\\s+\"));\r\n\t\tint maxInt = findMaxIntInArray(intArray);\r\n\t\tSystem.out.println(\"*** Initial Array ***\");\r\n\t\tSystem.out.println(Arrays.toString(intArray));\r\n\t\tSystem.out.println(\"*** Max number in array ***\");\r\n\t\tSystem.out.println(maxInt);\r\n\t}\r\n\r\n\t\r\n\tpublic static int findMaxIntInArray(int[] intArray) {\r\n\t\tint max = intArray[0];\r\n\t\tfor (int i : intArray) {\r\n\t\t\tif (i > max) {\r\n\t\t\t\tmax = i;\r\n\t\t\t}\r\n\t\t}\r\n\t\treturn max;\r\n\t}\r\n\r\n\tprivate static int[] convertStringArrayToIntArray(String[] stringArray) {\r\n\t\tint[] intArray = new int[stringArray.length];\r\n\t\tfor (int i = 0; i < stringArray.length; i++) {\r\n\t\t\tintArray[i] = Integer.valueOf(stringArray[i]);\r\n\t\t}\r\n\t\treturn intArray;\r\n\t}\r\n\t\r\n\t// ====================== SOLUTION WITH STREAM API\r\n\t\r\n\tpublic static int findMaxIntInArrayStreamApi(int[] intArray) {\r\n\t\treturn Arrays.stream(intArray).max().getAsInt();\r\n\t}\r\n\r\n\tprivate static int[] convertStringArrayToIntArrayStreamApi(String[] stringArray) {\r\n\t\treturn Arrays.stream(stringArray)\r\n\t\t\t\t.mapToInt(s -> Integer.valueOf(s))\r\n\t\t\t\t.toArray();\r\n\t}\r\n\r\n}\r\n"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/006 Draw empty rectangle.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "006 Draw empty rectangle", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program starts and asks user to input height of the rectangle</p></li><li><p>After user inputs heights of the rectangle, program asks to input width of the rectangle.</p></li><li><p>Program calls specific method which takes two parameters of int type which prints rectangle to the console:<br> ******</p></li></ol><p>&nbsp; &nbsp; &nbsp; &nbsp;&nbsp; *&nbsp; &nbsp; &nbsp; &nbsp; *</p><p>&nbsp; &nbsp; &nbsp;&nbsp;&nbsp; &nbsp;******</p></li></ol><p>In the example above height of the rectangle is 3, the width of the rectangle is 6</p><p><strong><em>N.B.: rectangle is empty inside</em></strong></p>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\n\nimport java.io.ByteArrayInputStream;\nimport java.io.ByteArrayOutputStream;\nimport java.io.InputStream;\nimport java.io.PrintStream;\n\nimport org.junit.After;\nimport org.junit.Before;\nimport org.junit.Test;\n\n\npublic class Evaluate {\nprivate static final String OUTPUT_RESULT_SEPARATOR = \"Please, enter width of rectangle: \";\n\tprivate final ByteArrayOutputStream out = new ByteArrayOutputStream();\n\tprivate final PrintStream originalOut = System.out;\n\t\n\t@Before\n\tpublic void setStreams() {\n\t\tSystem.setOut(new PrintStream(out));\n\t}\n\n\t@After\n\tpublic void restoreInitialStreams() {\n\t\tSystem.setOut(originalOut);\n\t}\n\n\t\n\t@Test\n\tpublic void shouldDrawRectangleEmptyInside() {\n\t\tString input = \"3 6\";\n\t\tInputStream in = new ByteArrayInputStream(input.getBytes());\n\t\tSystem.setIn(in);\n\n\t\tEmptyRectangle.main(new String[] {});\n\t\t\n\t\tString consoleOutput = out.toString().split(OUTPUT_RESULT_SEPARATOR)[1].strip();\n\t\tassertEquals(\"******\" + System.lineSeparator() +\n\t\t\t\t\"*    *\" + System.lineSeparator() +\n\t\t\t\t\"******\", consoleOutput);\n\t}\n\t\n\t@Test\n\tpublic void shouldDrawSquareEmptyInside() {\n\t\tString input = \"4 4\";\n\t\tInputStream in = new ByteArrayInputStream(input.getBytes());\n\t\tSystem.setIn(in);\n\n\t\tEmptyRectangle.main(new String[] {});\n\t\t\n\t\tString consoleOutput = out.toString().split(OUTPUT_RESULT_SEPARATOR)[1].strip();\n\t\tassertEquals(\"****\" + System.lineSeparator() +\n\t\t\t\t\"*  *\" + System.lineSeparator() +\n\t\t\t\t\"*  *\" + System.lineSeparator() +\n\t\t\t\t\"****\", consoleOutput);\n\t}\n\t\n\t@Test\n\tpublic void shouldDrawRectangleNotEmptyInside() {\n\t\tString input = \"2 6\";\n\t\tInputStream in = new ByteArrayInputStream(input.getBytes());\n\t\tSystem.setIn(in);\n\n\t\tEmptyRectangle.main(new String[] {});\n\t\t\n\t\tString consoleOutput = out.toString().split(OUTPUT_RESULT_SEPARATOR)[1].strip();\n\t\tassertEquals(\"******\" + System.lineSeparator() +\n\t\t\t\t\"******\", consoleOutput);\n\t}\n}\n"}], "solutions": [{"file_name": "EmptyRectangle.java", "content": "import java.util.Scanner;\r\n\r\npublic class EmptyRectangle {\r\n\t\r\n\tpublic static void main(String[] args) {\r\n\t\tScanner sc = new Scanner(System.in);\r\n\t\tSystem.out.print(\"Please, enter height of rectangle: \");\r\n\t\tint height = sc.nextInt();\r\n\t\tSystem.out.print(\"Please, enter width of rectangle: \");\r\n\t\tint width = sc.nextInt();\r\n\t\t\r\n\t\tdrawRectangle(height, width);\r\n\t}\r\n\r\n\tpublic static void drawRectangle(int height, int width) {\r\n\t\tfor (int i = 0; i < height; i++) {\r\n\t\t\tfor (int j = 0; j < width; j++) {\r\n\t\t\t\tif (j == 0 || j == width - 1 || i == 0 || i == height -1) {\r\n\t\t\t\t\tSystem.out.print(\"*\");\r\n\t\t\t\t} else {\r\n\t\t\t\t\tSystem.out.print(\" \");\r\n\t\t\t\t}\r\n\t\t\t\t\r\n\t\t\t}\r\n\t\t\tSystem.out.println();\r\n\t\t}\r\n\t}\r\n\r\n}"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/007 Calculate Amount of Words.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "007 Calculate Amount of Words", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program starts and asks user to enter text.</p></li><li><p>Program calls specific function which take one parameter of String type and returns amount of words in the text. <br><br>Method should look like this:<br><strong><em>public static int getWordsAmount(String text) {<br> &lt;write your code here&gt;<br>}</em></strong><br><br></p></li><li><p>Program prints amount of words to the console.</p></li></ol></li></ol>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\n\nimport org.junit.Test;\n\npublic class Evaluate {\n\t@Test\n\tpublic void shouldReturnAmountOfWordsForText() {\n\t\tString inputText = \"some random text! With punctuation, marks...\";\n\t\tassertEquals(6, AmountOfWords.getWordsAmount(inputText));\n\t}\n}\n"}], "solutions": [{"file_name": "AmountOfWords.java", "content": "import java.util.Scanner;\r\n\r\npublic class AmountOfWords {\r\n\r\n\tpublic static void main(String[] args) {\r\n\t\tScanner sc = new Scanner(System.in);\r\n\t\tSystem.out.print(\"Please, enter any text: \");\r\n\t\tString userInput = sc.nextLine();\r\n\t\t\r\n\t\tint amountOfWords = getWordsAmount(userInput);\r\n\t\tSystem.out.println(\"Amount of words in your text: \" + amountOfWords);\r\n\t}\r\n\r\n\tpublic static int getWordsAmount(String text) {\r\n\t\treturn text.split(\"[\\\\p{P}\\\\s]+\").length;\r\n\t}\r\n}"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/008 Filter String array.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "008 Filter String array", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program starts and asks user to enter random words separated by space</p></li><li><p>Program asks user to enter minimum length of string to filter words which were entered</p></li><li><p>Program creates array object from entered words</p></li><li><p>Program calls specific method which takes String[] as a parameter and returns array of strings which contains words that have length more or equal to value specified by user<br><br><strong><em>Method should look like this:<br>public static String[] filterWordsByLength(int minLength, String[] words) {</em></strong></p></li></ol><p><strong><em>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&lt;write your code here&gt;<br>&nbsp; &nbsp; &nbsp; }</em></strong><br><br><br></p><p>c. Program prints filtered array to the console output.</p></li></ol>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\n\nimport java.util.Arrays;\n\nimport org.junit.Test;\n\npublic class Evaluate {\n    @Test\n\tpublic void shouldFilterWordsByMinLength() {\n\t\tString[] input = { \"asd\", \"asdf\", \"as\", \"asdfg\", \"a\" };\n\t\tint minLength = 3;\n\n\t\tassertEquals(\"[asd, asdf, asdfg]\",\n\t\t\t\tArrays.toString(FilterStringArray.filterWordsByLength(minLength, input)));\n\t}\n\n\t@Test\n\tpublic void shouldFilterWordsByMinLength2() {\n\t\tString[] input = { \"asd\", \"asdf\", \"as\", \"asdfg\", \"a\" };\n\t\tint minLength = 2;\n\n\t\tassertEquals(\"[asd, asdf, as, asdfg]\",\n\t\t\t\tArrays.toString(FilterStringArray.filterWordsByLength(minLength, input)));\n\t}\n}\n"}], "solutions": [{"file_name": "FilterStringArray.java", "content": "import java.util.Arrays;\r\nimport java.util.Scanner;\r\n\r\npublic class FilterStringArray {\r\n\r\n\tpublic static void main(String[] args) {\r\n\t\tScanner sc = new Scanner(System.in);\r\n\t\tSystem.out.print(\"Please, enter any words separated by space: \");\r\n\t\tString userInput = sc.nextLine();\r\n\t\tSystem.out.print(\"Please, enter minumum word length to filter words: \");\r\n\t\tint minLength = sc.nextInt();\r\n\t\t\r\n\t\tString[] words = userInput.split(\"\\\\s+\");\r\n\t\tString[] filteredWords = filterWordsByLength(minLength, words);\r\n\t\tSystem.out.println(Arrays.toString(filteredWords));\r\n\t}\r\n\r\n\t\r\n\t\r\n\tpublic static String[] filterWordsByLength(int minLength, String[] words) {\r\n\t\tString[] filteredArray = new String[words.length];\r\n\t\tfor (int i = 0; i < words.length; i++) {\r\n\t\t\tif (words[i].length() >= minLength) {\r\n\t\t\t\tfilteredArray[i] = words[i];\r\n\t\t\t}\r\n\t\t}\r\n\t\t\r\n\t\tfilteredArray = filterNulls(filteredArray);\r\n\t\t\r\n\t\treturn filteredArray;\r\n\t}\r\n\r\n\r\n\r\n\tprivate static String[] filterNulls(String[] arr) {\r\n\t\tint newArraySize = 0;\r\n\t\tfor (String word : arr) {\r\n\t\t\tif (word != null) {\r\n\t\t\t\tnewArraySize++;\r\n\t\t\t}\r\n\t\t}\r\n\t\t\r\n\t\tString[] filteredArray = new String[newArraySize];\r\n\t\t\r\n\t\tint filteredArrayIndex = 0;\r\n\t\tfor (String word : arr) {\r\n\t\t\tif (word != null) {\r\n\t\t\t\tfilteredArray[filteredArrayIndex++] = word;\r\n\t\t\t}\r\n\t\t}\r\n\t\t\r\n\t\treturn filteredArray;\r\n\t}\r\n\r\n\r\n}\r\n"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/013 String Processor.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "013 String Processor", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program contains methods that can process input string according to requirements below</p></li><li><p>You have input string like this:<br><br><strong><em>Login;Name;Email</em></strong></p></li></ol></li></ol><p><strong><em>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;peterson;Chris Peterson;peterson@outlook.com</em></strong></p><p><strong><em>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;james;Derek James;james@gmail.com</em></strong></p><p><strong><em>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;jackson;Walter Jackson;jackson@gmail.com</em></strong></p><p><strong><em>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; gregory;Mike Gregory;gregory@yahoo.com<br><br></em></strong></p><ol><li><ol><li><p>You program has next method:<br><br><strong><em>public static String convert1(String input) {<br> &lt;write your code here&gt;<br>}</em></strong><br><br>Which formats input data like this:<br>peterson ==&gt; peterson@outlook.com</p></li></ol><p>&nbsp; &nbsp; &nbsp; james ==&gt; james@gmail.com</p><p>&nbsp; &nbsp; &nbsp; jackson ==&gt; jackson@gmail.com</p><p>&nbsp; &nbsp; &nbsp;gregory ==&gt; gregory@yahoo.com</p></li></ol><p><br><br></p><ol><li><ol><li><p>You program has next method:<br><br><strong><em>public static String convert2(String input) {<br> &lt;write your code here&gt;<br>}</em></strong><br><br>Which formats input data like this:</p><p>Chris Peterson (email: peterson@outlook.com)</p></li></ol><p>&nbsp; &nbsp; &nbsp; Derek James (email: james@gmail.com)</p><p>&nbsp; &nbsp; &nbsp; Walter Jackson (email: jackson@gmail.com)</p><p>&nbsp; &nbsp; &nbsp; Mike Gregory (email: gregory@yahoo.com)</p></li></ol><p><br></p>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\n\nimport org.junit.Test;\n\npublic class Evaluate {\n  private static final String INPUT_DATA = \"Login;Name;Email\" + System.lineSeparator()\n\t\t\t+ \"peton;Chris Peterso;person@outlook.com\" + System.lineSeparator()\n\t\t\t+ \"jas;Derek Jame;jes@gmail.com\" + System.lineSeparator()\n\t\t\t+ \"jack;Walter Jackso;jkson@gmail.com\" + System.lineSeparator()\n\t\t\t+ \"greg;Mike Gregor;ggory@yahoo.com\";\n\n\t@Test\n\tpublic void shouldFormatConvert1() {\n\t\tassertEquals(\n\t\t\t\t\"peton ==> person@outlook.com\" + System.lineSeparator()\n\t\t\t\t\t\t+ \"jas ==> jes@gmail.com\" + System.lineSeparator()\n\t\t\t\t\t\t+ \"jack ==> jkson@gmail.com\" + System.lineSeparator()\n\t\t\t\t\t\t+ \"greg ==> ggory@yahoo.com\",\n\t\t\t\tStringProcessor.convert1(INPUT_DATA).strip());\n\t}\n\n\t@Test\n\tpublic void shouldFormatConvert2() {\n\t\tassertEquals(\n\t\t\t\t\"Chris Peterso (email: person@outlook.com)\" + System.lineSeparator()\n\t\t\t\t\t\t+ \"Derek Jame (email: jes@gmail.com)\" + System.lineSeparator()\n\t\t\t\t\t\t+ \"Walter Jackso (email: jkson@gmail.com)\"\n\t\t\t\t\t\t+ System.lineSeparator() + \"Mike Gregor (email: ggory@yahoo.com)\",\n\t\t\t\tStringProcessor.convert2(INPUT_DATA).strip());\n\t}\n\n\n}\n"}], "solutions": [{"file_name": "StringProcessor.java", "content": "\r\npublic class StringProcessor {\r\n\t\r\n\t\r\n\tpublic static final String INPUT_DATA = \"Login;Name;Email\" + System.lineSeparator() +\r\n\t\t\t\"peterson;Chris Peterson;peterson@outlook.com\" + System.lineSeparator() +\r\n\t\t\t\"james;Derek James;james@gmail.com\" + System.lineSeparator() +\r\n\t\t\t\"jackson;Walter Jackson;jackson@gmail.com\" + System.lineSeparator() +\r\n\t\t\t\"gregory;Mike Gregory;gregory@yahoo.com\";\r\n\t\r\n\tpublic static void main(String[] args) {\r\n\t\tSystem.out.println(\"===== Convert 1 demo =====\");\r\n\t\tSystem.out.println(convert1(INPUT_DATA));\r\n\t\t\r\n\t\tSystem.out.println(\"===== Convert 2 demo =====\");\r\n\t\tSystem.out.println(convert2(INPUT_DATA));\r\n\t\t\r\n\t}\r\n\t\r\n\tpublic static String convert1(String input) {\r\n\t\tString result = \"\";\r\n\t\tString[] lines = input.split(System.lineSeparator());\r\n\t\tfor (int i = 1; i < lines.length; i++) {\r\n\t\t\tString[] wordsInLine = lines[i].split(\";\");\r\n\t\t\tresult += wordsInLine[0] + \" ==> \" + wordsInLine[2] + System.lineSeparator();\r\n\t\t}\r\n\t\treturn result;\r\n\t\r\n\t}\r\n\t\r\n\t\r\n\tpublic static String convert2(String input) {\r\n\t\tString result = new String();\r\n\t\tString[] lines = input.split(System.lineSeparator());\r\n\t\tfor (int i = 1; i < lines.length; i++) {\r\n\t\t\tString[] wordsInLine = lines[i].split(\";\");\r\n\t\t\tresult += wordsInLine[1] + \" (email: \" + wordsInLine[2] + \")\" + System.lineSeparator();\r\n\t\t}\r\n\t\treturn result;\r\n\t\r\n\t}\r\n\r\n}\r\n"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/014 Turn each first letter in a word to capital.html

@@ -0,0 +1,122 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta http-equiv="X-UA-Compatible" content="IE=edge" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Coding Assignment</title>
+
+        <style>
+            body {
+                font-family: sf pro text, -apple-system, BlinkMacSystemFont, Roboto, segoe ui, Helvetica, Arial,
+                    sans-serif, apple color emoji, segoe ui emoji, segoe ui symbol;
+                font-weight: 400;
+                line-height: 22.4px;
+                font-size: 16px;
+            }
+            p,
+            ul,
+            ol {
+                font-size: 16px;
+                font-weight: 400;
+            }
+            h1,
+            h2,
+            h3,
+            h4,
+            h5,
+            h6 {
+                font-weight: bold;
+            }
+            ul {
+                list-style: none;
+                margin: 0;
+                padding: 0;
+                max-width: none;
+            }
+            .code-snippet {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 90%;
+                padding: 0.2rem 0.4rem;
+            }
+            .code-block {
+                background-color: #fff;
+                color: #b4690e;
+                font-size: 90%;
+            }
+            .black-block {
+                color: #000000;
+            }
+            .italic-text {
+                font-style: italic;
+            }
+        </style>
+    </head>
+
+    <body onload="main()">
+        <h1 id="coding-title"></h1>
+        <div>
+            <h2>Instructions</h2>
+            <div id="coding-instructions"></div>
+        </div>
+        <div>
+            <h2>Test(s)</h2>
+            <div id="coding-tests"></div>
+        </div>
+        <div>
+            <h2>Solution(s)</h2>
+            <div id="coding-solutions"></div>
+        </div>
+
+        <script>
+            const quizData = {"title": "014 Turn each first letter in a word to capital", "hasInstructions": true, "hasTests": true, "hasSolutions": true, "instructions": "<ol><li><p>Implement console program which will meet the following requirements:</p><ol><li><p>Program starts and asks user to enter text</p></li><li><p>Program format text with the next rules:</p><ol><li><p>all characters in word should become lower case</p></li><li><p>the first letter in the word should become upper case</p></li></ol></li><li><p>Program prints result of formatting to console</p></li></ol></li></ol><p>To format text program uses next method:</p><p><strong><em>public static String firstCharToTitleCase(String string) {</em></strong></p><p><strong><em>&nbsp; &nbsp; &nbsp; &lt;write your code here&gt;</em></strong></p><p><strong><em>}</em></strong></p>", "tests": [{"file_name": "Evaluate.java", "content": "import org.junit.Test;\nimport org.junit.Assert;\nimport com.udemy.ucp.*;\nimport static org.junit.Assert.assertEquals;\nimport org.junit.Test;\n\npublic class Evaluate {\n    private static final String INPUT_DATA = \"When\\tI   was younger\" + System.lineSeparator()\n\t+ \"I never\\t  needed\";\n\t\n\t@Test\n\tpublic void shouldFormatString() {\n\t\tassertEquals(\"When\tI   Was Younger\" + System.lineSeparator() +\n\t\t\t\t\"I Never\t  Needed\", FirstCharCapital.firstCharToTitleCase(INPUT_DATA));\n\t}\n}\n"}], "solutions": [{"file_name": "FirstCharCapital.java", "content": "import java.util.Scanner;\r\n\r\npublic class FirstCharCapital {\r\n\r\n\t\r\n\r\n\tpublic static void main(String[] args) {\r\n\t\tScanner sc = new Scanner(System.in);\r\n\t\tSystem.out.print(\"Please, enter any text: \");\r\n\t\tString userInput = sc.nextLine();\r\n\t\t\r\n\t\t\r\n\t\tSystem.out.println(firstCharToTitleCase(userInput));\r\n\t}\r\n\r\n\tpublic static String firstCharToTitleCase(String string) {\r\n\t\tchar[] chars = string.toLowerCase().toCharArray();\r\n\t\tboolean found = false;\r\n\t\tfor (int i = 0; i < chars.length; i++) {\r\n\t\t\tif (!found && Character.isLetter(chars[i])) {\r\n\t\t\t\tchars[i] = Character.toUpperCase(chars[i]);\r\n\t\t\t\tfound = true;\r\n\t\t\t} else if (Character.isWhitespace(chars[i])) {\r\n\t\t\t\tfound = false;\r\n\t\t\t}\r\n\t\t}\r\n\t\treturn String.valueOf(chars);\r\n\t}\r\n}\r\n"}]};
+
+            function renderCodeList(rootElement, codeList, className, titlePrefix) {
+                for (var i = 0; i < codeList.length; i++) {
+                    var elem = codeList[i];
+                    var jsElem = document.createElement("div");
+                    jsElem.className = className;
+                    var jsElemTitle = document.createElement("h3");
+                    jsElemTitle.innerHTML = titlePrefix + " " + (i + 1);
+                    var jsElemBody = document.createElement("code");
+                    jsElemBody.className = "code-block black-block";
+                    jsElemBody.innerHTML = "<pre>" + elem.content + "</pre>";
+                    jsElem.appendChild(jsElemTitle);
+                    jsElem.appendChild(jsElemBody);
+                    rootElement.appendChild(jsElem);
+                }
+            }
+
+            function main() {
+                // display the assignment
+                var codingTitle = document.getElementById("coding-title");
+                codingTitle.innerHTML = quizData.title;
+
+                var codingInstructions = document.getElementById("coding-instructions");
+                if (quizData.hasInstructions) {
+                    codingInstructions.innerHTML = quizData.instructions;
+                } else {
+                    codingInstructions.innerHTML = '<span class="italic-text">' + quizData.instructions + "</span>";
+                }
+
+                // display the test(s)
+                var codingTests = document.getElementById("coding-tests");
+                if (!quizData.hasTests) {
+                    codingTests.innerHTML = '<span class="italic-text">' + quizData.tests + "</span>";
+                } else {
+                    renderCodeList(codingTests, quizData.tests, "coding-test", "Test");
+                }
+
+                // display the solution(s)
+                var codingSolutions = document.getElementById("coding-solutions");
+                if (!quizData.hasSolutions) {
+                    codingSolutions.innerHTML = '<span class="italic-text">' + quizData.solutions + "</span>";
+                } else {
+                    renderCodeList(codingSolutions, quizData.solutions, "coding-solution", "Solution");
+                }
+            }
+        </script>
+    </body>
+</html>

10 - Methods in Java/016 Homework review Methods_en.srt

@@ -0,0 +1,64 @@
+1
+00:00:00,000 --> 00:00:05,000
+Hello, Tim, I believe you did a great job and solve all coding exercises from your homework.
+
+2
+00:00:05,000 --> 00:00:10,000
+But if you don't mind, I'd like to share with you my own solutions to the same tasks.
+
+3
+00:00:10,000 --> 00:00:13,000
+I will leave links to the source code attached to this lesson.
+
+4
+00:00:14,000 --> 00:00:19,000
+Take your time to look through my solutions and left comments in the source code to make you understand
+
+5
+00:00:19,000 --> 00:00:21,000
+the most complicated pieces of code.
+
+6
+00:00:21,000 --> 00:00:26,000
+Also, in some solutions you can find something like this solution with stream API.
+
+7
+00:00:26,000 --> 00:00:31,000
+That is because the first solution, which I wrote for some tasks, was written with the help of stream
+
+8
+00:00:31,000 --> 00:00:31,000
+API.
+
+9
+00:00:32,000 --> 00:00:37,000
+After that, I remembered that we didn't learn stream API with you so far, and I wrote solution with
+
+10
+00:00:37,000 --> 00:00:39,000
+knowledge that you have right now.
+
+11
+00:00:39,000 --> 00:00:44,000
+Indeed, writing code with stream API takes less code to achieve the same goal.
+
+12
+00:00:45,000 --> 00:00:48,000
+Here is the solution without stream API v3.
+
+13
+00:00:50,000 --> 00:00:52,000
+And here is the solution with stream API.
+
+14
+00:00:53,000 --> 00:00:59,000
+Take this as motivation that soon will learn stream API and you will be able to write methods like this.
+
+15
+00:00:59,000 --> 00:01:04,000
+Feel free to ask questions and comment section regarding homework in case something is still unclear.
+
+16
+00:01:05,000 --> 00:01:08,000
+Thanks a lot for your attention and see you in the next lesson.
+

10 - Methods in Java/016 Solution-Calculate-amount-of-words.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/AmountOfWords.java

10 - Methods in Java/016 Solution-Convert-decimal-to-Roman-numbers-and-vice-versa.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/ConvertDecimalToRoman.java

10 - Methods in Java/016 Solution-Draw-rectanagle-empty-inside.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/EmptyRectangle.java

10 - Methods in Java/016 Solution-Extend-array.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/ArrayExtension.java

10 - Methods in Java/016 Solution-Filter-String-array-to-leave-words-no-less-than-specific-length.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FilterStringArray.java

10 - Methods in Java/016 Solution-Find-max-integer-in-array.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FindMaxInt.java

10 - Methods in Java/016 Solution-Greates-common-divisor.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/GreatestCommonDivisor.java

10 - Methods in Java/016 Solution-Rotate-Matrix.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/MatrixRotation.java

10 - Methods in Java/016 Solution-String-processor.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/StringProcessor.java

10 - Methods in Java/016 Solution-Sum-digits-in-a-number.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/SumDigitsInNumber.java

10 - Methods in Java/016 Solution-Turn-each-first-char-in-the-word-to-uppercase.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FirstCharCapital.java

10 - Methods in Java/017 Quiz Methods in Java - Check yourself.html

@@ -0,0 +1,69 @@
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8" />
+        <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+        <title>Quiz Methods in Java - Check yourself</title>
+
+        <style>
+            * {
+                box-sizing: border-box;
+                margin: 0;
+                padding: 0;
+            }
+            body {
+                font-family: var(--font-stack-text);
+                font-weight: 400;
+                line-height: 1.4;
+                font-size: 1.6rem;
+                color: #2d2f31;
+            }
+            .container {
+                position: relative;
+                height: 100%;
+                overflow-y: auto;
+            }
+            .content {
+                padding: 3.2rem 4.8rem;
+                word-break: break-word;
+                max-width: 69.6rem;
+                margin: 0 auto;
+            }
+            .heading {
+                margin-bottom: 24px;
+                font-family: -apple-system, BlinkMacSystemFont, Roboto, "Segoe UI", Helvetica, Arial, sans-serif,
+                    "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";
+                font-weight: 700;
+                line-height: 1.2;
+                letter-spacing: 0;
+                font-size: 32px;
+                max-width: 36em;
+            }
+            .article-asset-container {
+                padding: 2.4rem;
+            }
+            .article-asset-container p {
+                font-size: 19px;
+            }
+            code {
+                background-color: #fff;
+                border: 1px solid #d1d7dc;
+                color: #b4690e;
+                font-size: 80%;
+                padding: 0.2rem 0.4rem;
+                font-family: sfmono-regular, Consolas, liberation mono, Menlo, Courier, monospace;
+            }
+            p {
+                font-weight: 400;
+            }
+        </style>
+    </head>
+    <body>
+        <div class="container">
+            <div class="content">
+                <div class="heading">Quiz Methods in Java - Check yourself</div>
+                <div class="article-asset-container"><p><strong>Quiz Link: </strong><a href="https://forms.gle/a5SSyMsRVVL5ijaB9" rel="noopener noreferrer" target="_blank"><strong>https://forms.gle/a5SSyMsRVVL5ijaB9</strong></a></p><p>The quiz doesn’t require you to log in or submit any personal information. Your data privacy is my priority.</p><p>The Quiz covers content from the following lessons:</p><ul><li><p>Methods in Java: Overview</p></li><li><p>Parameter Passing Mechanism in Java</p></li><li><p>Recursive methods</p></li><li><p>Variable Length Arguments</p></li></ul><p><strong>Quiz Instructions:</strong></p><p>This quiz is designed to help you check your understanding of the content from the previous lessons. All questions are based only on material covered in the lessons you've already watched, so there’s no need to worry about unfamiliar topics.</p><ul><li><p>Each question has one correct answer. Select the best answer and click "Submit" when you're ready.</p></li><li><p>After submitting, you'll see your score. For each question, you earn 1 point.</p></li><li><p>Click "View Results" to see the explanations for each correct answer. Reviewing these explanations can help reinforce your understanding.</p></li><li><p>If you score below 70%, I recommend revisiting the previous lessons. However, if the explanations make sense and you’re confident in your understanding, feel free to move on.</p></li><li><p>If any explanations are unclear, please don’t hesitate to post questions in the Q&amp;A section—I’m here to help!</p></li></ul></div>
+            </div>
+        </div>
+    </body>
+</html>

10 - Methods in Java/external-links.txt

@@ -0,0 +1,48 @@
+
+001 Source-code-of-the-example-used-in-the-lesson
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/MethodsDemo.java
+
+002 Source-code-of-the-example-used-in-the-lesson
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/PassByValueDemo.java
+
+003 Source-code-of-the-example-used-in-the-lesson
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/RecursiveMethodsDemo.java
+
+004 Source-code-of-the-example-used-in-the-lesson
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/VarargsDemo.java
+
+004 Methods-in-Java-Homework
+https://docs.google.com/document/d/1ojb3UJGOZC14ULGXyrIonSWeBDKbGo5WKsDns4_crgo/edit?usp=sharing
+
+016 Solution-Calculate-amount-of-words
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/AmountOfWords.java
+
+016 Solution-Convert-decimal-to-Roman-numbers-and-vice-versa
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/ConvertDecimalToRoman.java
+
+016 Solution-Extend-array
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/ArrayExtension.java
+
+016 Solution-Draw-rectanagle-empty-inside
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/EmptyRectangle.java
+
+016 Solution-Filter-String-array-to-leave-words-no-less-than-specific-length
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FilterStringArray.java
+
+016 Solution-Find-max-integer-in-array
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FindMaxInt.java
+
+016 Solution-Turn-each-first-char-in-the-word-to-uppercase
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/FirstCharCapital.java
+
+016 Solution-Greates-common-divisor
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/GreatestCommonDivisor.java
+
+016 Solution-Rotate-Matrix
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/MatrixRotation.java
+
+016 Solution-String-processor
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/StringProcessor.java
+
+016 Solution-Sum-digits-in-a-number
+https://github.com/AndriiPiatakha/learnit_java_core/blob/master/src/com/itbulls/learnit/javacore/methods/hw/SumDigitsInNumber.java

100 - GPT + Slack + Jira + Gmail Integration/001 GPT + Slack + Jira Integration Work with Jira Datasource_en.srt

@@ -0,0 +1,1792 @@
+1
+00:00:05,000 --> 00:00:08,000
+Followed him by this moment in our course.
+
+2
+00:00:08,000 --> 00:00:10,000
+I assume that you already learned.
+
+3
+00:00:10,000 --> 00:00:12,000
+Your API in the previous lessons.
+
+4
+00:00:12,000 --> 00:00:16,000
+So now it is time to start implementation of Jira.
+
+5
+00:00:16,000 --> 00:00:22,000
+API integration with our web application that automates project management operations.
+
+6
+00:00:22,000 --> 00:00:31,000
+Namely, we are going to build integration between Jira, API, GP2, API, Slack and our web application.
+
+7
+00:00:31,000 --> 00:00:38,000
+I'm going to teach you how to connect Jira as a separate data source that you will use for gathering
+
+8
+00:00:38,000 --> 00:00:39,000
+required context.
+
+9
+00:00:39,000 --> 00:00:44,000
+In scope of this lesson, we will not post data to the Jira data source.
+
+10
+00:00:44,000 --> 00:00:48,000
+In other words, we will not create items there.
+
+11
+00:00:48,000 --> 00:00:54,000
+We will just read from Jira how to post data to Jira Data source.
+
+12
+00:00:54,000 --> 00:00:56,000
+We are going to learn in the next lessons.
+
+13
+00:00:57,000 --> 00:01:03,000
+I will show you on real examples how you can implement function calling feature to fetch required data
+
+14
+00:01:03,000 --> 00:01:06,000
+items to add them to the context.
+
+15
+00:01:06,000 --> 00:01:12,000
+By the end of this lesson, you are supposed to have completely functional and already useful application,
+
+16
+00:01:12,000 --> 00:01:19,000
+and I'm going to show you the demo of how you can use it as we will make an overview of the source code
+
+17
+00:01:19,000 --> 00:01:20,000
+line by line.
+
+18
+00:01:20,000 --> 00:01:26,000
+I will make an architecture overview and will share with you the best practices of web development.
+
+19
+00:01:26,000 --> 00:01:28,000
+Let's start our lesson.
+
+20
+00:01:29,000 --> 00:01:35,000
+If you follow the course and you don't skip any lessons, then you have enough theoretical knowledge
+
+21
+00:01:35,000 --> 00:01:41,000
+and practical skills to follow the live demo session that I'm going to hold in this lesson.
+
+22
+00:01:42,000 --> 00:01:47,000
+Today, we are going to have a lesson full of practical examples, and I will walk you through my solution
+
+23
+00:01:47,000 --> 00:01:52,000
+line by line, explaining you how we achieved this result.
+
+24
+00:01:52,000 --> 00:01:58,000
+Let's start from review of the result that we are going to achieve by the end of this lesson.
+
+25
+00:01:58,000 --> 00:02:04,000
+I believe that in this way you are going to be motivated to watch the lesson till the end and you will
+
+26
+00:02:04,000 --> 00:02:10,000
+have a clear picture in the head of what we are going to achieve when we will get to the source code.
+
+27
+00:02:10,000 --> 00:02:16,000
+Watching source code and keeping in the head the image of the end result will help you to stay focused,
+
+28
+00:02:16,000 --> 00:02:21,000
+concentrated and will help you to get answers to your questions.
+
+29
+00:02:22,000 --> 00:02:25,000
+I opened Slack and our imaginary case.
+
+30
+00:02:25,000 --> 00:02:29,000
+This is our team chat where we collaborate.
+
+31
+00:02:29,000 --> 00:02:35,000
+Imagine that during the conversation any participant of the chat would like to clarify more details
+
+32
+00:02:35,000 --> 00:02:42,000
+about the current progress of work, remaining items to do, or other information related to the product
+
+33
+00:02:42,000 --> 00:02:49,000
+that we work on with Jira Integration and GPT analysis capabilities, we can get the answers that we
+
+34
+00:02:49,000 --> 00:02:50,000
+expect to get.
+
+35
+00:02:50,000 --> 00:02:57,000
+I believe that you watched a lesson about function calling feature in ChatGPT because it is important
+
+36
+00:02:57,000 --> 00:02:58,000
+to understand the lesson.
+
+37
+00:02:58,000 --> 00:02:59,000
+Examples.
+
+38
+00:02:59,000 --> 00:03:02,000
+I will just remind you quickly what it is about.
+
+39
+00:03:02,000 --> 00:03:10,000
+We give to the GPT the array of functions descriptions to tell it what functions we have on our end
+
+40
+00:03:10,000 --> 00:03:13,000
+and which functions we can call in case it would be needed.
+
+41
+00:03:14,000 --> 00:03:20,000
+And GPT makes decision based on the context whether some function should be called and result of the
+
+42
+00:03:20,000 --> 00:03:22,000
+function should be provided as an input.
+
+43
+00:03:23,000 --> 00:03:29,000
+So basically in our today's demo, GPT will understand that we need to make a call to Jira API in order
+
+44
+00:03:29,000 --> 00:03:31,000
+to fetch some additional information.
+
+45
+00:03:31,000 --> 00:03:37,000
+And after providing requested information from Jira, GPT will provide an answer.
+
+46
+00:03:37,000 --> 00:03:41,000
+So imagine that in our work chat during the conversation.
+
+47
+00:03:41,000 --> 00:03:48,000
+Somebody needs to get additional information in order to make data driven decision and to save some
+
+48
+00:03:48,000 --> 00:03:54,000
+time on opening a Jira project, building a query, analyzing the data, I can simply ask my bot to
+
+49
+00:03:54,000 --> 00:03:57,000
+provide me with some information.
+
+50
+00:03:57,000 --> 00:04:01,000
+For example, I ask general question, something like this.
+
+51
+00:04:01,000 --> 00:04:10,000
+GPT How many work items do we have in Jira and receives a response that in Jira project there are six
+
+52
+00:04:10,000 --> 00:04:18,000
+work items and it doesn't matter for our bot whether it is six work items or 600 work items.
+
+53
+00:04:19,000 --> 00:04:26,000
+The only limit that we have in this scenario is the context token limit on GPT side, and in my previous
+
+54
+00:04:26,000 --> 00:04:33,000
+videos about GPT, we already learned that there are different GPT models with different context token
+
+55
+00:04:33,000 --> 00:04:34,000
+limit size.
+
+56
+00:04:34,000 --> 00:04:41,000
+So it is question of selecting the proper GPT model and applying GPT best practices about working with
+
+57
+00:04:41,000 --> 00:04:44,000
+context that we also learned in previous lessons.
+
+58
+00:04:44,000 --> 00:04:50,000
+Let me open Jira board in order to make sure that what provided me with correct response.
+
+59
+00:04:51,000 --> 00:04:57,000
+Of course, when I work with my team on production data, I don't check it each time I receive response
+
+60
+00:04:57,000 --> 00:04:58,000
+from the bot.
+
+61
+00:04:58,000 --> 00:05:03,000
+But right now let's perform some checks just for the sake of the demo.
+
+62
+00:05:03,000 --> 00:05:12,000
+And as you can see on our Jira board, we have also six work items, so all is good and bot's response
+
+63
+00:05:12,000 --> 00:05:13,000
+was correct.
+
+64
+00:05:13,000 --> 00:05:15,000
+Let's ask some other questions.
+
+65
+00:05:15,000 --> 00:05:21,000
+For example, GPT, how many work items and repertoire has assigned now?
+
+66
+00:05:21,000 --> 00:05:23,000
+And we receive response.
+
+67
+00:05:23,000 --> 00:05:24,000
+Is it in repertoire?
+
+68
+00:05:24,000 --> 00:05:32,000
+Currently has three work items assigned in Jira and we can open Jira board one more time just to check
+
+69
+00:05:32,000 --> 00:05:37,000
+and make sure that our bot doesn't lie us and we can see that.
+
+70
+00:05:37,000 --> 00:05:41,000
+And repertoire has only three work items assigned.
+
+71
+00:05:41,000 --> 00:05:41,000
+Cool.
+
+72
+00:05:42,000 --> 00:05:44,000
+Let's ask the next question.
+
+73
+00:05:45,000 --> 00:05:52,000
+GPT How many work items was the type Bach we have now in not completed status?
+
+74
+00:05:53,000 --> 00:06:00,000
+And it replies us that we have just two bugs will not check each response in order to have enough time
+
+75
+00:06:00,000 --> 00:06:02,000
+to show you more examples.
+
+76
+00:06:02,000 --> 00:06:08,000
+I already showed you our board and you can trust me that these responses are correct.
+
+77
+00:06:08,000 --> 00:06:11,000
+Or just go back and check the board one more time.
+
+78
+00:06:12,000 --> 00:06:16,000
+And in the meantime, let's continue and ask the next question.
+
+79
+00:06:16,000 --> 00:06:21,000
+GPT Can you group these bugs by status, please?
+
+80
+00:06:21,000 --> 00:06:25,000
+And we received breakdown of our box by their statuses.
+
+81
+00:06:26,000 --> 00:06:33,000
+One defect is in Todo status, one is in progress status and one bug is in done status.
+
+82
+00:06:34,000 --> 00:06:38,000
+And we can continue conversation in this chat with our team members.
+
+83
+00:06:38,000 --> 00:06:43,000
+And the coolest thing about this is that I don't use any GCL.
+
+84
+00:06:44,000 --> 00:06:50,000
+I mean Jira query language, I use conversational language like I would ask any Scrum master.
+
+85
+00:06:50,000 --> 00:06:51,000
+Okay.
+
+86
+00:06:51,000 --> 00:06:55,000
+In my team to get more information before I continue with the conversation.
+
+87
+00:06:56,000 --> 00:06:58,000
+Okay, let's continue.
+
+88
+00:06:58,000 --> 00:07:00,000
+Let's ask the next question.
+
+89
+00:07:00,000 --> 00:07:02,000
+GPT Check please.
+
+90
+00:07:02,000 --> 00:07:10,000
+Due dates of not completed bugs that you referred to and tell me when is the latest due date?
+
+91
+00:07:11,000 --> 00:07:17,000
+And we can see that the latest due date among not completed books is August 4th, which is logically
+
+92
+00:07:17,000 --> 00:07:18,000
+correct.
+
+93
+00:07:18,000 --> 00:07:21,000
+And we again received information that we need it.
+
+94
+00:07:22,000 --> 00:07:24,000
+The next question sounds like this.
+
+95
+00:07:25,000 --> 00:07:29,000
+GPT And who is the bug with the latest due date assigned to?
+
+96
+00:07:30,000 --> 00:07:33,000
+And we can see that the bug is assigned to Andre Petaja.
+
+97
+00:07:34,000 --> 00:07:38,000
+Usually in responses you can receive different volume of information.
+
+98
+00:07:38,000 --> 00:07:42,000
+It depends on the GPT configurations that we already learned.
+
+99
+00:07:42,000 --> 00:07:48,000
+This example just shows you that you can check some basic information about items that was mentioned
+
+100
+00:07:48,000 --> 00:07:50,000
+previously in the context.
+
+101
+00:07:51,000 --> 00:07:58,000
+You can ask some more interesting questions like for example, the following one GPT Can you remind
+
+102
+00:07:58,000 --> 00:08:06,000
+me of the issue ID please, in the scope of which we planned to implement the persistent card state
+
+103
+00:08:06,000 --> 00:08:08,000
+across multiple browsers for the same account?
+
+104
+00:08:09,000 --> 00:08:14,000
+When I deal with hundreds and sometimes with thousands of tickets on the big programs that I manage,
+
+105
+00:08:14,000 --> 00:08:22,000
+sometimes it is hard to remember specific work item ID of the task that I'm interested in, but still
+
+106
+00:08:22,000 --> 00:08:27,000
+I might need the reference to the ticket in order to check the latest status comments.
+
+107
+00:08:27,000 --> 00:08:28,000
+Szczekociny Seine.
+
+108
+00:08:28,000 --> 00:08:29,000
+ET cetera.
+
+109
+00:08:30,000 --> 00:08:33,000
+And I can ask GPT to help me with this task.
+
+110
+00:08:33,000 --> 00:08:34,000
+Like you saw in the demo.
+
+111
+00:08:35,000 --> 00:08:43,000
+Now I can get navigated to the work item with ID dash four and check information that I need.
+
+112
+00:08:43,000 --> 00:08:50,000
+In this particular case, GPT analyzed the description field in order to find the ID of the task that
+
+113
+00:08:50,000 --> 00:08:52,000
+I am interested in.
+
+114
+00:08:52,000 --> 00:08:58,000
+You can see that in this case the summary of the task doesn't provide us with a lot of details and useful
+
+115
+00:08:58,000 --> 00:08:59,000
+information.
+
+116
+00:08:59,000 --> 00:09:06,000
+But in the description I see that this task is about implementation of card component, ensuring that
+
+117
+00:09:06,000 --> 00:09:12,000
+the state of the card persists in different tabs and in different browsers for the same account.
+
+118
+00:09:12,000 --> 00:09:17,000
+Basically, you can ask any other questions that you may think of.
+
+119
+00:09:17,000 --> 00:09:19,000
+Like for example, GPT.
+
+120
+00:09:19,000 --> 00:09:26,000
+Can you group all items from Jira by assigning you please, and you will be provided with a breakdown
+
+121
+00:09:26,000 --> 00:09:27,000
+by assignee.
+
+122
+00:09:27,000 --> 00:09:34,000
+Please note that we have one work item on the dashboard that does not assigned to anyone.
+
+123
+00:09:34,000 --> 00:09:36,000
+This is done for testing purposes.
+
+124
+00:09:36,000 --> 00:09:43,000
+Of course, in real life you would avoid cases like this, but we can see that in total GPT provided
+
+125
+00:09:43,000 --> 00:09:50,000
+us with details about five tickets because the sixth ticket doesn't have assignee.
+
+126
+00:09:51,000 --> 00:09:55,000
+GPT Can you group all items by issue type, please?
+
+127
+00:09:55,000 --> 00:10:03,000
+Almost the similar kind of task, but in this case we group by issue type and we again receive correct
+
+128
+00:10:03,000 --> 00:10:04,000
+response.
+
+129
+00:10:04,000 --> 00:10:11,000
+These are just few questions that I use on my Scrum Masters use when managing the teamwork and during
+
+130
+00:10:11,000 --> 00:10:12,000
+the collaboration.
+
+131
+00:10:13,000 --> 00:10:19,000
+Of course, if your experienced project delivery manager or delivery director, you can imagine the
+
+132
+00:10:19,000 --> 00:10:23,000
+whole potential power and set of use cases.
+
+133
+00:10:23,000 --> 00:10:28,000
+How you can apply this both to boost collaboration within the team to the next level.
+
+134
+00:10:29,000 --> 00:10:32,000
+So I am done with the demo and with review.
+
+135
+00:10:32,000 --> 00:10:40,000
+Now it is time to understand how all this is implemented and I open my IDE to walk you through the source
+
+136
+00:10:40,000 --> 00:10:40,000
+code.
+
+137
+00:10:40,000 --> 00:10:45,000
+As always, you can find the reference to the solution in attachments to this video.
+
+138
+00:10:46,000 --> 00:10:53,000
+I'm going to make an overview of how I implemented the application in order to achieve the end result
+
+139
+00:10:53,000 --> 00:10:54,000
+that I have just demoed.
+
+140
+00:10:54,000 --> 00:11:01,000
+And in case you would have any questions, you are always welcome to post your questions below the video
+
+141
+00:11:01,000 --> 00:11:03,000
+and I will be happy to answer.
+
+142
+00:11:04,000 --> 00:11:10,000
+So this is the application that I developed from scratch during the previous lessons and share it with
+
+143
+00:11:10,000 --> 00:11:12,000
+you each line.
+
+144
+00:11:12,000 --> 00:11:19,000
+That's why today I wouldn't stop too much on the pieces of code and configurations that we already reviewed
+
+145
+00:11:19,000 --> 00:11:25,000
+in the previous lessons, and I will focus your attention only on things that were changed here.
+
+146
+00:11:25,000 --> 00:11:29,000
+I open Slack integration Controller class here.
+
+147
+00:11:29,000 --> 00:11:34,000
+We get notified about the Slack events we are subscribed for.
+
+148
+00:11:34,000 --> 00:11:41,000
+So far we are subscribed only for the one event that is application mentioning event.
+
+149
+00:11:41,000 --> 00:11:47,000
+We receive it here and we call process on mention event of the slack service.
+
+150
+00:11:47,000 --> 00:11:49,000
+Nothing is changed here.
+
+151
+00:11:49,000 --> 00:11:53,000
+Let me then go to the default slack service.
+
+152
+00:11:53,000 --> 00:11:57,000
+I opened the default slack service class here.
+
+153
+00:11:57,000 --> 00:11:58,000
+I introduced improvement.
+
+154
+00:11:59,000 --> 00:12:05,000
+When we call get answer to single query method, I pass an array of GPT functions.
+
+155
+00:12:05,000 --> 00:12:12,000
+You can see that functions variable refers to the list that I let spring to initialize by putting autopilot
+
+156
+00:12:12,000 --> 00:12:21,000
+annotation like this spring gathered all beans of GPT function type discovered within the context and
+
+157
+00:12:21,000 --> 00:12:27,000
+taking into account get answer to a single query takes variable argument of GPT function type.
+
+158
+00:12:27,000 --> 00:12:35,000
+I need to convert the list to array because as you know in Java variable argument type may be treated
+
+159
+00:12:35,000 --> 00:12:37,000
+as an array within the message.
+
+160
+00:12:37,000 --> 00:12:40,000
+Currently in this list we have two functions.
+
+161
+00:12:40,000 --> 00:12:45,000
+One function is from the lesson when we learned function calling feature in GPT.
+
+162
+00:12:46,000 --> 00:12:53,000
+Do you remember we had GPT was a function that supposed to return us the weather in the specific location.
+
+163
+00:12:53,000 --> 00:12:55,000
+That was an easy example.
+
+164
+00:12:55,000 --> 00:13:02,000
+I decided not to remove it from the source code because this is the learning project and I want to store
+
+165
+00:13:02,000 --> 00:13:05,000
+this example in the repository for future students.
+
+166
+00:13:05,000 --> 00:13:12,000
+And the second function is the main one for today's lesson in the Beans configuration class.
+
+167
+00:13:12,000 --> 00:13:17,000
+At the bottom you can see that I declared GPT Jira issues function.
+
+168
+00:13:17,000 --> 00:13:19,000
+Let's understand what it is.
+
+169
+00:13:20,000 --> 00:13:22,000
+Just in case you skipped function.
+
+170
+00:13:22,000 --> 00:13:23,000
+Call in lesson.
+
+171
+00:13:23,000 --> 00:13:30,000
+Please come back and watch that lesson because understanding function calling in GPT is the key for
+
+172
+00:13:30,000 --> 00:13:32,000
+understanding of this implementation.
+
+173
+00:13:32,000 --> 00:13:36,000
+So function should have name and description.
+
+174
+00:13:36,000 --> 00:13:39,000
+I decided to store this information in properties file.
+
+175
+00:13:40,000 --> 00:13:42,000
+I open my app properties file.
+
+176
+00:13:42,000 --> 00:13:46,000
+The function name is simple get Jira issues function.
+
+177
+00:13:46,000 --> 00:13:53,000
+The most important is the description of the function because based on the description analysis, GPT
+
+178
+00:13:53,000 --> 00:13:57,000
+will make a decision whether it makes sense to call this function or not.
+
+179
+00:13:58,000 --> 00:14:04,000
+Based on the description, you can understand that I have a function with such name that can return
+
+180
+00:14:04,000 --> 00:14:11,000
+all Jira issues of all issue types from the configured Jira project in order to provide information
+
+181
+00:14:11,000 --> 00:14:12,000
+about team members.
+
+182
+00:14:12,000 --> 00:14:20,000
+Workload, current plan, due dates, issue descriptions, summaries, total amount of issues finds
+
+183
+00:14:20,000 --> 00:14:25,000
+the issues of the specific type finds the idea of the issue by its summary or description.
+
+184
+00:14:25,000 --> 00:14:29,000
+Calculate team workload and lots more.
+
+185
+00:14:29,000 --> 00:14:31,000
+Few comments here.
+
+186
+00:14:31,000 --> 00:14:38,000
+Definitely this is learning course, but we learn the principles that may be applied on production scale.
+
+187
+00:14:38,000 --> 00:14:43,000
+For example, in my production application, I have multiple functions created.
+
+188
+00:14:43,000 --> 00:14:49,000
+There might be cases when you need the whole amount of Jira tickets, but there is also might be need
+
+189
+00:14:49,000 --> 00:14:57,000
+to fetch child items of some epic or items that contains only specific label and we don't need the rest
+
+190
+00:14:57,000 --> 00:14:58,000
+of the items.
+
+191
+00:14:58,000 --> 00:15:03,000
+It will help you to optimize performance and costs spent on GPT.
+
+192
+00:15:04,000 --> 00:15:06,000
+Just an idea to think about.
+
+193
+00:15:06,000 --> 00:15:11,000
+In scope of this lesson, we are going to review implementation of this function because all other functions
+
+194
+00:15:11,000 --> 00:15:15,000
+can be implemented exactly in the similar way.
+
+195
+00:15:15,000 --> 00:15:20,000
+The second thing here it is set for configured project.
+
+196
+00:15:20,000 --> 00:15:24,000
+There are different ways of how you can configure project.
+
+197
+00:15:24,000 --> 00:15:30,000
+It all depends on the business requirements and how you want to build interaction with the end user.
+
+198
+00:15:31,000 --> 00:15:37,000
+For example, you can ask end user to provide all necessary configuration during the interaction with
+
+199
+00:15:37,000 --> 00:15:44,000
+the bot, or you can make this configuration as mandatory one during the bot installation or any other
+
+200
+00:15:44,000 --> 00:15:45,000
+option.
+
+201
+00:15:45,000 --> 00:15:51,000
+In our particular case, I configured the project key in the properties a little bit later.
+
+202
+00:15:51,000 --> 00:15:56,000
+In this lesson I'm going to show you where it will be needed and where we are going to use it.
+
+203
+00:15:57,000 --> 00:16:04,000
+So coming back to Bean's configuration class, I used value annotation to inject values from the properties
+
+204
+00:16:04,000 --> 00:16:05,000
+file.
+
+205
+00:16:05,000 --> 00:16:10,000
+Then I set the name and description values into GPT function object.
+
+206
+00:16:10,000 --> 00:16:19,000
+Then important point GPT should identify input arguments for our function and we need to tell GPU what
+
+207
+00:16:19,000 --> 00:16:22,000
+type of arguments with what name we need.
+
+208
+00:16:22,000 --> 00:16:29,000
+But in this specific case we don't need any input arguments because we don't need to filter result or
+
+209
+00:16:29,000 --> 00:16:33,000
+sort result or any other input arguments to work with.
+
+210
+00:16:33,000 --> 00:16:37,000
+We just need to fetch all work items from Jira project.
+
+211
+00:16:37,000 --> 00:16:38,000
+That's it.
+
+212
+00:16:38,000 --> 00:16:46,000
+But still, according to GPT, API parameters attribute is required and in documentation it is said
+
+213
+00:16:46,000 --> 00:16:53,000
+that in case I need to describe a function that accepts no parameters, we need to provide the value
+
+214
+00:16:53,000 --> 00:17:00,000
+object for type property of the parameters object and empty object value for properties attribute.
+
+215
+00:17:00,000 --> 00:17:07,000
+Taking into account that Kazan library doesn't take into consideration fields with no values during
+
+216
+00:17:07,000 --> 00:17:09,000
+the conversion from object to Json format.
+
+217
+00:17:09,000 --> 00:17:16,000
+By default, we need to come up with some workaround in order to meet the requirements of GPT API.
+
+218
+00:17:17,000 --> 00:17:21,000
+So I set object type for parameters object.
+
+219
+00:17:21,000 --> 00:17:29,000
+I did it according to documentation and for properties, I created an object of type no properties that
+
+220
+00:17:29,000 --> 00:17:31,000
+basically doesn't have any state.
+
+221
+00:17:31,000 --> 00:17:36,000
+It is compatible with our marker interface parameter properties.
+
+222
+00:17:36,000 --> 00:17:38,000
+So I can set properties now.
+
+223
+00:17:38,000 --> 00:17:42,000
+Set parameters to the function and function is ready.
+
+224
+00:17:42,000 --> 00:17:44,000
+What happens next?
+
+225
+00:17:44,000 --> 00:17:53,000
+I open default GPT service class inside the get answer to single query method I call prepare request
+
+226
+00:17:53,000 --> 00:17:58,000
+method where I add functions to the GPT request object.
+
+227
+00:17:58,000 --> 00:18:01,000
+Now we are ready to send the request to GPT.
+
+228
+00:18:01,000 --> 00:18:04,000
+Request is sent inside the method.
+
+229
+00:18:04,000 --> 00:18:05,000
+Get response from GPT.
+
+230
+00:18:05,000 --> 00:18:09,000
+Here we have method extract GPT response content.
+
+231
+00:18:10,000 --> 00:18:15,000
+I don't stop on the detailed review of these methods because nothing was changed.
+
+232
+00:18:16,000 --> 00:18:20,000
+If you forgot the implementation, please check previous lessons.
+
+233
+00:18:20,000 --> 00:18:25,000
+The only thing that was changed here is the part related to the function call.
+
+234
+00:18:25,000 --> 00:18:28,000
+Let's review it again from the top.
+
+235
+00:18:28,000 --> 00:18:33,000
+So in case in the GPT response there is a request for function call.
+
+236
+00:18:33,000 --> 00:18:42,000
+We enter this if block and get the reference to the function been using the function name I implemented
+
+237
+00:18:42,000 --> 00:18:44,000
+here factory method pattern.
+
+238
+00:18:45,000 --> 00:18:48,000
+I just did small improvements in the function factory class.
+
+239
+00:18:48,000 --> 00:18:50,000
+Let me show it quickly.
+
+240
+00:18:51,000 --> 00:18:55,000
+I implemented application context aware interface.
+
+241
+00:18:55,000 --> 00:19:01,000
+This is done on purpose in order to use the existing context instead of creating the new one.
+
+242
+00:19:01,000 --> 00:19:08,000
+I just need to implement set, application context, method and initialize application context field.
+
+243
+00:19:08,000 --> 00:19:11,000
+All necessary injections will be done by spring.
+
+244
+00:19:12,000 --> 00:19:13,000
+This is done in order.
+
+245
+00:19:13,000 --> 00:19:20,000
+I can extract the bean from the current application context because my get Jira issues function.
+
+246
+00:19:20,000 --> 00:19:27,000
+Bean contains injections from the properties files and depends on the Jira service bean.
+
+247
+00:19:27,000 --> 00:19:33,000
+So this improvement in the function factory is exactly what we need to do in order to make this solution
+
+248
+00:19:33,000 --> 00:19:35,000
+work as expected.
+
+249
+00:19:35,000 --> 00:19:41,000
+If you want to learn more how it is implemented, we reviewed it in the previous lesson when we learned
+
+250
+00:19:41,000 --> 00:19:43,000
+function calling feature.
+
+251
+00:19:44,000 --> 00:19:51,000
+Once he receives the reference to the function I call execute method in the beans configuration file,
+
+252
+00:19:51,000 --> 00:19:55,000
+you can find the bean with name get Jira issues function.
+
+253
+00:19:55,000 --> 00:19:58,000
+Let's review it and understand how it works.
+
+254
+00:19:59,000 --> 00:20:04,000
+I injected Jira service and into this class in execute method.
+
+255
+00:20:04,000 --> 00:20:08,000
+I use Jira service to extract all Jira issues.
+
+256
+00:20:08,000 --> 00:20:15,000
+I need to make a review of multiple types here Jira service, default, Jira service and Jira issue.
+
+257
+00:20:15,000 --> 00:20:17,000
+So let's go one by one.
+
+258
+00:20:18,000 --> 00:20:19,000
+Jira Service.
+
+259
+00:20:20,000 --> 00:20:23,000
+Let me start an overview from Jira service interface.
+
+260
+00:20:24,000 --> 00:20:29,000
+So this is a high level interface that defines the contract for two masses for now.
+
+261
+00:20:30,000 --> 00:20:38,000
+The first method is get full Json issue by ID, it appears here as part of the refactoring.
+
+262
+00:20:38,000 --> 00:20:45,000
+After lessons about Jira API, I moved the code from the controller to the service and now we can use
+
+263
+00:20:45,000 --> 00:20:53,000
+Jira service bean to extract the Json of any work item by its ID and in general integration controller
+
+264
+00:20:53,000 --> 00:21:02,000
+clause, you can see that get issue by method was refactored to now delegate the call to the Jira service
+
+265
+00:21:02,000 --> 00:21:08,000
+and the code that was here in the previous lesson was moved to the default Jira service.
+
+266
+00:21:09,000 --> 00:21:15,000
+I also did some minor cosmetic changes in the URL and query parameter name.
+
+267
+00:21:15,000 --> 00:21:16,000
+In attachments.
+
+268
+00:21:16,000 --> 00:21:23,000
+You can find the reference to the commit with all changes and you can investigate it and remember that
+
+269
+00:21:23,000 --> 00:21:29,000
+in the case of any questions, please do not hesitate to post your questions below the video and I will
+
+270
+00:21:29,000 --> 00:21:30,000
+be happy to answer.
+
+271
+00:21:31,000 --> 00:21:37,000
+Also, for demo and development purposes, I created here endpoints that allows you to fetch all issues
+
+272
+00:21:37,000 --> 00:21:40,000
+from Jira without posting message to Slack.
+
+273
+00:21:40,000 --> 00:21:47,000
+You can do it by directly accessing this resource and here you can see that get Jira issues method is
+
+274
+00:21:47,000 --> 00:21:49,000
+also invoked.
+
+275
+00:21:49,000 --> 00:21:54,000
+The same method is invoked in the get Jira issues function.
+
+276
+00:21:54,000 --> 00:22:00,000
+So let me open now the default service class and review it together with you.
+
+277
+00:22:00,000 --> 00:22:04,000
+Let's start review from the get Jira issues method.
+
+278
+00:22:05,000 --> 00:22:08,000
+The first thing that I do in this method I build a URL.
+
+279
+00:22:09,000 --> 00:22:17,000
+The URL consists from the Jira API base URL, such resource URL and query parameter with name JKL.
+
+280
+00:22:18,000 --> 00:22:25,000
+JKL stands for Jira Query language so I can pass Gql here to fetch issues that I need.
+
+281
+00:22:26,000 --> 00:22:30,000
+I pass here project name and Max results.
+
+282
+00:22:30,000 --> 00:22:32,000
+I have few comments here.
+
+283
+00:22:33,000 --> 00:22:39,000
+First of all, all these field values are initialized with values from the Myapp properties file.
+
+284
+00:22:39,000 --> 00:22:40,000
+Let's check it.
+
+285
+00:22:41,000 --> 00:22:46,000
+I open my app properties file, I changed version of Jira API to the second one.
+
+286
+00:22:47,000 --> 00:22:53,000
+We reviewed the Jira API versions in the separate lesson and I said that Jira API version two is the
+
+287
+00:22:53,000 --> 00:23:01,000
+most stable one based on today and also it provides the content of description field as one single value,
+
+288
+00:23:01,000 --> 00:23:08,000
+which is easier to parse in the Jira API Version three The content of description field is divided into
+
+289
+00:23:08,000 --> 00:23:09,000
+blocks.
+
+290
+00:23:09,000 --> 00:23:11,000
+It is also available to parse it.
+
+291
+00:23:11,000 --> 00:23:19,000
+But I don't like complicate things when there is an easier option available, so I decided to change
+
+292
+00:23:19,000 --> 00:23:20,000
+it to version two.
+
+293
+00:23:21,000 --> 00:23:27,000
+Then you can find here each resource and search resource and Jira project name too.
+
+294
+00:23:27,000 --> 00:23:29,000
+That is our project key.
+
+295
+00:23:30,000 --> 00:23:34,000
+Probably you might be wondering where to get Project key.
+
+296
+00:23:34,000 --> 00:23:36,000
+Let me show it to you.
+
+297
+00:23:36,000 --> 00:23:40,000
+Basically, you can easily find it in each work item ID.
+
+298
+00:23:41,000 --> 00:23:48,000
+Each ID consists from the short abbreviation dash and number like you can see here.
+
+299
+00:23:49,000 --> 00:23:57,000
+Let dash four or here, let dash five to change the project key or just to find it.
+
+300
+00:23:57,000 --> 00:24:00,000
+You can go to the project settings.
+
+301
+00:24:00,000 --> 00:24:08,000
+So click on the project settings in the left pane and here in Details section, you can check the key
+
+302
+00:24:08,000 --> 00:24:08,000
+field.
+
+303
+00:24:08,000 --> 00:24:14,000
+So just copy it and paste it in your web app configuration.
+
+304
+00:24:14,000 --> 00:24:21,000
+Max Results Property is also configured in the Myapp properties, and as you can see, this equal to
+
+305
+00:24:21,000 --> 00:24:29,000
+100 because according to the documentation, it is max amount of items allowed to return within one
+
+306
+00:24:29,000 --> 00:24:30,000
+single query.
+
+307
+00:24:31,000 --> 00:24:35,000
+But what to do in case there is more than 100 work items in the backlog?
+
+308
+00:24:36,000 --> 00:24:39,000
+Jira API implements pagination approach.
+
+309
+00:24:39,000 --> 00:24:45,000
+Basically, in each response, you receive information about the total amount of records available in
+
+310
+00:24:45,000 --> 00:24:47,000
+scope of the current request.
+
+311
+00:24:47,000 --> 00:24:52,000
+For example, you receive 100 value in the max results attribute.
+
+312
+00:24:53,000 --> 00:24:58,000
+That means that you can count on reading 100 records from the current request.
+
+313
+00:24:59,000 --> 00:25:02,000
+Also, in each request you will receive the total field.
+
+314
+00:25:03,000 --> 00:25:08,000
+This field contains information about total amount of records that match your query.
+
+315
+00:25:09,000 --> 00:25:16,000
+And you also receive start at attribute that tells you from which record the list is started.
+
+316
+00:25:16,000 --> 00:25:25,000
+Usually it is zero here, but during the request you can specify start at attribute and thus receives
+
+317
+00:25:25,000 --> 00:25:27,000
+the next 100 tickets.
+
+318
+00:25:27,000 --> 00:25:34,000
+But be attentive with numbers because indexing is started not from one but from zero.
+
+319
+00:25:34,000 --> 00:25:41,000
+And your homework would be to implement just enough code to handle pagination of Jira API.
+
+320
+00:25:42,000 --> 00:25:43,000
+Write it down.
+
+321
+00:25:43,000 --> 00:25:46,000
+It would be a nice practice for you.
+
+322
+00:25:46,000 --> 00:25:53,000
+So basically you would need to call Jira API multiple times in case total amount of records is higher
+
+323
+00:25:53,000 --> 00:26:00,000
+than value in Max results attribute that actually specifies the amount of records returned.
+
+324
+00:26:01,000 --> 00:26:06,000
+Okay, so it looks like we are done with building the query.
+
+325
+00:26:06,000 --> 00:26:09,000
+Then create a http get request.
+
+326
+00:26:09,000 --> 00:26:16,000
+In this example I use Apache http client library, but you are welcome to use any library you wish.
+
+327
+00:26:17,000 --> 00:26:18,000
+Everything.
+
+328
+00:26:18,000 --> 00:26:24,000
+What happens next is passing the returned Json and namely array with issues.
+
+329
+00:26:24,000 --> 00:26:29,000
+But pay attention that I don't send the GPG as a whole array of issues.
+
+330
+00:26:29,000 --> 00:26:30,000
+Why?
+
+331
+00:26:30,000 --> 00:26:31,000
+Because of multiple reasons.
+
+332
+00:26:32,000 --> 00:26:34,000
+First of all, cost optimization.
+
+333
+00:26:34,000 --> 00:26:41,000
+I don't want to give redundant content that will not help me to find the answer on my questions.
+
+334
+00:26:41,000 --> 00:26:47,000
+So it is up to you to hold the business analysis and define what fields of Jira issues do you really
+
+335
+00:26:47,000 --> 00:26:54,000
+need that would make impact and will help to provide the end user with the answer to the question.
+
+336
+00:26:55,000 --> 00:26:59,000
+The second reason we need to remember about GPT model token limit.
+
+337
+00:27:00,000 --> 00:27:08,000
+I just try to not send redundant tokens and instead I would prefer to leave some space for useful information
+
+338
+00:27:08,000 --> 00:27:10,000
+that I can send to GPT.
+
+339
+00:27:11,000 --> 00:27:14,000
+The third reason I care about performance.
+
+340
+00:27:14,000 --> 00:27:23,000
+Again, I don't want to ask GPT to analyze more data than actually needed, thus will win some time
+
+341
+00:27:23,000 --> 00:27:24,000
+during the data processing.
+
+342
+00:27:25,000 --> 00:27:31,000
+The fourth reason I follow clean architecture principles in the OP design.
+
+343
+00:27:31,000 --> 00:27:40,000
+The idea that I follow here is to have business oriented entities, namely in case API will change.
+
+344
+00:27:40,000 --> 00:27:47,000
+I don't need to refactor the rest of the code in the application because the rest of my code depends
+
+345
+00:27:47,000 --> 00:27:48,000
+on the business entity.
+
+346
+00:27:48,000 --> 00:27:56,000
+Jira issue and in case API will change in the next versions, I would need to adjust only this method
+
+347
+00:27:56,000 --> 00:28:03,000
+in order to pass all required information and data properly and the rest of the code will keep using
+
+348
+00:28:03,000 --> 00:28:06,000
+the Jira issue programming interface.
+
+349
+00:28:07,000 --> 00:28:14,000
+To understand my answer better, you need just explore the details of each Jira issue and understand
+
+350
+00:28:14,000 --> 00:28:18,000
+how it looks like in the API in the lessons about Jira API.
+
+351
+00:28:18,000 --> 00:28:26,000
+Well, when you saw an example of extracting issue by key, as you can see while parsing the Json here,
+
+352
+00:28:26,000 --> 00:28:30,000
+I set the values into Jira issue objects properties.
+
+353
+00:28:30,000 --> 00:28:36,000
+Jira issue is my business entity that contains only that information that is needed according to my
+
+354
+00:28:36,000 --> 00:28:38,000
+business requirements.
+
+355
+00:28:38,000 --> 00:28:40,000
+Let's review the source code of this type.
+
+356
+00:28:41,000 --> 00:28:45,000
+So Jira issue contains such information.
+
+357
+00:28:45,000 --> 00:28:52,000
+Key Assignee Description Summary Status Due Date Project Key.
+
+358
+00:28:52,000 --> 00:28:58,000
+It belongs to project ID Project name Priority Issue Type.
+
+359
+00:28:59,000 --> 00:29:06,000
+And just to let you know that I also made some changes in my Jira and extended standard set of fields
+
+360
+00:29:06,000 --> 00:29:09,000
+with such fields as priority.
+
+361
+00:29:09,000 --> 00:29:11,000
+How to add new fields to Jira.
+
+362
+00:29:11,000 --> 00:29:14,000
+You can learn in the Jira section of this course.
+
+363
+00:29:15,000 --> 00:29:18,000
+Taking into account we already discussed this in the previous lesson.
+
+364
+00:29:18,000 --> 00:29:21,000
+I wouldn't stop on this too much.
+
+365
+00:29:21,000 --> 00:29:27,000
+Everything else in this class is just getters and setters, so nothing special.
+
+366
+00:29:27,000 --> 00:29:30,000
+Let's get back to the default Jira service.
+
+367
+00:29:30,000 --> 00:29:34,000
+So I iterated over each issue in Json.
+
+368
+00:29:34,000 --> 00:29:39,000
+I converted it to the Jira issue and added those to the list.
+
+369
+00:29:39,000 --> 00:29:43,000
+The list of Jira issues is returned from this method.
+
+370
+00:29:43,000 --> 00:29:48,000
+The second method is the one that we already reviewed in the previous lesson.
+
+371
+00:29:48,000 --> 00:29:55,000
+As I already mentioned, I refactored code a bit and moved this code from the Jira integration controller
+
+372
+00:29:55,000 --> 00:29:57,000
+to the default service.
+
+373
+00:29:58,000 --> 00:30:01,000
+Now you know how get Jira issues function works.
+
+374
+00:30:02,000 --> 00:30:10,000
+Once Jira service returns a list of Jira issues, I just need to convert it to the Json because we would
+
+375
+00:30:10,000 --> 00:30:17,000
+need to send this list back to the GP2 in order it could prepare an answer for us based on the whole
+
+376
+00:30:17,000 --> 00:30:18,000
+required data gathered.
+
+377
+00:30:19,000 --> 00:30:22,000
+Let's get back to the default GP2 service.
+
+378
+00:30:22,000 --> 00:30:28,000
+So function got executed and we received the Json response of this function.
+
+379
+00:30:29,000 --> 00:30:31,000
+Amazing what happens next.
+
+380
+00:30:32,000 --> 00:30:35,000
+Then I do the same thing that we already learned in the function call.
+
+381
+00:30:35,000 --> 00:30:43,000
+In lesson I create GP2 message because right now I need to add one more message to the context and then
+
+382
+00:30:43,000 --> 00:30:44,000
+send it back to the GP2.
+
+383
+00:30:45,000 --> 00:30:49,000
+The message should have function role in GP2 API.
+
+384
+00:30:49,000 --> 00:30:53,000
+There is a dedicated role reserved for the function responses.
+
+385
+00:30:54,000 --> 00:30:59,000
+Another interesting message that I implement here is called add message with token limit.
+
+386
+00:30:59,000 --> 00:31:06,000
+Imagine the case when you have a lot of messages already in the context and you fetched 100 zero items,
+
+387
+00:31:06,000 --> 00:31:09,000
+for example, or 200.
+
+388
+00:31:09,000 --> 00:31:16,000
+The main thing is that response that you received and want to add to the context will go over the tokens
+
+389
+00:31:16,000 --> 00:31:20,000
+limit defined by the GPT model selected.
+
+390
+00:31:20,000 --> 00:31:21,000
+Understood.
+
+391
+00:31:21,000 --> 00:31:27,000
+For example, imagine that your model selected supports 4000 tokens.
+
+392
+00:31:27,000 --> 00:31:34,000
+Just an example and you are going to add a response after function call that is 3000 tokens long.
+
+393
+00:31:34,000 --> 00:31:39,000
+But you also already have more than 2000 tokens in the context.
+
+394
+00:31:39,000 --> 00:31:41,000
+What to do in this case.
+
+395
+00:31:41,000 --> 00:31:49,000
+In order to not catch the exception, it is better to remove all the messages from the context and put
+
+396
+00:31:49,000 --> 00:31:52,000
+response after the function call into the context.
+
+397
+00:31:52,000 --> 00:32:00,000
+Agree that the purpose of this message In this method, I check whether the total length of the context,
+
+398
+00:32:00,000 --> 00:32:05,000
+together with the length of the new message that I want to add, will stay within the limits.
+
+399
+00:32:05,000 --> 00:32:11,000
+If no, then I remove from the context the element at zero index position.
+
+400
+00:32:12,000 --> 00:32:16,000
+That would be the oldest message in our Slack conversation.
+
+401
+00:32:16,000 --> 00:32:24,000
+And then I call this message again until we would be able to add the message taking into account we
+
+402
+00:32:24,000 --> 00:32:26,000
+have recursion here in this method.
+
+403
+00:32:26,000 --> 00:32:31,000
+It is super important to define the conditions of breaking the recursion.
+
+404
+00:32:31,000 --> 00:32:38,000
+So in case there are no messages left in the context, that means the size of the list with context
+
+405
+00:32:38,000 --> 00:32:42,000
+messages is equal to zero, then return from the message.
+
+406
+00:32:43,000 --> 00:32:50,000
+Then I set functions to null in order to optimize the costs by saving some more tokens on functions.
+
+407
+00:32:50,000 --> 00:32:51,000
+Description.
+
+408
+00:32:51,000 --> 00:32:58,000
+And then I have another important point in case there is zero messages left in the context.
+
+409
+00:32:58,000 --> 00:33:00,000
+I will not send requests to the GPU.
+
+410
+00:33:01,000 --> 00:33:08,000
+Instead, I will just return error message that tells end user to contact with the administrator because
+
+411
+00:33:08,000 --> 00:33:16,000
+the current model selected run out of tokens limit for this request and request cannot be processed.
+
+412
+00:33:16,000 --> 00:33:22,000
+Definitely you can implement all the logic to handle situations like this.
+
+413
+00:33:22,000 --> 00:33:29,000
+When this case may happen, imagine the case that the length of the function call response from Jira
+
+414
+00:33:29,000 --> 00:33:32,000
+would be 20,000 tokens.
+
+415
+00:33:32,000 --> 00:33:38,000
+Well, for example, it will depend on how you implement your solution and how big your Jira project
+
+416
+00:33:38,000 --> 00:33:39,000
+is.
+
+417
+00:33:39,000 --> 00:33:43,000
+But your model can handle only 4000 tokens.
+
+418
+00:33:43,000 --> 00:33:49,000
+Then our previous method that we have just reviewed will not add anything.
+
+419
+00:33:49,000 --> 00:33:56,000
+After removing all the context messages and in case we removed all messages from the context trying
+
+420
+00:33:56,000 --> 00:34:02,000
+to add the response from the function call, there is no sense to send request to GPU because we already
+
+421
+00:34:02,000 --> 00:34:09,000
+removed original request and our response after function call may also not fit within the context limits.
+
+422
+00:34:10,000 --> 00:34:13,000
+These are limitations that we need to accept.
+
+423
+00:34:13,000 --> 00:34:20,000
+So what I would recommend you to do in this case, if you follow the lesson attentively, you might
+
+424
+00:34:20,000 --> 00:34:27,000
+already hear the answer in the production version of this application that I use with my team to manage
+
+425
+00:34:27,000 --> 00:34:29,000
+our work and to collaborate.
+
+426
+00:34:29,000 --> 00:34:37,000
+I have multiple functions defined that return not all work items, but return only those items that
+
+427
+00:34:37,000 --> 00:34:38,000
+meets criteria.
+
+428
+00:34:38,000 --> 00:34:46,000
+Basically the query to filter on the requested issue is created and I don't need overwhelm the context
+
+429
+00:34:46,000 --> 00:34:51,000
+with data that is not required to answer my question.
+
+430
+00:34:51,000 --> 00:34:59,000
+For example, if I request information about the box, there is no need to fetch epics stories tasks.
+
+431
+00:34:59,000 --> 00:35:05,000
+If I ask question about child items of the specific epic, I don't need to provide GPU with all the
+
+432
+00:35:05,000 --> 00:35:10,000
+tickets that are not related to the specific epic from the conversation history.
+
+433
+00:35:10,000 --> 00:35:12,000
+Do you understand my point?
+
+434
+00:35:12,000 --> 00:35:19,000
+Anyway, in case something is not clear, please let me know about your questions in comments below
+
+435
+00:35:19,000 --> 00:35:22,000
+this video and I will be happy to answer.
+
+436
+00:35:22,000 --> 00:35:30,000
+In our code we send requests to the GPU and receive response and return this response to the end user.
+
+437
+00:35:30,000 --> 00:35:34,000
+That's how it works and we already hold the demo.
+
+438
+00:35:34,000 --> 00:35:38,000
+So I believe you can imagine end to end scenario.
+
+439
+00:35:38,000 --> 00:35:41,000
+Let's recap what we have learned in this lesson.
+
+440
+00:35:42,000 --> 00:35:48,000
+In this lesson, we learned how to integrate Jira Slack and our web application.
+
+441
+00:35:48,000 --> 00:35:52,000
+We implement function calling feature in our application.
+
+442
+00:35:53,000 --> 00:35:59,000
+I hold an end to end demo of web applications that's supposed to boost the collaboration within your
+
+443
+00:35:59,000 --> 00:35:59,000
+team.
+
+444
+00:36:00,000 --> 00:36:07,000
+We reviewed architecture of our application and discussed best practices of web application development
+
+445
+00:36:07,000 --> 00:36:09,000
+and saw how to apply them in practice.
+
+446
+00:36:10,000 --> 00:36:12,000
+That's all what I wanted to share with you in this lesson.
+
+447
+00:36:13,000 --> 00:36:14,000
+Thanks a lot for your attention.
+
+448
+00:36:14,000 --> 00:36:17,000
+Have a great day and see you in the next lesson.
+

100 - GPT + Slack + Jira + Gmail Integration/001 Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/cb46583146e15ddd4782ab32315c14ca55f73edf

100 - GPT + Slack + Jira + Gmail Integration/002 Create-Jira-ticket-from-chat-Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/7853021ac12529fcb1c58f49f6d9495624f8850b

100 - GPT + Slack + Jira + Gmail Integration/002 Generate Tickets in Jira & Send Email from Slack via Chat Interface_en.srt

@@ -0,0 +1,1136 @@
+1
+00:00:05,000 --> 00:00:06,000
+Hello, Tim.
+
+2
+00:00:06,000 --> 00:00:12,000
+In this video we are going to learn a few more use cases of using our board together with a ChatGPT.
+
+3
+00:00:12,000 --> 00:00:18,000
+We don't have any long agenda for today because today we're actually not going to learn new tools,
+
+4
+00:00:18,000 --> 00:00:20,000
+new approaches, new theory.
+
+5
+00:00:20,000 --> 00:00:27,000
+Basically, we are going to use a similar approach that we used in previous lesson, namely using GPT
+
+6
+00:00:27,000 --> 00:00:33,000
+function called feature in order to invoke the correct function based on the context when it will be
+
+7
+00:00:33,000 --> 00:00:33,000
+needed.
+
+8
+00:00:33,000 --> 00:00:39,000
+That's why in this video we are going to review simple but still important and very useful business
+
+9
+00:00:39,000 --> 00:00:40,000
+use cases.
+
+10
+00:00:40,000 --> 00:00:46,000
+They are preparing, generating and send an email using chat interface.
+
+11
+00:00:46,000 --> 00:00:53,000
+In this scenario, we will ask our bot to prepare an email and it will use GPT to generate an email
+
+12
+00:00:53,000 --> 00:01:00,000
+for us and then would ask bot to send an email to a specific user telling us the name of the user and
+
+13
+00:01:00,000 --> 00:01:08,000
+my bot together with GPT, will handle the rest of the things I'm going to show you in this lesson how
+
+14
+00:01:08,000 --> 00:01:09,000
+it works.
+
+15
+00:01:09,000 --> 00:01:16,000
+And I believe that this is really useful scenario to implement because it will save a lot of hours at
+
+16
+00:01:16,000 --> 00:01:16,000
+work.
+
+17
+00:01:16,000 --> 00:01:23,000
+And the second scenario is that we are going to review today is generating a work items in Jira using
+
+18
+00:01:23,000 --> 00:01:25,000
+chat interface.
+
+19
+00:01:25,000 --> 00:01:32,000
+In previous lesson I showed you how to use Jira as data source and get additional context to answer
+
+20
+00:01:32,000 --> 00:01:33,000
+and user requests.
+
+21
+00:01:33,000 --> 00:01:41,000
+Basically, when the user asked our bot about state of the work, we did a request to Jira and added
+
+22
+00:01:41,000 --> 00:01:44,000
+information about work items to the context.
+
+23
+00:01:44,000 --> 00:01:52,000
+So in this lesson I am going to show you how to post data to Jira using our bot and ChatGPT will keep
+
+24
+00:01:52,000 --> 00:01:57,000
+using Jira API for such purposes and function calling mechanism.
+
+25
+00:01:57,000 --> 00:02:03,000
+So technically speaking we are going to use exactly the same approach we already used in previous lesson.
+
+26
+00:02:03,000 --> 00:02:06,000
+The same approach but different end result.
+
+27
+00:02:06,000 --> 00:02:12,000
+Of course, during the course I will not be able physically to cover all possible variety of potential
+
+28
+00:02:12,000 --> 00:02:16,000
+use cases where ChatGPT may be applied.
+
+29
+00:02:16,000 --> 00:02:22,000
+But with this lesson, I just want to give you a few more examples of how you can build powerful application
+
+30
+00:02:22,000 --> 00:02:24,000
+using AI support.
+
+31
+00:02:24,000 --> 00:02:25,000
+Let's start our lesson.
+
+32
+00:02:26,000 --> 00:02:30,000
+And as I already mentioned today, we are going to have a lot of practical examples.
+
+33
+00:02:30,000 --> 00:02:35,000
+So let me show you the end result that we are going to achieve by the end of the lesson.
+
+34
+00:02:36,000 --> 00:02:38,000
+Imagine that you need to send an email.
+
+35
+00:02:38,000 --> 00:02:41,000
+You clearly know what you want to send.
+
+36
+00:02:41,000 --> 00:02:45,000
+You also know whom you want to send the email to.
+
+37
+00:02:45,000 --> 00:02:52,000
+You just don't have too much time to put your thoughts on paper, Prepare an email and send it or you
+
+38
+00:02:52,000 --> 00:02:53,000
+are as lazy as I am.
+
+39
+00:02:54,000 --> 00:03:00,000
+Let me show you how both developed by me that takes advantage of GPT capabilities will help us to solve
+
+40
+00:03:00,000 --> 00:03:01,000
+this task.
+
+41
+00:03:02,000 --> 00:03:03,000
+I integrated.
+
+42
+00:03:03,000 --> 00:03:05,000
+GPT was developed by me web applications.
+
+43
+00:03:05,000 --> 00:03:08,000
+It is integrated with the Slack messenger too.
+
+44
+00:03:09,000 --> 00:03:15,000
+So at work I just put commands and slack every time I need help from GPT.
+
+45
+00:03:15,000 --> 00:03:23,000
+In this case, let me ask GPT to prepare an email for John, who is manager in my company and want to
+
+46
+00:03:23,000 --> 00:03:30,000
+invite John to the meeting to review the project he is working on discuss KPIs, risks and other project
+
+47
+00:03:30,000 --> 00:03:31,000
+related things.
+
+48
+00:03:32,000 --> 00:03:39,000
+I need to warn him to be prepared for this meeting and tell him the exact time, date and place of our
+
+49
+00:03:39,000 --> 00:03:39,000
+meeting.
+
+50
+00:03:39,000 --> 00:03:43,000
+So I just ask GPT to help me with this.
+
+51
+00:03:43,000 --> 00:03:52,000
+Then GPT generates an email for me and I must admit that the way how GPT structured this email is impressive.
+
+52
+00:03:52,000 --> 00:03:56,000
+I have email subject here, I have email body here.
+
+53
+00:03:56,000 --> 00:03:59,000
+I just need to put my signature and send it.
+
+54
+00:03:59,000 --> 00:04:06,000
+But sometimes I'm so lazy that I just want to have big red button under my hand that will do everything
+
+55
+00:04:06,000 --> 00:04:07,000
+what I need.
+
+56
+00:04:07,000 --> 00:04:13,000
+So I even don't want to open my outlook or Gmail or any other mail client.
+
+57
+00:04:13,000 --> 00:04:17,000
+I just ask GPT to send this email to John Doe.
+
+58
+00:04:17,000 --> 00:04:21,000
+I just add my signature at the end of the email and that's it.
+
+59
+00:04:21,000 --> 00:04:24,000
+What does my bot do on the background?
+
+60
+00:04:24,000 --> 00:04:30,000
+It understands based on the context that I want to send this email to John Doe.
+
+61
+00:04:30,000 --> 00:04:35,000
+That's why it searches for his email address in the data source configured.
+
+62
+00:04:35,000 --> 00:04:43,000
+It also understands where subject is finished and where email body is started, so the board gives me
+
+63
+00:04:43,000 --> 00:04:49,000
+proper arguments for my function that is in charge of sending emails based on the understanding of the
+
+64
+00:04:49,000 --> 00:04:50,000
+context.
+
+65
+00:04:51,000 --> 00:04:55,000
+And then I see confirmation that email was sent successfully.
+
+66
+00:04:55,000 --> 00:04:56,000
+Amazing.
+
+67
+00:04:56,000 --> 00:04:59,000
+Let me check the mailbox of John Doe.
+
+68
+00:05:00,000 --> 00:05:02,000
+As you may guess, this is a demo user with a.
+
+69
+00:05:03,000 --> 00:05:06,000
+Mailbox that created for the demo purposes.
+
+70
+00:05:06,000 --> 00:05:09,000
+That's why have access to his email.
+
+71
+00:05:09,000 --> 00:05:14,000
+And as we can see on the screen, John received my mail zero minutes ago.
+
+72
+00:05:14,000 --> 00:05:19,000
+The receiving time is the same as confirmation in my slack.
+
+73
+00:05:19,000 --> 00:05:24,000
+The remember when you last time prepared and sent emails so fast?
+
+74
+00:05:24,000 --> 00:05:28,000
+And of course opportunities for customization are endless.
+
+75
+00:05:28,000 --> 00:05:33,000
+You can book the online meeting if you wish or place meeting event to calendar.
+
+76
+00:05:33,000 --> 00:05:36,000
+There are no limits on the customization.
+
+77
+00:05:37,000 --> 00:05:43,000
+Like I always say that GPT is not a magic, but it is definitely a calculator for your brain.
+
+78
+00:05:43,000 --> 00:05:50,000
+By saying this, I mean like obviously you can do the math in your head and calculate long numbers on
+
+79
+00:05:50,000 --> 00:05:55,000
+paper, but you have calculator for this to boost your productivity, right?
+
+80
+00:05:55,000 --> 00:05:58,000
+So we live in an era of AI.
+
+81
+00:05:58,000 --> 00:06:02,000
+Let's use these tools to become even more productive and efficient.
+
+82
+00:06:03,000 --> 00:06:05,000
+Let me show you another example.
+
+83
+00:06:05,000 --> 00:06:11,000
+Imagine that you have engineering team and you need to create a bunch of user stories to fill out backlog
+
+84
+00:06:11,000 --> 00:06:12,000
+for them.
+
+85
+00:06:12,000 --> 00:06:17,000
+Of course you can do it manually, but let's take advantage of ChatGPT.
+
+86
+00:06:17,000 --> 00:06:22,000
+And again, using our team messenger, I just tell GPT what I need.
+
+87
+00:06:23,000 --> 00:06:27,000
+I ask it to create a user story that should be assigned to John Doe.
+
+88
+00:06:28,000 --> 00:06:31,000
+I explain briefly what this user story is about.
+
+89
+00:06:31,000 --> 00:06:37,000
+It is about card component in the e-commerce web app that we work on right now.
+
+90
+00:06:37,000 --> 00:06:40,000
+I also specified due date for this task.
+
+91
+00:06:40,000 --> 00:06:48,000
+After waiting a few moments GPT generated for me user story description with acceptance criteria and
+
+92
+00:06:48,000 --> 00:06:50,000
+even generated a ticket in the Jira.
+
+93
+00:06:50,000 --> 00:06:52,000
+Yeah, it is really fantastic.
+
+94
+00:06:53,000 --> 00:06:59,000
+I use this tool that I developed in my company and it boosted the productivity in my team significantly.
+
+95
+00:06:59,000 --> 00:07:04,000
+From now on we don't have user stories without complete description.
+
+96
+00:07:04,000 --> 00:07:07,000
+Like I said, GPT is not a magic tool.
+
+97
+00:07:07,000 --> 00:07:13,000
+I still recommend to verify everything what was generated here, but it doesn't take so much time in
+
+98
+00:07:13,000 --> 00:07:16,000
+comparison when we create requirements from scratch.
+
+99
+00:07:17,000 --> 00:07:21,000
+Here is a direct link to the Jira ticket generated too.
+
+100
+00:07:21,000 --> 00:07:24,000
+Let's click on it and see what we have there.
+
+101
+00:07:25,000 --> 00:07:27,000
+You can see that issue type is story.
+
+102
+00:07:28,000 --> 00:07:34,000
+I will teach you how to develop your bot in such a way that it would recognize what issue type you want
+
+103
+00:07:34,000 --> 00:07:35,000
+to create in Jira.
+
+104
+00:07:35,000 --> 00:07:37,000
+Summary Description.
+
+105
+00:07:37,000 --> 00:07:39,000
+Everything is on place.
+
+106
+00:07:39,000 --> 00:07:45,000
+Pay attention that assignee is also configured and due date is set to August 31st.
+
+107
+00:07:45,000 --> 00:07:46,000
+Like we asked.
+
+108
+00:07:47,000 --> 00:07:48,000
+Amazing.
+
+109
+00:07:48,000 --> 00:07:52,000
+And now imagine how this will boost productivity of your engineering teams.
+
+110
+00:07:52,000 --> 00:07:57,000
+How much time and thus money will you able to save?
+
+111
+00:07:57,000 --> 00:08:00,000
+And of course, this is just one simple example.
+
+112
+00:08:00,000 --> 00:08:04,000
+Imagine that you can boost your planning or refinement sessions.
+
+113
+00:08:04,000 --> 00:08:11,000
+For example, you can ask GPT to analyze the context and complexity of previously estimated user stories
+
+114
+00:08:11,000 --> 00:08:16,000
+and ask it to make estimation of not estimated user stories.
+
+115
+00:08:16,000 --> 00:08:22,000
+Of course, you can verify the estimation with the team and change it later if you wish, but again,
+
+116
+00:08:22,000 --> 00:08:24,000
+this will boost your productivity.
+
+117
+00:08:25,000 --> 00:08:31,000
+You can also ask the board that uses GPT capabilities to calculate average teams velocity and prepare
+
+118
+00:08:31,000 --> 00:08:39,000
+the scope of the next sprint based on the estimated tickets from the backlog and even to make preliminary
+
+119
+00:08:39,000 --> 00:08:42,000
+user stories assignment on team members.
+
+120
+00:08:42,000 --> 00:08:48,000
+If only I was an owner of Jira Software or Azure DevOps, I would implement inbuilt support of so many
+
+121
+00:08:48,000 --> 00:08:49,000
+cool features.
+
+122
+00:08:49,000 --> 00:08:55,000
+But while they are still working on them inside my company, I implemented custom solutions that I use
+
+123
+00:08:55,000 --> 00:08:56,000
+with my team.
+
+124
+00:08:57,000 --> 00:09:01,000
+Now it is time to review the source code and implementation.
+
+125
+00:09:01,000 --> 00:09:06,000
+As always, you can find the source code in attachments to the video.
+
+126
+00:09:06,000 --> 00:09:11,000
+Taking into account that in this lesson we are going to use the same approach that we already learned
+
+127
+00:09:11,000 --> 00:09:13,000
+in previous lesson.
+
+128
+00:09:13,000 --> 00:09:19,000
+I will show you demo and will make an overview of the implementation without stopping on the things
+
+129
+00:09:19,000 --> 00:09:22,000
+that we already learned in previous lesson.
+
+130
+00:09:22,000 --> 00:09:28,000
+And even in case you would have any questions, please do not hesitate to ask your questions below the
+
+131
+00:09:28,000 --> 00:09:31,000
+video and I will be happy to answer.
+
+132
+00:09:31,000 --> 00:09:37,000
+Let me start the review from the implementation of scenarios that will generate tickets for us in Jira.
+
+133
+00:09:38,000 --> 00:09:44,000
+Basically, together, we did a great job in previous lessons by implementing scalable architecture
+
+134
+00:09:45,000 --> 00:09:51,000
+that is advantage of all approach in comparison to pure functional programming approach.
+
+135
+00:09:51,000 --> 00:10:00,000
+The advantage becomes obvious on a bigger scale and as amount of code grows, the logic and principles
+
+136
+00:10:00,000 --> 00:10:07,000
+used to build our program helps us to scale faster and reliably without the risk of breaking the code
+
+137
+00:10:07,000 --> 00:10:09,000
+that we wrote before.
+
+138
+00:10:09,000 --> 00:10:15,000
+So the only thing that we need to do is to add additional functions that would be in charge of creation
+
+139
+00:10:15,000 --> 00:10:22,000
+ticket in Jira, because we already implemented code that gathers all beans of GPT functions and then
+
+140
+00:10:22,000 --> 00:10:24,000
+request to call functions.
+
+141
+00:10:24,000 --> 00:10:27,000
+All this was already implemented.
+
+142
+00:10:27,000 --> 00:10:31,000
+So let me open beans configuration class here.
+
+143
+00:10:31,000 --> 00:10:39,000
+I declared bin of GPT function and called it GPT Create Jira issue function as usual.
+
+144
+00:10:39,000 --> 00:10:42,000
+I created function, set, name and description.
+
+145
+00:10:42,000 --> 00:10:45,000
+And now let's talk about parameters.
+
+146
+00:10:45,000 --> 00:10:52,000
+So my function would require GPT to define the following parameters based on the context and give me
+
+147
+00:10:52,000 --> 00:10:53,000
+them in order.
+
+148
+00:10:53,000 --> 00:11:01,000
+I can make a request to Jira API, so I need to assign name work item description.
+
+149
+00:11:01,000 --> 00:11:07,000
+By the way, I expect that GPT will help me with both things to find the description for work item in
+
+150
+00:11:07,000 --> 00:11:12,000
+the context and actually to generate it based on my request first.
+
+151
+00:11:12,000 --> 00:11:19,000
+So GPT will generate the full description for the work item like you saw in the demo and then it will
+
+152
+00:11:19,000 --> 00:11:23,000
+give me the description in order I can create a work item in Jira.
+
+153
+00:11:24,000 --> 00:11:32,000
+Also, I need issue type attribute in order to understand what work item I need to create user story
+
+154
+00:11:32,000 --> 00:11:34,000
+task epic bug.
+
+155
+00:11:34,000 --> 00:11:42,000
+You can see that I also pass array of possible issue types in case ChatGPT can find information about
+
+156
+00:11:42,000 --> 00:11:45,000
+you date based on the context.
+
+157
+00:11:45,000 --> 00:11:52,000
+I would also ask it to provide me with it and the summary of the work item for the title in parameters.
+
+158
+00:11:52,000 --> 00:11:58,000
+I also specified that summary description, issue type and due date are required attributes, but if
+
+159
+00:11:58,000 --> 00:12:05,000
+you already worked with ChatGPT in previous lessons, I believe that you already understand that all
+
+160
+00:12:05,000 --> 00:12:08,000
+of this is just a recommendation for ChatGPT.
+
+161
+00:12:09,000 --> 00:12:16,000
+So even in case it wouldn't find due date, it still can request to call the function, but without
+
+162
+00:12:16,000 --> 00:12:16,000
+due date.
+
+163
+00:12:17,000 --> 00:12:24,000
+I had such cases and I will be honest with you, I hope that guys from OpenAI will keep improving their
+
+164
+00:12:24,000 --> 00:12:24,000
+product.
+
+165
+00:12:24,000 --> 00:12:27,000
+But based on today it works like this.
+
+166
+00:12:28,000 --> 00:12:31,000
+Still, I decided to keep due date as required attribute.
+
+167
+00:12:32,000 --> 00:12:38,000
+You can explore my source code and try different options and select the ones that works the best for
+
+168
+00:12:38,000 --> 00:12:40,000
+you to address your business needs.
+
+169
+00:12:40,000 --> 00:12:46,000
+You can see that all values are configurable and located in my properties file.
+
+170
+00:12:46,000 --> 00:12:53,000
+Feel free to explore the attribute descriptions, but I bet there will not be any surprises for you.
+
+171
+00:12:53,000 --> 00:13:01,000
+I just described for ChatGPT what each attribute means and in similar way I created classes of required
+
+172
+00:13:01,000 --> 00:13:02,000
+entities.
+
+173
+00:13:02,000 --> 00:13:08,000
+I wouldn't stop on that too, because in previous lesson we also created classes for entities.
+
+174
+00:13:08,000 --> 00:13:15,000
+So I can just confirm that I used this same approach and principles here in case Jira will decide to
+
+175
+00:13:15,000 --> 00:13:23,000
+invoke the function I will take from the context create Jira issue function been that I also configure
+
+176
+00:13:23,000 --> 00:13:24,000
+it here.
+
+177
+00:13:24,000 --> 00:13:33,000
+Let's check this function according to the interface it implements one method execute so I pass arguments
+
+178
+00:13:33,000 --> 00:13:40,000
+in order to not pass arguments one by one, I created separate entity and pass all arguments directly
+
+179
+00:13:40,000 --> 00:13:41,000
+to Jira.
+
+180
+00:13:41,000 --> 00:13:44,000
+Issue fields object in general service.
+
+181
+00:13:44,000 --> 00:13:48,000
+I implemented method that is called create Jira issue.
+
+182
+00:13:48,000 --> 00:13:53,000
+It will be in charge of creation of work item in Jira.
+
+183
+00:13:54,000 --> 00:13:56,000
+Let me open the source code of the default.
+
+184
+00:13:57,000 --> 00:14:00,000
+Service class and show you the implementation of that message.
+
+185
+00:14:01,000 --> 00:14:08,000
+And here you can see implementation of the message using Jira API documentation that we reviewed in
+
+186
+00:14:08,000 --> 00:14:09,000
+the lesson about Jira API.
+
+187
+00:14:10,000 --> 00:14:17,000
+I created URL and http post request object and here goes specifics of Jira API.
+
+188
+00:14:18,000 --> 00:14:23,000
+I just need to prepare the body of the post request and also for assignee.
+
+189
+00:14:23,000 --> 00:14:30,000
+I need to extract Jira user ID because according to API it is not enough to parse the name.
+
+190
+00:14:31,000 --> 00:14:37,000
+So based on the name I extract Jira user id and I have separate method for that.
+
+191
+00:14:37,000 --> 00:14:44,000
+In Jira Cloud there is an endpoint that returns the array of users and you can work with it as you wish.
+
+192
+00:14:44,000 --> 00:14:47,000
+In order to find Jira user ID.
+
+193
+00:14:47,000 --> 00:14:54,000
+So I created here Jira request body object that contains fields property.
+
+194
+00:14:54,000 --> 00:15:00,000
+Fields contains assignee issue type project description due date summary.
+
+195
+00:15:01,000 --> 00:15:06,000
+Then set the Jira request body to the Http post object as an entity.
+
+196
+00:15:07,000 --> 00:15:13,000
+This is interface of Apache http client that we already used and learned before.
+
+197
+00:15:13,000 --> 00:15:16,000
+That's why don't stop on this too much.
+
+198
+00:15:16,000 --> 00:15:19,000
+Then I should receive response from Jira.
+
+199
+00:15:19,000 --> 00:15:25,000
+And here in Commented Lines you can see the example of response that I expect to receive.
+
+200
+00:15:26,000 --> 00:15:33,000
+I extract only value of the key attribute and in the create Jira issue function class you saw that I
+
+201
+00:15:33,000 --> 00:15:41,000
+returned back the year of the ticket in know the GPT can let our end user know the idea of the ticket
+
+202
+00:15:41,000 --> 00:15:42,000
+that was created.
+
+203
+00:15:42,000 --> 00:15:49,000
+And also another important point in order to improve user experience in the default slack service method,
+
+204
+00:15:49,000 --> 00:15:54,000
+I created a method that is called add hyper references to Jira.
+
+205
+00:15:55,000 --> 00:16:02,000
+This method finds the Jira ticket IDs based on the pattern configured in the regular expression that
+
+206
+00:16:02,000 --> 00:16:10,000
+takes into account Jira, project name, dash and number and substitute all Jira tickets mentions by
+
+207
+00:16:10,000 --> 00:16:12,000
+adding hyper reference in the parentheses.
+
+208
+00:16:12,000 --> 00:16:14,000
+Next is the Jira key.
+
+209
+00:16:14,000 --> 00:16:20,000
+This is just an additional feature that I added in order to improve user experience for end user by
+
+210
+00:16:20,000 --> 00:16:24,000
+having linked to the Jira in the chat like you saw in the demo.
+
+211
+00:16:25,000 --> 00:16:26,000
+Basically that's it.
+
+212
+00:16:27,000 --> 00:16:32,000
+Like I said, you can find the reference to the source code in attachments to the video.
+
+213
+00:16:32,000 --> 00:16:34,000
+Feel free to look through the solution.
+
+214
+00:16:34,000 --> 00:16:40,000
+I believe there will be nothing new to you because we learned already everything what is required to
+
+215
+00:16:40,000 --> 00:16:42,000
+implement this solution.
+
+216
+00:16:42,000 --> 00:16:47,000
+But still, in case you would have any specific question, I'm here to help.
+
+217
+00:16:47,000 --> 00:16:52,000
+Let's review now the second scenario with sending an email in the similar way.
+
+218
+00:16:52,000 --> 00:16:55,000
+Let me start overview from the beans configuration file.
+
+219
+00:16:56,000 --> 00:17:00,000
+I declare the beans that is called GPT Send email function.
+
+220
+00:17:00,000 --> 00:17:08,000
+I tell GPT which parameters I expect in order I can send an email so it is either addressee, email
+
+221
+00:17:08,000 --> 00:17:10,000
+or addressee name.
+
+222
+00:17:10,000 --> 00:17:16,000
+And again, everything will depend on the environment you work in and business requirements.
+
+223
+00:17:16,000 --> 00:17:23,000
+For example, current solution will send an email in case there will be email address in the context
+
+224
+00:17:23,000 --> 00:17:30,000
+and it will also send an email in case there will not be email address in the context, but in case
+
+225
+00:17:30,000 --> 00:17:34,000
+ChatGPT can get the full name of the addressee from the context.
+
+226
+00:17:34,000 --> 00:17:42,000
+In this case, ChatGPT will give me a name and I will find email address in the third party data source
+
+227
+00:17:42,000 --> 00:17:49,000
+and in my case I can integrate with internal data source where I can find email by full name.
+
+228
+00:17:49,000 --> 00:17:57,000
+For example, I can find email in Active Directory or whatever tool your organization uses, even from
+
+229
+00:17:57,000 --> 00:17:59,000
+Excel file if needed.
+
+230
+00:17:59,000 --> 00:18:05,000
+So just check the business requirements and decide how you want to implement this part of functionality.
+
+231
+00:18:05,000 --> 00:18:11,000
+And there are also two other attributes that we need email, subject and email content.
+
+232
+00:18:11,000 --> 00:18:12,000
+That's it.
+
+233
+00:18:12,000 --> 00:18:21,000
+Required attributes are content and subject in case ChatGPT will not be able to identify an email address
+
+234
+00:18:21,000 --> 00:18:22,000
+nor name of the addressee.
+
+235
+00:18:22,000 --> 00:18:29,000
+In this case, I will let end user know that I can't execute this function and that I need is an email
+
+236
+00:18:29,000 --> 00:18:30,000
+or full name.
+
+237
+00:18:31,000 --> 00:18:38,000
+Then, in case ChatGPT would ask me to call this function, I will extract send email function being
+
+238
+00:18:38,000 --> 00:18:39,000
+from the context.
+
+239
+00:18:39,000 --> 00:18:47,000
+Let's check how it is implemented in the send email function been injected mail service.
+
+240
+00:18:47,000 --> 00:18:54,000
+This is the type that will be responsible for sending an email in the execute method during the passing
+
+241
+00:18:54,000 --> 00:19:01,000
+of the arguments in case there is no email or addressee name, I returns a message that tells that email
+
+242
+00:19:01,000 --> 00:19:02,000
+wasn't sent and why.
+
+243
+00:19:03,000 --> 00:19:11,000
+In case there is no email but there is an addressee name, I try to extract email address first of our
+
+244
+00:19:11,000 --> 00:19:15,000
+user in case full email address can't be found.
+
+245
+00:19:15,000 --> 00:19:21,000
+I send back information that email wasn't sent and explains the reason why.
+
+246
+00:19:21,000 --> 00:19:27,000
+You can see that I use extract email by full name method from email service in order to extract the
+
+247
+00:19:27,000 --> 00:19:28,000
+email.
+
+248
+00:19:29,000 --> 00:19:33,000
+This is exactly the message that you need to decide how you would implement it.
+
+249
+00:19:33,000 --> 00:19:40,000
+You can read email addresses and full names from external data source, like for example, from the
+
+250
+00:19:40,000 --> 00:19:47,000
+database, from Active Directory, from Excel file, whatever data source you like the most.
+
+251
+00:19:47,000 --> 00:19:54,000
+In this specific learning example, I hardcoded map with full names and email addresses taking into
+
+252
+00:19:54,000 --> 00:19:57,000
+account this is application for my students.
+
+253
+00:19:57,000 --> 00:20:02,000
+I don't want to overcomplicate this example with one more integration.
+
+254
+00:20:02,000 --> 00:20:09,000
+One day we will learn about integration with Active Directory and we will update this example because
+
+255
+00:20:09,000 --> 00:20:13,000
+it is a separate, larger topic to talk about, not in scope of this lesson.
+
+256
+00:20:14,000 --> 00:20:19,000
+And send email method is implemented right above in the same file.
+
+257
+00:20:19,000 --> 00:20:27,000
+I configured gmail Smtp server for sending emails and again, this method can be implemented in different
+
+258
+00:20:27,000 --> 00:20:28,000
+ways.
+
+259
+00:20:28,000 --> 00:20:35,000
+You can use dedicated email client for sending emails from your board, or you can find a way to authenticate
+
+260
+00:20:35,000 --> 00:20:40,000
+and authorize access of the end user to his or her email.
+
+261
+00:20:40,000 --> 00:20:46,000
+And this configuration can be done during the bot installation or when start using the bot.
+
+262
+00:20:46,000 --> 00:20:52,000
+Again, there is unlimited variety of options of how you can build this integration.
+
+263
+00:20:52,000 --> 00:20:59,000
+The main thing is that you understand the key principles and if you understand them well, you are limited
+
+264
+00:20:59,000 --> 00:21:06,000
+only with your fantasy of how you want to implement this in the send email function source code.
+
+265
+00:21:06,000 --> 00:21:11,000
+You can see that in case email was sent and sent, email message returned.
+
+266
+00:21:11,000 --> 00:21:11,000
+True.
+
+267
+00:21:12,000 --> 00:21:19,000
+In case of success, I notify my board and GPG that email was sent in case there was some exception
+
+268
+00:21:19,000 --> 00:21:21,000
+during email sending.
+
+269
+00:21:21,000 --> 00:21:28,000
+Send email message will return false and will confirm that email wasn't sent with adding reasoning message.
+
+270
+00:21:29,000 --> 00:21:30,000
+Basically that's it.
+
+271
+00:21:31,000 --> 00:21:36,000
+If you are interested in how I send an email, you can spend a little bit more time reviewing the source
+
+272
+00:21:36,000 --> 00:21:38,000
+code of the default mail service.
+
+273
+00:21:39,000 --> 00:21:44,000
+Even in case you would have any questions or in case you would like me to review this topic in details,
+
+274
+00:21:44,000 --> 00:21:46,000
+I am just two clicks away.
+
+275
+00:21:47,000 --> 00:21:51,000
+You can write your question below this video and I will be happy to answer.
+
+276
+00:21:51,000 --> 00:21:54,000
+Okay, that's it for this lesson.
+
+277
+00:21:54,000 --> 00:21:57,000
+Let's recap what we have learned today.
+
+278
+00:21:57,000 --> 00:22:03,000
+In this lesson, we reviewed two scenarios to business use cases of how you can apply your knowledge
+
+279
+00:22:03,000 --> 00:22:05,000
+and practice with ChatGPT.
+
+280
+00:22:05,000 --> 00:22:12,000
+They are creating work item in Jira, generating description and summary for Jira ticket directly from
+
+281
+00:22:12,000 --> 00:22:20,000
+the slack and the second scenario preparation of email and sending it via a chat interface.
+
+282
+00:22:20,000 --> 00:22:22,000
+That's all for today.
+
+283
+00:22:22,000 --> 00:22:24,000
+Thanks a lot for your attention.
+
+284
+00:22:24,000 --> 00:22:27,000
+Have a great day and see you in the next lesson.
+

100 - GPT + Slack + Jira + Gmail Integration/002 Generate-email-and-send-it-from-chat-Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/5fa99c435a722527fccb152de538e113cb397985

100 - GPT + Slack + Jira + Gmail Integration/external-links.txt

@@ -0,0 +1,9 @@
+
+001 Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/cb46583146e15ddd4782ab32315c14ca55f73edf
+
+002 Create-Jira-ticket-from-chat-Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/7853021ac12529fcb1c58f49f6d9495624f8850b
+
+002 Generate-email-and-send-it-from-chat-Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/5fa99c435a722527fccb152de538e113cb397985

101 - Manage a Scrum Team with ChatGPT/001 Average-velocity-calculation-Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/e4266b6cc3cb9c1021bd8810471fd6fa7c21ca64

101 - Manage a Scrum Team with ChatGPT/001 Managing Scrum & Risk Management with Custom Bot, Slack & GPT_en.srt

@@ -0,0 +1,560 @@
+1
+00:00:05,000 --> 00:00:07,000
+Hello, my name is Andre Petaja.
+
+2
+00:00:07,000 --> 00:00:14,000
+Let me show you how I use boards that I develop that use capabilities of ChatGPT in order to perform
+
+3
+00:00:14,000 --> 00:00:16,000
+project management operational activities.
+
+4
+00:00:17,000 --> 00:00:23,000
+Right now, I want to show you the scenario, how I use board with ChatGPT to boost my sprint planning
+
+5
+00:00:23,000 --> 00:00:25,000
+and sprint preparation activities.
+
+6
+00:00:25,000 --> 00:00:31,000
+If you are a project manager who has experience with Scrum methodology, you know that during spring
+
+7
+00:00:31,000 --> 00:00:37,000
+planning it is better to know teams capacity for the next sprint in order to be more accurate during
+
+8
+00:00:37,000 --> 00:00:38,000
+the planning.
+
+9
+00:00:38,000 --> 00:00:44,000
+There are different approaches how to calculate teams capacity, but one of the approaches that I use
+
+10
+00:00:44,000 --> 00:00:49,000
+is calculation of sprint capacity based on the average velocity in story points.
+
+11
+00:00:49,000 --> 00:00:56,000
+So usually I take amount of story points delivered in the last 2 or 3 sprints and calculate average
+
+12
+00:00:56,000 --> 00:00:57,000
+velocity.
+
+13
+00:00:57,000 --> 00:01:03,000
+Then, based on the average velocity, I define teams capacity in order to plan sprint.
+
+14
+00:01:04,000 --> 00:01:10,000
+In the current example you can see that I created fake sprints with fake data in order to have some
+
+15
+00:01:10,000 --> 00:01:12,000
+data for the use case that I want to demo.
+
+16
+00:01:13,000 --> 00:01:20,000
+I closed three sprints with the user stories and I received the following data in the sprint.
+
+17
+00:01:20,000 --> 00:01:24,000
+Number one, my team delivered 16 story points in the sprint.
+
+18
+00:01:24,000 --> 00:01:32,000
+Number two, my team managed to complete 14 story points and in the sprint three also 14 story points.
+
+19
+00:01:33,000 --> 00:01:38,000
+We are going to use this data in our example in order to review end to end scenario.
+
+20
+00:01:39,000 --> 00:01:45,000
+And now imagine that I have a conversation in Slack Messenger with my team members or with my Scrum
+
+21
+00:01:45,000 --> 00:01:50,000
+Masters and I just want to get more information about the average velocity.
+
+22
+00:01:50,000 --> 00:01:53,000
+I need this information to hold planning.
+
+23
+00:01:53,000 --> 00:01:56,000
+I can just ask my bot in chat.
+
+24
+00:01:56,000 --> 00:02:03,000
+What is the average velocity based on the previously completed three sprints and receive a response.
+
+25
+00:02:03,000 --> 00:02:10,000
+14.67 story points, which is correct based on the data we received from Jira.
+
+26
+00:02:11,000 --> 00:02:19,000
+I can clarify the average velocity based on the previous two sprints and the answer is 14 story points
+
+27
+00:02:19,000 --> 00:02:27,000
+is it is also correct because in Sprint two and Sprint three, my team delivered 14 story points per
+
+28
+00:02:27,000 --> 00:02:34,000
+each sprint and I know that I can open velocity charts in Jira and take information from there.
+
+29
+00:02:34,000 --> 00:02:39,000
+But what if this is a live chat and I want to receive information faster?
+
+30
+00:02:39,000 --> 00:02:46,000
+And this scenario is very simplified demo because in reality you can integrate these commands into more
+
+31
+00:02:46,000 --> 00:02:54,000
+complex scenarios like email preparation and its auto generation reports generation and many other different
+
+32
+00:02:54,000 --> 00:02:55,000
+generation scenarios.
+
+33
+00:02:55,000 --> 00:03:01,000
+But even using it in chat is pretty comfortable knowing this information.
+
+34
+00:03:01,000 --> 00:03:07,000
+I can proceed with planning, but in order you can understand the use case, let me show you how my
+
+35
+00:03:07,000 --> 00:03:09,000
+backlog looks like now.
+
+36
+00:03:09,000 --> 00:03:16,000
+In my backlog, I already have estimated workitems that by the way, also may be estimated with the
+
+37
+00:03:16,000 --> 00:03:22,000
+help of board and ChatGPT based on the analysis of the text description of the previous user stories.
+
+38
+00:03:23,000 --> 00:03:31,000
+Or we can train and fine tune our model in order to teach it how to define complexity based on the requirements.
+
+39
+00:03:31,000 --> 00:03:35,000
+But that case will be described and shown in the separate demo.
+
+40
+00:03:36,000 --> 00:03:41,000
+So here in the backlog, you can also see that user stories have priorities.
+
+41
+00:03:41,000 --> 00:03:48,000
+And when I plan Sprint, I want to take into consideration priorities and plan the user stories with
+
+42
+00:03:48,000 --> 00:03:56,000
+highest priorities and then with lowest, you can see that there are two heavy user stories with high
+
+43
+00:03:56,000 --> 00:03:59,000
+priority eight story points each.
+
+44
+00:03:59,000 --> 00:04:04,000
+That means, unfortunately, both of them doesn't fit into Sprint.
+
+45
+00:04:04,000 --> 00:04:12,000
+If our capacity is 14 story points and in this case, remaining space in Sprint should be filled out
+
+46
+00:04:12,000 --> 00:04:14,000
+by user stories with lower priorities.
+
+47
+00:04:14,000 --> 00:04:16,000
+Does it sound logical for you?
+
+48
+00:04:16,000 --> 00:04:26,000
+So let me ask now our board to plan Sprint four Considering that our capacity is 14 story points, I
+
+49
+00:04:26,000 --> 00:04:32,000
+decided to use average velocity based on the previous two sprints to calculate our capacity for the
+
+50
+00:04:32,000 --> 00:04:33,000
+next sprint.
+
+51
+00:04:33,000 --> 00:04:39,000
+And after a few moments my boss replies me that sprint was planned and such stories are included into
+
+52
+00:04:39,000 --> 00:04:44,000
+the sprint for the plan was done with respect to teams capacity.
+
+53
+00:04:45,000 --> 00:04:49,000
+Let me open Jira now and make sure that these changes are reflected.
+
+54
+00:04:50,000 --> 00:04:57,000
+So I refresh the page and I can see that Sprint four is planned with top priority items for certain
+
+55
+00:04:57,000 --> 00:05:02,000
+story points, and the only thing that is left is to start the sprint.
+
+56
+00:05:03,000 --> 00:05:09,000
+Another user story was high priority for eight story points, unfortunately doesn't fit into the sprint
+
+57
+00:05:09,000 --> 00:05:12,000
+and will be left for the following sprint planning.
+
+58
+00:05:13,000 --> 00:05:20,000
+That's how using chat interface I can check average velocity in story points and plan the following
+
+59
+00:05:20,000 --> 00:05:20,000
+sprint.
+
+60
+00:05:20,000 --> 00:05:28,000
+And just keep in mind that this is just one example and variety of other combinations or options for
+
+61
+00:05:28,000 --> 00:05:31,000
+how you can build your own board with chat.
+
+62
+00:05:31,000 --> 00:05:33,000
+GPT integration is unlimited.
+
+63
+00:05:34,000 --> 00:05:36,000
+Okay, let's review another example.
+
+64
+00:05:36,000 --> 00:05:42,000
+Now, I would like to show you how my board that is integrated with ChatGPT helps me to manage risks
+
+65
+00:05:42,000 --> 00:05:45,000
+and work with risks on daily basis.
+
+66
+00:05:45,000 --> 00:05:46,000
+Risk management.
+
+67
+00:05:46,000 --> 00:05:50,000
+That is something what is not specific only for scrum projects.
+
+68
+00:05:50,000 --> 00:05:57,000
+That's why no matter whether you worked with Scrum before or no, this example should be clear for you.
+
+69
+00:05:57,000 --> 00:06:01,000
+In case you are familiar with project management techniques according to PMI.
+
+70
+00:06:02,000 --> 00:06:05,000
+Let me start from the preconditions.
+
+71
+00:06:05,000 --> 00:06:09,000
+In our specific example, I configured separate board for risk management.
+
+72
+00:06:09,000 --> 00:06:13,000
+On this board I have only items with risk issue type.
+
+73
+00:06:14,000 --> 00:06:16,000
+This is a custom type that I created.
+
+74
+00:06:17,000 --> 00:06:18,000
+There are six risks.
+
+75
+00:06:18,000 --> 00:06:22,000
+Each risk has its own impact and probability.
+
+76
+00:06:22,000 --> 00:06:30,000
+Configure it and imagine that I want to streamline our conversation and have a chat about risks directly
+
+77
+00:06:30,000 --> 00:06:37,000
+in the messenger with my colleagues so I can use my board to ask questions about risks that we have
+
+78
+00:06:37,000 --> 00:06:38,000
+on our project.
+
+79
+00:06:39,000 --> 00:06:42,000
+For example, let's start with simple questions.
+
+80
+00:06:42,000 --> 00:06:45,000
+How many risks we have in total?
+
+81
+00:06:45,000 --> 00:06:49,000
+And when Jira is connected as a data source to the board.
+
+82
+00:06:49,000 --> 00:06:56,000
+I use ChatGPT linguistic capabilities to understand my request and provide me with the response.
+
+83
+00:06:56,000 --> 00:07:02,000
+In case you set priorities to risks and you need to review risks by priorities, I can ask board to
+
+84
+00:07:02,000 --> 00:07:09,000
+provide me with the requested information and you can see that I have one ticket with the highest priority,
+
+85
+00:07:09,000 --> 00:07:16,000
+one with high priority, and the rest of risks have the medium priority on the Kanban board.
+
+86
+00:07:16,000 --> 00:07:23,000
+You can check how accurate our board is and you can see that both provided us with correct information
+
+87
+00:07:23,000 --> 00:07:24,000
+about priorities.
+
+88
+00:07:24,000 --> 00:07:30,000
+That is because it is smart enough to process the data that is provided to the board on request.
+
+89
+00:07:30,000 --> 00:07:38,000
+ChatGPT identifies what information is needed and asks my board to provide necessary information.
+
+90
+00:07:38,000 --> 00:07:45,000
+My board fetches required information and provides it back to the GPT for the further analysis.
+
+91
+00:07:45,000 --> 00:07:47,000
+That is what is happening in the background.
+
+92
+00:07:48,000 --> 00:07:56,000
+If we want, we can ask to group all risks by statuses and a few moments after I receive the response,
+
+93
+00:07:56,000 --> 00:08:01,000
+we can see here clear breakdown of risk items by statuses.
+
+94
+00:08:01,000 --> 00:08:09,000
+I can request to group risks by impact and again, I receive information that I need that was analyzed
+
+95
+00:08:09,000 --> 00:08:11,000
+and processed by ChatGPT.
+
+96
+00:08:11,000 --> 00:08:15,000
+And you can ask to analyze any data you need.
+
+97
+00:08:15,000 --> 00:08:18,000
+If you need to group risks by probabilities.
+
+98
+00:08:18,000 --> 00:08:20,000
+You can also ask what to do.
+
+99
+00:08:20,000 --> 00:08:23,000
+So believe me, result will be correct.
+
+100
+00:08:24,000 --> 00:08:30,000
+And instead of this, imagine that you want to identify responsible team members who are in charge of
+
+101
+00:08:30,000 --> 00:08:33,000
+risks that are in backlog state.
+
+102
+00:08:33,000 --> 00:08:39,000
+I just want to know names of these heroes and team members in order to follow up with them and ask them
+
+103
+00:08:39,000 --> 00:08:45,000
+to process risks as soon as possible and ask this question to my board.
+
+104
+00:08:45,000 --> 00:08:52,000
+And it replies me that Andre Petaja is assigned to two risks that are in the backlog state now.
+
+105
+00:08:52,000 --> 00:08:56,000
+Now I know the name with whom I need to follow up.
+
+106
+00:08:56,000 --> 00:09:01,000
+I can tag the risk owner here or configure cron job with reminder.
+
+107
+00:09:02,000 --> 00:09:09,000
+For example, each morning responsible people are attacked in case they have risks that are still not
+
+108
+00:09:09,000 --> 00:09:10,000
+processed.
+
+109
+00:09:10,000 --> 00:09:17,000
+This would be ideal for big programs and it will help to boost the communication inside the team significantly.
+
+110
+00:09:18,000 --> 00:09:22,000
+Any team member, even without opening a Jira, can ask a follow up questions.
+
+111
+00:09:22,000 --> 00:09:29,000
+Like, for example, we can request additional details about these two risks in order to remember what
+
+112
+00:09:29,000 --> 00:09:36,000
+are they about and what provides me with only necessary information In order I can get better context
+
+113
+00:09:36,000 --> 00:09:37,000
+of the risks.
+
+114
+00:09:37,000 --> 00:09:43,000
+In response, I can find information about assignee, risk, description, impact and probability.
+
+115
+00:09:44,000 --> 00:09:51,000
+GPT also knows that each of these risks has its own mitigation plan and response strategies.
+
+116
+00:09:51,000 --> 00:09:56,000
+So let me then clarify these details and ask for additional details.
+
+117
+00:09:58,000 --> 00:10:03,000
+And this time I see information about mitigation plan and response strategies included.
+
+118
+00:10:03,000 --> 00:10:10,000
+The interesting thing to remember about is that ChatGPT sometimes generates additional details.
+
+119
+00:10:10,000 --> 00:10:18,000
+For example, in our specific case response strategy is just a single value from the dropdown.
+
+120
+00:10:18,000 --> 00:10:22,000
+But instead of writing that response strategy is mitigate.
+
+121
+00:10:22,000 --> 00:10:30,000
+We can see that additional text is added like in this example, implement measures to mitigate risk
+
+122
+00:10:30,000 --> 00:10:32,000
+instead of just mitigate.
+
+123
+00:10:32,000 --> 00:10:39,000
+The main idea is delivered correctly, but be aware that GPT can generate additional wording sometimes.
+
+124
+00:10:39,000 --> 00:10:46,000
+In the previous video I already showed you how we can create Jira issues using chat interface.
+
+125
+00:10:46,000 --> 00:10:48,000
+The same things may be applied here.
+
+126
+00:10:48,000 --> 00:10:51,000
+We can use our bot together with a chat.
+
+127
+00:10:51,000 --> 00:10:54,000
+GPT capabilities to generate risk Description.
+
+128
+00:10:54,000 --> 00:11:01,000
+Ask GPT to suggest US mitigation plan and put this data directly into the ticket.
+
+129
+00:11:01,000 --> 00:11:05,000
+I just don't see the reason to demo all seamless scenarios.
+
+130
+00:11:05,000 --> 00:11:10,000
+The main thing that I wanted to show you in this video is that you are not bound to the default types
+
+131
+00:11:10,000 --> 00:11:11,000
+in Jira.
+
+132
+00:11:11,000 --> 00:11:18,000
+You can customize your bot and take advantage of ChatGPT capabilities to build any customized business
+
+133
+00:11:18,000 --> 00:11:20,000
+logic or business flow that you need.
+
+134
+00:11:21,000 --> 00:11:22,000
+That's all for this video.
+
+135
+00:11:23,000 --> 00:11:29,000
+In case you are interested in the topic, check video description and feel free to ask your questions
+
+136
+00:11:29,000 --> 00:11:30,000
+in comments to the video.
+
+137
+00:11:31,000 --> 00:11:33,000
+I hope you enjoyed the video.
+
+138
+00:11:33,000 --> 00:11:39,000
+Put your thumbs up, leave the comments and follow the channel to not miss other interesting videos.
+
+139
+00:11:39,000 --> 00:11:40,000
+Have a great day.
+
+140
+00:11:40,000 --> 00:11:41,000
+Bye.
+

101 - Manage a Scrum Team with ChatGPT/001 Risk-management-Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/3110d2ed017c5c6ab196916359cb8c3983f063b5

101 - Manage a Scrum Team with ChatGPT/001 Sprint-Planning-Source-code-of-examples-from-the-lesson-commit-with-changes-.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/commit/3f3e698451ac043734f0b0b56974a49385c9865d

101 - Manage a Scrum Team with ChatGPT/external-links.txt

@@ -0,0 +1,9 @@
+
+001 Average-velocity-calculation-Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/e4266b6cc3cb9c1021bd8810471fd6fa7c21ca64
+
+001 Sprint-Planning-Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/3f3e698451ac043734f0b0b56974a49385c9865d
+
+001 Risk-management-Source-code-of-examples-from-the-lesson-commit-with-changes-
+https://github.com/AndriiPiatakha/openai-learnit/commit/3110d2ed017c5c6ab196916359cb8c3983f063b5

102 - DALL-E - Text to image AI Model by OpenAI/001 API-Reference-for-DALL-E-Model.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://platform.openai.com/docs/api-reference/images

102 - DALL-E - Text to image AI Model by OpenAI/001 DALL-E Model & API Overview With Examples in Postman_en.srt

@@ -0,0 +1,1676 @@
+1
+00:00:05,000 --> 00:00:06,000
+Hello, team.
+
+2
+00:00:06,000 --> 00:00:12,000
+In this video we are going to learn how to work with the API and understand what it is.
+
+3
+00:00:12,000 --> 00:00:16,000
+We are going to start this lesson from the general overview of the model.
+
+4
+00:00:16,000 --> 00:00:21,000
+Then I will explain you what a decoder and encoder are.
+
+5
+00:00:21,000 --> 00:00:23,000
+In transforming models.
+
+6
+00:00:23,000 --> 00:00:27,000
+We will review business use cases and limitations of the model.
+
+7
+00:00:27,000 --> 00:00:34,000
+I'm going to show you the API documentation and we will focus attention on the most important things.
+
+8
+00:00:34,000 --> 00:00:40,000
+Then we will learn how to use API and real examples I will share with you Postman collections that I
+
+9
+00:00:40,000 --> 00:00:45,000
+prepared specially for this lesson and we will work with the API.
+
+10
+00:00:45,000 --> 00:00:48,000
+We will generate images based on a text prompt.
+
+11
+00:00:48,000 --> 00:00:53,000
+Will edit images based on a text prompt and will create variations of an existing image.
+
+12
+00:00:53,000 --> 00:00:55,000
+Let's start our lesson.
+
+13
+00:00:56,000 --> 00:00:59,000
+Let's start from understanding of what Dali is.
+
+14
+00:00:59,000 --> 00:01:07,000
+Dali is a neural network model developed by OpenAI for generating images from textual descriptions.
+
+15
+00:01:07,000 --> 00:01:15,000
+This model is trained on a dataset of text image pairs and has the capability to generate a wide range
+
+16
+00:01:15,000 --> 00:01:18,000
+of images based on natural language descriptions.
+
+17
+00:01:19,000 --> 00:01:27,000
+Dali is a decoder only transformer models that takes both text and image data as input and models them
+
+18
+00:01:27,000 --> 00:01:32,000
+autoregressively to generate images based on a textual descriptions.
+
+19
+00:01:33,000 --> 00:01:39,000
+The model's architecture and training procedure are detailed in OpenAI research paper.
+
+20
+00:01:40,000 --> 00:01:46,000
+In a minute, I'm going to explain you what decoder and encoder concepts are in transforming models.
+
+21
+00:01:46,000 --> 00:01:53,000
+The capabilities of the opened up possibilities for various applications, including content generation,
+
+22
+00:01:53,000 --> 00:02:01,000
+design, art and potentially many more areas where the translation of textual ideas into visual representations
+
+23
+00:02:01,000 --> 00:02:02,000
+is required.
+
+24
+00:02:03,000 --> 00:02:08,000
+Let's understand what a decoder and encoder are in transforming models.
+
+25
+00:02:08,000 --> 00:02:11,000
+It will be useful for your general education.
+
+26
+00:02:11,000 --> 00:02:16,000
+Also, it will help you to understand architectures of other transforming models.
+
+27
+00:02:16,000 --> 00:02:20,000
+I said that Delhi is a decoder only transformer model.
+
+28
+00:02:20,000 --> 00:02:27,000
+It means that the model consists only of the decoder part of the transformer architecture without the
+
+29
+00:02:27,000 --> 00:02:34,000
+encoder in the context of transformer models, the encoder is typically responsible for processing input
+
+30
+00:02:34,000 --> 00:02:40,000
+data, while the decoder generates output data based on that processed input.
+
+31
+00:02:40,000 --> 00:02:46,000
+Here is breakdown of the roles of the encoder and decoder in a typical transformer architecture.
+
+32
+00:02:46,000 --> 00:02:53,000
+The encoder is responsible for processing and encoding the input data in natural language processing
+
+33
+00:02:53,000 --> 00:02:53,000
+tasks.
+
+34
+00:02:53,000 --> 00:02:57,000
+The input data is usually a sequence of words or tokens.
+
+35
+00:02:58,000 --> 00:03:05,000
+The encoder processes this input sequence and generates a set of hidden representations or embeddings
+
+36
+00:03:05,000 --> 00:03:06,000
+for each token.
+
+37
+00:03:06,000 --> 00:03:13,000
+These representations capture contextual information about each token in relation to the others in the
+
+38
+00:03:13,000 --> 00:03:14,000
+sequence.
+
+39
+00:03:14,000 --> 00:03:18,000
+The encoder's output is then used as an input to the decoder.
+
+40
+00:03:19,000 --> 00:03:25,000
+The decoder takes the encoded information from the encoder and generates the output sequence.
+
+41
+00:03:25,000 --> 00:03:31,000
+In the case of text generation, this output sequence could be a sequence of words or tokens.
+
+42
+00:03:31,000 --> 00:03:38,000
+The decoder processes the encoded information autoregressively generating one token at a time while
+
+43
+00:03:38,000 --> 00:03:40,000
+considering the previously generated tokens.
+
+44
+00:03:41,000 --> 00:03:46,000
+This autoregressive process continues until the desired output sequence is generated.
+
+45
+00:03:47,000 --> 00:03:53,000
+In the context of the and similar models, we generate images from textual descriptions.
+
+46
+00:03:53,000 --> 00:04:01,000
+The decoder only architecture means that the model takes the textual description as an input and directly
+
+47
+00:04:01,000 --> 00:04:06,000
+generates a corresponding image without needing an encoder to process the input text.
+
+48
+00:04:07,000 --> 00:04:13,000
+The model's decoder is responsible for both understanding the text and generating the image based on
+
+49
+00:04:13,000 --> 00:04:14,000
+the understanding.
+
+50
+00:04:14,000 --> 00:04:21,000
+This architecture simplifies the model structure and makes it more suitable for tasks where the primary
+
+51
+00:04:21,000 --> 00:04:23,000
+goal is to generate content.
+
+52
+00:04:23,000 --> 00:04:28,000
+In this case, images based on textual input descriptions.
+
+53
+00:04:28,000 --> 00:04:35,000
+It eliminates the need for an encoder that might be used in tasks like text understanding or text to
+
+54
+00:04:35,000 --> 00:04:36,000
+text translation.
+
+55
+00:04:37,000 --> 00:04:39,000
+To understand the better.
+
+56
+00:04:39,000 --> 00:04:41,000
+Let's learn its key features.
+
+57
+00:04:42,000 --> 00:04:50,000
+The Li is a remarkable neural network model developed by OpenAI that combines natural language understanding
+
+58
+00:04:50,000 --> 00:04:52,000
+with image generation.
+
+59
+00:04:52,000 --> 00:04:58,000
+It is designed to generate images from textual descriptions and exhibits, several key features and
+
+60
+00:04:58,000 --> 00:04:59,000
+capabilities.
+
+61
+00:05:00,000 --> 00:05:01,000
+Let's review them.
+
+62
+00:05:02,000 --> 00:05:03,000
+Text to image generation.
+
+63
+00:05:03,000 --> 00:05:10,000
+The Li can generate images from textual prompts allowing users to describe a concept or scene in natural
+
+64
+00:05:10,000 --> 00:05:14,000
+language, and the model produces corresponding images.
+
+65
+00:05:14,000 --> 00:05:16,000
+Creative Imagery.
+
+66
+00:05:16,000 --> 00:05:24,000
+One of its standout features is its ability to generate highly creative images based on textual descriptions,
+
+67
+00:05:24,000 --> 00:05:29,000
+often combining concepts in unexpected and novel ways.
+
+68
+00:05:30,000 --> 00:05:32,000
+Large model size.
+
+69
+00:05:32,000 --> 00:05:39,000
+The Li is a substantial model with 12 billion parameters, making it powerful and capable model for
+
+70
+00:05:39,000 --> 00:05:42,000
+generating high quality images.
+
+71
+00:05:43,000 --> 00:05:44,000
+Diverse capabilities.
+
+72
+00:05:45,000 --> 00:05:52,000
+It has a diverse set of capabilities, including creating anthropomorphized versions of animals and
+
+73
+00:05:52,000 --> 00:06:00,000
+objects, combining unrelated concepts in plausible ways, rendering text and applying transformations
+
+74
+00:06:00,000 --> 00:06:02,000
+to existing images.
+
+75
+00:06:03,000 --> 00:06:04,000
+A resolution.
+
+76
+00:06:04,000 --> 00:06:10,000
+The can generate images with high resolution up to 1024 by 1024 pixels.
+
+77
+00:06:10,000 --> 00:06:11,000
+Fine control.
+
+78
+00:06:12,000 --> 00:06:18,000
+The model offers fine control over attributes and positions of objects within the generated images,
+
+79
+00:06:18,000 --> 00:06:23,000
+allowing for specific adjustments such as changing colors, sizes and relative positions.
+
+80
+00:06:24,000 --> 00:06:31,000
+Viewpoint and 3D rendering the leak and control the viewpoint of a scene and render scenes in different
+
+81
+00:06:31,000 --> 00:06:37,000
+3D styles, adding a sense of depth and perspective to the generated images.
+
+82
+00:06:38,000 --> 00:06:40,000
+Variable binding.
+
+83
+00:06:40,000 --> 00:06:47,000
+It can interpret and correctly compose complex textual descriptions involving multiple objects, their
+
+84
+00:06:47,000 --> 00:06:50,000
+attributes and spatial relationships.
+
+85
+00:06:51,000 --> 00:06:57,000
+Combining unrelated concepts, the league can creatively combine unrelated ideas to synthesize unique
+
+86
+00:06:57,000 --> 00:07:03,000
+objects or scenes demonstrating its ability to generate imaginative content.
+
+87
+00:07:04,000 --> 00:07:06,000
+Geographic and temporal knowledge.
+
+88
+00:07:06,000 --> 00:07:13,000
+The model has learned about geographic facts, landmarks, neighborhoods and concepts that vary over
+
+89
+00:07:13,000 --> 00:07:18,000
+time, making it versatile for generating images related to different locations and areas.
+
+90
+00:07:19,000 --> 00:07:25,000
+The Li represents a significant progress in AI's ability to bridge the gap between natural language
+
+91
+00:07:25,000 --> 00:07:28,000
+understanding and image generation.
+
+92
+00:07:29,000 --> 00:07:30,000
+I believe that you may be wondered.
+
+93
+00:07:30,000 --> 00:07:32,000
+Okay, the lead generates pictures.
+
+94
+00:07:32,000 --> 00:07:33,000
+Okay.
+
+95
+00:07:33,000 --> 00:07:35,000
+It converts text to image.
+
+96
+00:07:35,000 --> 00:07:37,000
+But why do I need it?
+
+97
+00:07:37,000 --> 00:07:44,000
+If you ask this question, let's review some common business cases and let's try to understand when
+
+98
+00:07:44,000 --> 00:07:49,000
+it is applicable to use Delhi, in which cases and applications.
+
+99
+00:07:49,000 --> 00:07:55,000
+The cases that we are going to review will be mostly declarative and will describe the cases where text
+
+100
+00:07:55,000 --> 00:08:01,000
+to image models might be used in general and how the model is promoted on the market.
+
+101
+00:08:01,000 --> 00:08:09,000
+I need to make this note because based on today, Delhi still has some challenges with being suitable
+
+102
+00:08:09,000 --> 00:08:14,000
+to most of business cases it's supposed to be used for and during.
+
+103
+00:08:14,000 --> 00:08:21,000
+The lesson will try to understand whether Delhi really can handle cases on the expected level of quality
+
+104
+00:08:21,000 --> 00:08:25,000
+and quality is a key when we talk about transforming models.
+
+105
+00:08:26,000 --> 00:08:30,000
+I tried to brainstorm and group cases by categories.
+
+106
+00:08:30,000 --> 00:08:33,000
+The first category content generation.
+
+107
+00:08:33,000 --> 00:08:41,000
+It includes graphic design, advertising, marketing materials, for example, create custom graphics,
+
+108
+00:08:41,000 --> 00:08:45,000
+logos or illustrations based on textual description.
+
+109
+00:08:45,000 --> 00:08:50,000
+Or generate eye catching visuals and advertisements with the help of the Lee.
+
+110
+00:08:51,000 --> 00:08:55,000
+Create visuals for brochures, posters and social media posts.
+
+111
+00:08:55,000 --> 00:08:58,000
+By the way, you can use the leave for social media content generation.
+
+112
+00:08:58,000 --> 00:09:03,000
+It is really helpful tool in case you need creative visuals.
+
+113
+00:09:03,000 --> 00:09:04,000
+Product design.
+
+114
+00:09:05,000 --> 00:09:12,000
+You can use the li for interior designing, fashion, designing, industrial design, like, for example,
+
+115
+00:09:12,000 --> 00:09:16,000
+to generate concept art for product prototypes.
+
+116
+00:09:16,000 --> 00:09:18,000
+Entertainment and media.
+
+117
+00:09:18,000 --> 00:09:20,000
+Video games.
+
+118
+00:09:20,000 --> 00:09:24,000
+Generate game assets, characters and scenes from narrative descriptions.
+
+119
+00:09:25,000 --> 00:09:26,000
+Film and animation.
+
+120
+00:09:27,000 --> 00:09:31,000
+Create storyboards and concept art for movies and animations.
+
+121
+00:09:31,000 --> 00:09:33,000
+Book Covers.
+
+122
+00:09:33,000 --> 00:09:34,000
+Design Book Cover.
+
+123
+00:09:34,000 --> 00:09:37,000
+Illustrations based on book summaries.
+
+124
+00:09:38,000 --> 00:09:39,000
+Art and Creativity.
+
+125
+00:09:39,000 --> 00:09:45,000
+Artists can use the lead to explore creative ideas and generate novel artwork.
+
+126
+00:09:46,000 --> 00:09:51,000
+Educational resources you can use daily for textbook illustrations.
+
+127
+00:09:51,000 --> 00:09:54,000
+Interactive learning and PowerPoint presentations.
+
+128
+00:09:54,000 --> 00:09:55,000
+For example.
+
+129
+00:09:55,000 --> 00:10:00,000
+If you want, you can use the Li to create visuals for your PowerPoint presentation.
+
+130
+00:10:00,000 --> 00:10:04,000
+This will help you to stay consistent in style of your visuals.
+
+131
+00:10:04,000 --> 00:10:12,000
+Or you can even create a web application that helps to generate slides using chat GPT model for text
+
+132
+00:10:12,000 --> 00:10:15,000
+generation and the Li for image generation.
+
+133
+00:10:15,000 --> 00:10:16,000
+Think about it.
+
+134
+00:10:17,000 --> 00:10:19,000
+Storytelling and writing.
+
+135
+00:10:19,000 --> 00:10:26,000
+You can use the model to generate visual references for scenes and characters in stories, enhance interactive
+
+136
+00:10:26,000 --> 00:10:28,000
+fiction and games with dynamic visuals.
+
+137
+00:10:29,000 --> 00:10:37,000
+Concept prototyping product prototypes generate visual representations of product concepts for early
+
+138
+00:10:37,000 --> 00:10:38,000
+stage prototyping.
+
+139
+00:10:38,000 --> 00:10:44,000
+Visualize architectural design ideas before construction and lots more.
+
+140
+00:10:44,000 --> 00:10:51,000
+Basically any task for visualization, you can challenge the lead to help you and to master the lead.
+
+141
+00:10:51,000 --> 00:10:55,000
+Like a pro, you should know its limitations.
+
+142
+00:10:55,000 --> 00:11:00,000
+Like all models, the Li has its own challenges and limitations.
+
+143
+00:11:00,000 --> 00:11:01,000
+What is a.
+
+144
+00:11:02,000 --> 00:11:04,000
+The first one Data dependency.
+
+145
+00:11:04,000 --> 00:11:09,000
+The li relies on text image pairs it was trained on.
+
+146
+00:11:09,000 --> 00:11:16,000
+It may have difficulty generating images for concepts or scenes that were not well represented in its
+
+147
+00:11:16,000 --> 00:11:17,000
+training data.
+
+148
+00:11:17,000 --> 00:11:21,000
+You should always remember about this ambiguity.
+
+149
+00:11:21,000 --> 00:11:28,000
+If a textual description is ambiguous or can be interpreted in multiple ways, the Li might generate
+
+150
+00:11:28,000 --> 00:11:32,000
+images that align with one interpretation but not another.
+
+151
+00:11:33,000 --> 00:11:35,000
+Complex descriptions.
+
+152
+00:11:35,000 --> 00:11:43,000
+While the league can handle complex descriptions, extremely detailed prompts might not bring you satisfactory
+
+153
+00:11:43,000 --> 00:11:44,000
+results.
+
+154
+00:11:44,000 --> 00:11:45,000
+Image Realism.
+
+155
+00:11:46,000 --> 00:11:52,000
+While the league generates creative images, they may not always appear realistic or high quality,
+
+156
+00:11:52,000 --> 00:11:56,000
+especially for scenes with complex textures and details.
+
+157
+00:11:56,000 --> 00:11:59,000
+And probably this is one of the key challenges.
+
+158
+00:11:59,000 --> 00:12:03,000
+You will understand it better when you will start using the league more actively.
+
+159
+00:12:05,000 --> 00:12:06,000
+Control challenges.
+
+160
+00:12:06,000 --> 00:12:11,000
+Fine grained control over attributes such as color or position can be challenging.
+
+161
+00:12:12,000 --> 00:12:16,000
+The model may not always precisely capture user specified details.
+
+162
+00:12:17,000 --> 00:12:19,000
+Creative interpretations.
+
+163
+00:12:19,000 --> 00:12:23,000
+Daily creativity can be a double edged sword.
+
+164
+00:12:23,000 --> 00:12:31,000
+While it can produce novel and imaginative images, it may not always adhere to conventional or expected
+
+165
+00:12:31,000 --> 00:12:33,000
+interpretations.
+
+166
+00:12:33,000 --> 00:12:34,000
+Interactivity.
+
+167
+00:12:34,000 --> 00:12:41,000
+While the leat generates images based on text prompts, it doesn't have the capability for interactive
+
+168
+00:12:41,000 --> 00:12:46,000
+back and forth conversations like some other AI models by OpenAI.
+
+169
+00:12:46,000 --> 00:12:50,000
+You can't edit generated image with Dall-e.
+
+170
+00:12:51,000 --> 00:12:55,000
+You need to submit another prompt and generate a new image.
+
+171
+00:12:55,000 --> 00:13:02,000
+It's important to be aware of these limitations and use the leave within its intended scope and with
+
+172
+00:13:02,000 --> 00:13:03,000
+a critical eye.
+
+173
+00:13:03,000 --> 00:13:09,000
+OpenAI is the organization behind Dall-e continues to research and improve AI models to address these
+
+174
+00:13:09,000 --> 00:13:16,000
+challenges, but users should exercise caution and apply human judgment when utilizing AI generated
+
+175
+00:13:16,000 --> 00:13:17,000
+content.
+
+176
+00:13:18,000 --> 00:13:24,000
+I believe that we have learned enough Syrian we can proceed with learning details and practical examples
+
+177
+00:13:24,000 --> 00:13:26,000
+in the attachments to the video.
+
+178
+00:13:26,000 --> 00:13:29,000
+You will be able to find link to the API reference.
+
+179
+00:13:29,000 --> 00:13:32,000
+Let me make a brief overview of an API.
+
+180
+00:13:33,000 --> 00:13:38,000
+The first important thing that I would like to mention is about three key methods for interacting with
+
+181
+00:13:38,000 --> 00:13:39,000
+the Li API.
+
+182
+00:13:39,000 --> 00:13:44,000
+They are creating images from scratch based on a text prompt.
+
+183
+00:13:44,000 --> 00:13:48,000
+Creating edits of an existing image based on a new text prompt.
+
+184
+00:13:49,000 --> 00:13:51,000
+Creating variations of an existing image.
+
+185
+00:13:52,000 --> 00:13:56,000
+And here you can find three end points in the API reference.
+
+186
+00:13:56,000 --> 00:13:58,000
+Let's start from the simple one.
+
+187
+00:13:58,000 --> 00:14:00,000
+Image creation.
+
+188
+00:14:00,000 --> 00:14:05,000
+Basically, you just need to send the prompt with a post message to the specific resource.
+
+189
+00:14:05,000 --> 00:14:11,000
+Also, I prepared for you postman collection that you can just import into your postman and execute
+
+190
+00:14:11,000 --> 00:14:12,000
+queries.
+
+191
+00:14:12,000 --> 00:14:18,000
+You can either import my postman collection or create a new request from scratch.
+
+192
+00:14:18,000 --> 00:14:22,000
+It doesn't matter because we don't have super complex requests here.
+
+193
+00:14:23,000 --> 00:14:29,000
+After importing postman collection, don't forget to add your OpenAI API key.
+
+194
+00:14:29,000 --> 00:14:32,000
+I use variable for API key.
+
+195
+00:14:32,000 --> 00:14:35,000
+You have to add it into your authorization header.
+
+196
+00:14:36,000 --> 00:14:43,000
+Content type is application Json because OpenAI API will reply us with a generated URL.
+
+197
+00:14:44,000 --> 00:14:46,000
+Let's go to the body now.
+
+198
+00:14:46,000 --> 00:14:49,000
+There is only one required attribute that is prompt.
+
+199
+00:14:50,000 --> 00:14:54,000
+You need to describe what you want to see on the generated picture.
+
+200
+00:14:54,000 --> 00:14:56,000
+Let's try something creative.
+
+201
+00:14:56,000 --> 00:15:03,000
+For example, an astronaut rides a motorcycle on Mars in photorealistic style.
+
+202
+00:15:03,000 --> 00:15:09,000
+As you already understood, you can specify details in the prompt like colors, position of elements
+
+203
+00:15:09,000 --> 00:15:12,000
+or even style like I did now.
+
+204
+00:15:13,000 --> 00:15:16,000
+And we'll see how the imagines this.
+
+205
+00:15:16,000 --> 00:15:21,000
+Another attribute that is called n, it is the number of items to be generated.
+
+206
+00:15:21,000 --> 00:15:24,000
+It can be between 1 and 10.
+
+207
+00:15:24,000 --> 00:15:31,000
+By default, it is equal to one, but decided to keep it here and request just for the learning purposes.
+
+208
+00:15:31,000 --> 00:15:33,000
+The next attribute is size.
+
+209
+00:15:34,000 --> 00:15:45,000
+I selected 1024 by 1024 pixels, which is default value, but you can also set here to 156 by 256 or
+
+210
+00:15:45,000 --> 00:15:48,000
+512 by 512.
+
+211
+00:15:49,000 --> 00:15:55,000
+This size will impact costs per image in the model you pay per image generated.
+
+212
+00:15:56,000 --> 00:16:01,000
+As always, you can check the most relevant and up to date pricing on the pricing page.
+
+213
+00:16:01,000 --> 00:16:06,000
+As you can see, pricing will depend on the image size that you want to generate.
+
+214
+00:16:06,000 --> 00:16:09,000
+So I configure it size.
+
+215
+00:16:09,000 --> 00:16:16,000
+There is also user attributes that you can use as a unique identifier representing your end user, which
+
+216
+00:16:16,000 --> 00:16:20,000
+can help OpenAI to monitor and detect abuse.
+
+217
+00:16:20,000 --> 00:16:25,000
+In our example, I would just use an email as a fake unique identifier.
+
+218
+00:16:26,000 --> 00:16:32,000
+Do not forget that this attribute is optional, and if you don't need it, you can just omit it.
+
+219
+00:16:33,000 --> 00:16:36,000
+And also there is an attribute that is called response format.
+
+220
+00:16:36,000 --> 00:16:44,000
+It can be just one out of two possible options is a URL or b64 Json.
+
+221
+00:16:45,000 --> 00:16:51,000
+If you use b64 Json, the image is encoded in b64 json.
+
+222
+00:16:51,000 --> 00:16:53,000
+Not a normal image file type.
+
+223
+00:16:53,000 --> 00:16:57,000
+By default it is URL and I'm going to show you how it works.
+
+224
+00:16:58,000 --> 00:17:05,000
+Let me vividly leave this attribute here, but as I said, it is optional and you can rely on default
+
+225
+00:17:05,000 --> 00:17:06,000
+values.
+
+226
+00:17:06,000 --> 00:17:10,000
+Let's send this query now and see what we would receive.
+
+227
+00:17:11,000 --> 00:17:19,000
+We received an URL, we can copy this URL and paste it into the browser and see what we received.
+
+228
+00:17:19,000 --> 00:17:25,000
+Once you reviewed the image in the browser, you can save it to the local computer.
+
+229
+00:17:25,000 --> 00:17:29,000
+As you can see, images may be of different quality.
+
+230
+00:17:29,000 --> 00:17:32,000
+Some lines may be not perfect.
+
+231
+00:17:32,000 --> 00:17:38,000
+That's why sometimes you can just change the number of images to generate and just try one more time.
+
+232
+00:17:39,000 --> 00:17:43,000
+For example, you can ask the lead to generate ten images per prompt.
+
+233
+00:17:43,000 --> 00:17:50,000
+In this case, you are going to have higher chances that at least some options will look as you expected.
+
+234
+00:17:51,000 --> 00:17:52,000
+Attention.
+
+235
+00:17:52,000 --> 00:17:57,000
+I recommend you to save images just in case, because they can be lost easily.
+
+236
+00:17:57,000 --> 00:18:01,000
+The link is available only for a limited amount of time.
+
+237
+00:18:01,000 --> 00:18:07,000
+Some time passed when you use the link to open image, the link will be broken.
+
+238
+00:18:07,000 --> 00:18:14,000
+I believe that is done on purpose because in case OpenAI would store all possible generated images,
+
+239
+00:18:14,000 --> 00:18:19,000
+it wouldn't be enough space to support image generation in the whole world on daily basis.
+
+240
+00:18:20,000 --> 00:18:22,000
+That's how easy you can generate images.
+
+241
+00:18:23,000 --> 00:18:28,000
+Let's review the second method of interaction with the OpenAI API.
+
+242
+00:18:29,000 --> 00:18:33,000
+Creating edits of an existing image based on a new text prompt.
+
+243
+00:18:34,000 --> 00:18:35,000
+Before we review.
+
+244
+00:18:36,000 --> 00:18:38,000
+Let me explain you how edits work.
+
+245
+00:18:38,000 --> 00:18:45,000
+The image editing endpoint offers the capability to enhance and modify an image through the inclusion
+
+246
+00:18:45,000 --> 00:18:48,000
+of a mask during the upload process.
+
+247
+00:18:48,000 --> 00:18:56,000
+Transparent regions within the mask serve as guidelines for where adjustments should be made to the
+
+248
+00:18:56,000 --> 00:18:56,000
+image.
+
+249
+00:18:57,000 --> 00:19:04,000
+Additionally, it's important to note that the prompt should describe the whole new image rather than
+
+250
+00:19:04,000 --> 00:19:07,000
+focusing solely on the erased spaces.
+
+251
+00:19:08,000 --> 00:19:12,000
+On the slide you can see example from OpenAI Dall-e documentation.
+
+252
+00:19:12,000 --> 00:19:14,000
+Pay attention to the original image.
+
+253
+00:19:15,000 --> 00:19:18,000
+Then create mask by saying mask.
+
+254
+00:19:18,000 --> 00:19:22,000
+I mean creation of transparent areas on the same image.
+
+255
+00:19:22,000 --> 00:19:27,000
+You can do this in Photoshop, for example, or in any other image editor.
+
+256
+00:19:28,000 --> 00:19:38,000
+PNG file format allows you to save transparency on the image and then describe in the prompt is a whole
+
+257
+00:19:38,000 --> 00:19:44,000
+picture, a sunlit indoor lounge area with a pool containing a flamingo.
+
+258
+00:19:44,000 --> 00:19:50,000
+As you can see, Dall-e gave us back edited image taken into consideration.
+
+259
+00:19:50,000 --> 00:19:52,000
+Our prompt is in the school.
+
+260
+00:19:53,000 --> 00:19:59,000
+Both the uploaded image and the related mosque must adhere to specific criteria.
+
+261
+00:19:59,000 --> 00:20:06,000
+They should be square PNG images with file sizes under four megabytes and their dimensions must match
+
+262
+00:20:06,000 --> 00:20:07,000
+each other.
+
+263
+00:20:08,000 --> 00:20:15,000
+It's worth highlighting that the non transparent sections of the mosque do not play a role in generating
+
+264
+00:20:15,000 --> 00:20:16,000
+the output.
+
+265
+00:20:16,000 --> 00:20:22,000
+As a result, there is no strict requirement for them to align precisely with the original image.
+
+266
+00:20:22,000 --> 00:20:25,000
+Unlike the example provided on the slide.
+
+267
+00:20:26,000 --> 00:20:32,000
+Now let's take a look at demo and learn how this works in practice, especially for you.
+
+268
+00:20:32,000 --> 00:20:36,000
+I prepared a set of original images and related images with mask.
+
+269
+00:20:36,000 --> 00:20:43,000
+This can save your time during the preparing examples to test an API, you can find images for this
+
+270
+00:20:43,000 --> 00:20:47,000
+example in attachments to the lesson in the postman.
+
+271
+00:20:47,000 --> 00:20:49,000
+You can find template requests for edits.
+
+272
+00:20:50,000 --> 00:20:57,000
+There are few attributes here that you can configure and only two out of all these attributes are required,
+
+273
+00:20:57,000 --> 00:21:00,000
+so only image and prompt attributes are required.
+
+274
+00:21:01,000 --> 00:21:04,000
+Image attribute is an object a file.
+
+275
+00:21:04,000 --> 00:21:10,000
+The image to edit must be a valid PNG file less than four megabytes and square.
+
+276
+00:21:11,000 --> 00:21:15,000
+And you see that mask attribute is an optional one.
+
+277
+00:21:15,000 --> 00:21:21,000
+If mask is not provided, image must have transparency which will be used as the mask.
+
+278
+00:21:22,000 --> 00:21:26,000
+I'm going to show you this example too, at the end of the lesson.
+
+279
+00:21:26,000 --> 00:21:33,000
+In postman I select form data and here for each attribute I can change attribute type.
+
+280
+00:21:33,000 --> 00:21:39,000
+You can see that for image and for mask attribute I selected file type.
+
+281
+00:21:39,000 --> 00:21:44,000
+Just hover mouse over the field and change the selector here.
+
+282
+00:21:45,000 --> 00:21:48,000
+And now I can select file as a value.
+
+283
+00:21:49,000 --> 00:21:55,000
+For mask attribute, you need to select image with transparent area like I showed you on the slide.
+
+284
+00:21:55,000 --> 00:22:02,000
+In a minute we are going to review a dataset that I prepared for you and results that we received.
+
+285
+00:22:02,000 --> 00:22:11,000
+So Mask is an additional image whose fully transparent areas where alpha is zero indicate where image
+
+286
+00:22:11,000 --> 00:22:17,000
+should be edited, must be a valid PNG file less than four megabytes and have the same dimensions as
+
+287
+00:22:17,000 --> 00:22:18,000
+image.
+
+288
+00:22:19,000 --> 00:22:21,000
+Who is prompt attribute everything a symbol.
+
+289
+00:22:21,000 --> 00:22:25,000
+It is a text description of the desired image.
+
+290
+00:22:25,000 --> 00:22:30,000
+And as we already discussed, you need to describe the whole image.
+
+291
+00:22:30,000 --> 00:22:32,000
+But not only the transparent part.
+
+292
+00:22:33,000 --> 00:22:37,000
+There's only one important thing that I didn't mention yet is a prompt limit.
+
+293
+00:22:38,000 --> 00:22:41,000
+This is a maximum length is 1000 characters.
+
+294
+00:22:42,000 --> 00:22:47,000
+An attribute is for amount of images to be generated must be between 1 and 10.
+
+295
+00:22:48,000 --> 00:22:55,000
+Size attribute sets the size of generated images the same as with image generation.
+
+296
+00:22:55,000 --> 00:22:59,000
+There are three size variations are supported at the moment.
+
+297
+00:22:59,000 --> 00:23:02,000
+256 by 256.
+
+298
+00:23:02,000 --> 00:23:08,000
+512 by 512 and ten, 24 by 1024.
+
+299
+00:23:08,000 --> 00:23:15,000
+And in the similar way, like in Create API that we already reviewed, we have response format with
+
+300
+00:23:15,000 --> 00:23:17,000
+two possible values and user.
+
+301
+00:23:18,000 --> 00:23:21,000
+These are all available attributes based on today.
+
+302
+00:23:21,000 --> 00:23:23,000
+So let's get back to the postman now.
+
+303
+00:23:24,000 --> 00:23:30,000
+I configured all attributes in the form data and I'm ready to send the request.
+
+304
+00:23:30,000 --> 00:23:38,000
+In response, I receive URL that I can copy and paste to the browser for better visualization and better
+
+305
+00:23:38,000 --> 00:23:40,000
+comparison of input and output.
+
+306
+00:23:40,000 --> 00:23:44,000
+Let me present the results with the help of slides.
+
+307
+00:23:44,000 --> 00:23:47,000
+Let's start from the successful example.
+
+308
+00:23:47,000 --> 00:23:55,000
+I used Unsplash to download royalty free images and work with them and I tried to find the similar image
+
+309
+00:23:55,000 --> 00:23:57,000
+like I showed you from OpenAI.
+
+310
+00:23:57,000 --> 00:23:59,000
+API documentation.
+
+311
+00:23:59,000 --> 00:24:06,000
+You can see here we also have swimming pool and the graphic is a cartoon like style.
+
+312
+00:24:06,000 --> 00:24:09,000
+As you may see, it looks unrealistic.
+
+313
+00:24:10,000 --> 00:24:17,000
+And in case I submit this page with the mask like you can see on the slide and prompt swimming pool
+
+314
+00:24:17,000 --> 00:24:20,000
+with Flamingo, then I receive such result.
+
+315
+00:24:20,000 --> 00:24:26,000
+And to be honest, this is the best result that I managed to achieve based on today.
+
+316
+00:24:26,000 --> 00:24:33,000
+Edit Feature of Dali is more for marketing purposes rather than to handle real use cases.
+
+317
+00:24:33,000 --> 00:24:39,000
+I really believe that this will be improved over time, but it is not the fact, based on the day of
+
+318
+00:24:39,000 --> 00:24:41,000
+the creation of this video lesson.
+
+319
+00:24:41,000 --> 00:24:42,000
+Why I say so?
+
+320
+00:24:42,000 --> 00:24:49,000
+First of all, to receive an image of at least this quality, I had to generate few images.
+
+321
+00:24:49,000 --> 00:24:53,000
+You can also see the previous generated version, which is not good.
+
+322
+00:24:54,000 --> 00:24:59,000
+The second thing is look at the examples that I say is the most successful one.
+
+323
+00:24:59,000 --> 00:25:06,000
+If you would scale it and start look to the details, you can see that the drawing itself is not the
+
+324
+00:25:06,000 --> 00:25:07,000
+best.
+
+325
+00:25:08,000 --> 00:25:15,000
+Somewhere lines are broken, somewhere it just looks not natural or other things.
+
+326
+00:25:15,000 --> 00:25:23,000
+And the third thing why I think that added endpoint in the list, super raw, is the generated images.
+
+327
+00:25:23,000 --> 00:25:27,000
+Let's review a few more examples for the next case.
+
+328
+00:25:27,000 --> 00:25:33,000
+I wanted to edit the image in order it would look like a man riding a bicycle in the park in autumn.
+
+329
+00:25:34,000 --> 00:25:40,000
+You can see that the lead tried to draw a man riding a bike, but at the same time we have to admit
+
+330
+00:25:40,000 --> 00:25:45,000
+that it didn't succeed, at least as I expected.
+
+331
+00:25:45,000 --> 00:25:49,000
+Sometimes the league simply ignores my prompts.
+
+332
+00:25:49,000 --> 00:25:50,000
+Like I asked.
+
+333
+00:25:50,000 --> 00:25:51,000
+Nothing.
+
+334
+00:25:51,000 --> 00:25:59,000
+For example, I expected to have a small yard here on the image, but instead of yard is a transparent
+
+335
+00:25:59,000 --> 00:26:02,000
+area was filled out with the sea and that's it.
+
+336
+00:26:03,000 --> 00:26:08,000
+Also, I wanted to paste a street musician on this image in first option.
+
+337
+00:26:08,000 --> 00:26:13,000
+The list simply ignored my prompt and just draw a wall almost like it was.
+
+338
+00:26:13,000 --> 00:26:19,000
+And during the second attempt, the league created something what was supposed to look like a musician.
+
+339
+00:26:19,000 --> 00:26:21,000
+But apparently it is not.
+
+340
+00:26:22,000 --> 00:26:29,000
+And one more example is about a group of people who works together in the office in this imaginary example
+
+341
+00:26:29,000 --> 00:26:32,000
+imagines that we want to substitute one colleague.
+
+342
+00:26:32,000 --> 00:26:38,000
+For example, our colleague took a day off that day, but came to the team building activities and Pizza
+
+343
+00:26:38,000 --> 00:26:40,000
+Friday party.
+
+344
+00:26:40,000 --> 00:26:45,000
+So I want to edit this image with the Dali in order to erase this colleague.
+
+345
+00:26:46,000 --> 00:26:49,000
+As you can see, result is not realistic at all.
+
+346
+00:26:49,000 --> 00:26:57,000
+And even for cartoon like style or for painting, it doesn't look mature enough to be shared with somebody.
+
+347
+00:26:58,000 --> 00:27:03,000
+You also know that the mask attribute is not required and not a mandatory one.
+
+348
+00:27:03,000 --> 00:27:09,000
+We learned this during an API review, but let's try now how it works in practice.
+
+349
+00:27:10,000 --> 00:27:13,000
+What would be in case I would upload just mask straight away.
+
+350
+00:27:13,000 --> 00:27:19,000
+In such case, model shouldn't understand that it should draw a person on that place.
+
+351
+00:27:19,000 --> 00:27:20,000
+What do you think?
+
+352
+00:27:21,000 --> 00:27:28,000
+So in case I upload image with transparent areas into an image attribute and don't send any mask at
+
+353
+00:27:28,000 --> 00:27:31,000
+all, we receive the following results.
+
+354
+00:27:31,000 --> 00:27:37,000
+I would say that the first one is slightly better now, but it still draws a person during the second
+
+355
+00:27:37,000 --> 00:27:38,000
+attempt.
+
+356
+00:27:38,000 --> 00:27:45,000
+So now you understand my conclusion about the lee, but at least we learned an API which is most likely
+
+357
+00:27:45,000 --> 00:27:48,000
+will remain the same or at least similar.
+
+358
+00:27:49,000 --> 00:27:54,000
+And with the new release of the Lee, you can try the added feature one more time.
+
+359
+00:27:55,000 --> 00:27:56,000
+Who knows?
+
+360
+00:27:56,000 --> 00:28:03,000
+Probably next version of added feature in the Lee will work much better and the set method of interaction
+
+361
+00:28:03,000 --> 00:28:07,000
+with the Lee API is about variations of an existing image.
+
+362
+00:28:07,000 --> 00:28:09,000
+Everything is simple here.
+
+363
+00:28:09,000 --> 00:28:15,000
+Imagine that you have an image and you need to create another variation of the image.
+
+364
+00:28:15,000 --> 00:28:17,000
+You may have a different motivation.
+
+365
+00:28:17,000 --> 00:28:24,000
+Is it to get inspiration or just have a legal workaround about using the original image and create its
+
+366
+00:28:24,000 --> 00:28:26,000
+unique variation?
+
+367
+00:28:26,000 --> 00:28:28,000
+This is just an example.
+
+368
+00:28:28,000 --> 00:28:30,000
+There can be different cases.
+
+369
+00:28:31,000 --> 00:28:35,000
+Let's review API reference and we'll try to understand it better.
+
+370
+00:28:35,000 --> 00:28:39,000
+We have one required attribute that is image.
+
+371
+00:28:39,000 --> 00:28:41,000
+Requirements are similar.
+
+372
+00:28:41,000 --> 00:28:50,000
+Image should be a valid PNG file square and not more than four megabytes and all other attributes like
+
+373
+00:28:50,000 --> 00:28:52,000
+n size response format.
+
+374
+00:28:52,000 --> 00:29:00,000
+We already discussed during the review of previous endpoints, so let's check how it looks like in Postman
+
+375
+00:29:00,000 --> 00:29:03,000
+in the Postman collection that you can find in attachment.
+
+376
+00:29:03,000 --> 00:29:07,000
+You will be able also to find request for image variation endpoint.
+
+377
+00:29:07,000 --> 00:29:09,000
+Nothing special here.
+
+378
+00:29:09,000 --> 00:29:14,000
+For the sake of example, let's make variation for different images.
+
+379
+00:29:15,000 --> 00:29:19,000
+Mona Lisa, probably the most famous piece of art in the world.
+
+380
+00:29:19,000 --> 00:29:24,000
+Let's ask the lead to make its own variation of this masterpiece.
+
+381
+00:29:24,000 --> 00:29:32,000
+And as you can see, it is hard to beat Da Vinci in general when you are dealing with human images using
+
+382
+00:29:32,000 --> 00:29:32,000
+Dali.
+
+383
+00:29:33,000 --> 00:29:39,000
+It works approximately in the way like you can see on the slides, whether it is good or bad, it is
+
+384
+00:29:39,000 --> 00:29:40,000
+you to decide.
+
+385
+00:29:40,000 --> 00:29:42,000
+Because all of us have different tastes.
+
+386
+00:29:43,000 --> 00:29:49,000
+And you remember we had example with the pork, as you can see, in case there are no requirements to
+
+387
+00:29:49,000 --> 00:29:52,000
+draw human or extra small details.
+
+388
+00:29:52,000 --> 00:29:55,000
+Dali can handle such cases.
+
+389
+00:29:55,000 --> 00:29:59,000
+So in this case, we can say that we receive expected variations.
+
+390
+00:30:00,000 --> 00:30:02,000
+So what would be the conclusion?
+
+391
+00:30:02,000 --> 00:30:07,000
+As you can see, I always try to avoid making conclusions instead of you.
+
+392
+00:30:07,000 --> 00:30:16,000
+Instead, I tried to do the best what I can to give you 360 degree overview of the topic and let you
+
+393
+00:30:16,000 --> 00:30:18,000
+form your own opinion.
+
+394
+00:30:18,000 --> 00:30:24,000
+But if you are interested, I can just share my opinion about Delhi in general, and this is related
+
+395
+00:30:24,000 --> 00:30:26,000
+to all Delhi features.
+
+396
+00:30:26,000 --> 00:30:31,000
+The lower requirements for details you have the better really?
+
+397
+00:30:31,000 --> 00:30:35,000
+Because you saw that not all scenarios Delhi can handle with ease.
+
+398
+00:30:36,000 --> 00:30:43,000
+Obviously it has huge advantages and it is really an amazing thing for brainstorming, for inspiration
+
+399
+00:30:43,000 --> 00:30:45,000
+and other things.
+
+400
+00:30:45,000 --> 00:30:51,000
+For example, I used the lead to take inspiration during the design of the logo for one of my projects.
+
+401
+00:30:51,000 --> 00:30:59,000
+I was wondered what associations can arise with different words, and Delhi helped me with this goal.
+
+402
+00:31:00,000 --> 00:31:05,000
+So really a lot will depend on your expectations and how you plan to use Delhi.
+
+403
+00:31:05,000 --> 00:31:11,000
+And today we did a useful deep dive into API in order to help you understand how you can use Delhi in
+
+404
+00:31:11,000 --> 00:31:12,000
+your work.
+
+405
+00:31:12,000 --> 00:31:16,000
+So let's recap now what we have learned during the lesson.
+
+406
+00:31:17,000 --> 00:31:19,000
+We learned a lot of things today.
+
+407
+00:31:19,000 --> 00:31:22,000
+Now you know what the Lee model is.
+
+408
+00:31:22,000 --> 00:31:27,000
+I explained what a decoder and encoder are in transforming models.
+
+409
+00:31:27,000 --> 00:31:34,000
+This is some basic terminology and will help you with understanding of the topic and architecture of
+
+410
+00:31:34,000 --> 00:31:35,000
+other models.
+
+411
+00:31:35,000 --> 00:31:39,000
+We've used key features of Delhi to help you understand the topic better.
+
+412
+00:31:39,000 --> 00:31:46,000
+We learned business use cases when the league can be applied, and then we reviewed documentation and
+
+413
+00:31:46,000 --> 00:31:51,000
+practical examples about three ways of interaction with the league API.
+
+414
+00:31:51,000 --> 00:31:59,000
+Namely, we learned how to create images based on text prompt, how to edit images based on a text prompt,
+
+415
+00:31:59,000 --> 00:32:03,000
+and how to create variations based on an existing image.
+
+416
+00:32:03,000 --> 00:32:08,000
+And for each case, we reviewed documentation and practical examples.
+
+417
+00:32:08,000 --> 00:32:11,000
+That's all what I wanted to share with you today.
+
+418
+00:32:11,000 --> 00:32:13,000
+Thanks a lot for your attention.
+
+419
+00:32:13,000 --> 00:32:16,000
+Have a great day and see you in the next lesson.
+

102 - DALL-E - Text to image AI Model by OpenAI/001 Examples-of-images-and-masks.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/tree/master/dall-e

102 - DALL-E - Text to image AI Model by OpenAI/001 Postman-collection-used-in-lesson.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/blob/master/dall-e/OpenAI%20DALL-E.postman_collection.json

102 - DALL-E - Text to image AI Model by OpenAI/001 Pricing.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://openai.com/pricing

102 - DALL-E - Text to image AI Model by OpenAI/external-links.txt

@@ -0,0 +1,12 @@
+
+001 API-Reference-for-DALL-E-Model
+https://platform.openai.com/docs/api-reference/images
+
+001 Postman-collection-used-in-lesson
+https://github.com/AndriiPiatakha/openai-learnit/blob/master/dall-e/OpenAI%20DALL-E.postman_collection.json
+
+001 Examples-of-images-and-masks
+https://github.com/AndriiPiatakha/openai-learnit/tree/master/dall-e
+
+001 Pricing
+https://openai.com/pricing

103 - Whisper - Speech to text AI model by OpenAI/001 API-Reference.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://platform.openai.com/docs/api-reference/audio

103 - Whisper - Speech to text AI model by OpenAI/001 Audio-file-1-used-in-the-lesson-for-transcription-demo.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/blob/master/whisper/01_demo_file.mp3

103 - Whisper - Speech to text AI model by OpenAI/001 Audio-file-2-used-in-the-lesson-for-translation-demo.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/blob/master/whisper/02_demo_file.mp3

103 - Whisper - Speech to text AI model by OpenAI/001 Postman-collection.url

@@ -0,0 +1,2 @@
+[InternetShortcut]
+URL=https://github.com/AndriiPiatakha/openai-learnit/blob/master/whisper/OpenAI%20Whisper.postman_collection.json

103 - Whisper - Speech to text AI model by OpenAI/001 Whisper Model & API Overview With Examples in Postman_en.srt

@@ -0,0 +1,1552 @@
+1
+00:00:05,000 --> 00:00:06,000
+Hello, Tim.
+
+2
+00:00:06,000 --> 00:00:11,000
+In this lesson we are going to talk about such OpenAI model as Whisper.
+
+3
+00:00:11,000 --> 00:00:16,000
+We are going to start lesson from the general overview of OpenAI whisper model.
+
+4
+00:00:17,000 --> 00:00:22,000
+I will explain you its key features to understand what a whisper model is.
+
+5
+00:00:22,000 --> 00:00:26,000
+We will review business use cases when whisper may be applicable.
+
+6
+00:00:26,000 --> 00:00:33,000
+Also I will tell you limitations of the model that you need to consider if you plan to work with Whisper.
+
+7
+00:00:33,000 --> 00:00:39,000
+While learning about Whisper, we are going to learn some basic concepts about speech recognition as
+
+8
+00:00:39,000 --> 00:00:41,000
+word error rate.
+
+9
+00:00:41,000 --> 00:00:46,000
+Also in the lesson, we are going to have a lot of practical exercises.
+
+10
+00:00:46,000 --> 00:00:53,000
+We will learn whisper API from the documentation and when we will complete review of the API, we will
+
+11
+00:00:53,000 --> 00:00:55,000
+proceed with practical examples.
+
+12
+00:00:55,000 --> 00:01:00,000
+I'm going to show you transcription and translation demo with Whisper API.
+
+13
+00:01:00,000 --> 00:01:07,000
+We'll use different attributes and will understand how using of different attributes can impact the
+
+14
+00:01:07,000 --> 00:01:08,000
+response.
+
+15
+00:01:08,000 --> 00:01:10,000
+Let's start our lesson.
+
+16
+00:01:10,000 --> 00:01:13,000
+Let's understand first what a whisper model is.
+
+17
+00:01:13,000 --> 00:01:18,000
+Whisper is an automatic speech recognition system developed by OpenAI.
+
+18
+00:01:19,000 --> 00:01:24,000
+Trained on a vast and diverse dataset collected from the Internet.
+
+19
+00:01:24,000 --> 00:01:34,000
+Whisper is trained on an extensive dataset comprising 680,000 hours of multilingual and multi-task supervised
+
+20
+00:01:34,000 --> 00:01:35,000
+data.
+
+21
+00:01:35,000 --> 00:01:43,000
+This large and diverse dataset contributes to its robustness in handling various accents, background
+
+22
+00:01:43,000 --> 00:01:47,000
+noise, technical vocabulary and multiple languages.
+
+23
+00:01:47,000 --> 00:01:54,000
+The whisper architecture follows a simple end to end approach implemented as an encoder decoder.
+
+24
+00:01:54,000 --> 00:02:04,000
+Transformer input audio is divided into 32nd segments, converted into log mel spectrograms and processed
+
+25
+00:02:04,000 --> 00:02:05,000
+through an encoder.
+
+26
+00:02:06,000 --> 00:02:07,000
+What is log?
+
+27
+00:02:07,000 --> 00:02:09,000
+Mel spectrogram a log.
+
+28
+00:02:09,000 --> 00:02:17,000
+Mel spectrogram short for logarithmic mel frequency spectrogram is a representation of the spectrum
+
+29
+00:02:17,000 --> 00:02:24,000
+of a signal typically an audio signal that is widely used in speech and audio processing.
+
+30
+00:02:24,000 --> 00:02:32,000
+It is a valuable tool for various tasks such as speech recognition, music analysis and sound classification.
+
+31
+00:02:32,000 --> 00:02:36,000
+At this stage, I don't see a lot of benefits for you going deeper in Learn of Log.
+
+32
+00:02:36,000 --> 00:02:37,000
+Mel Spectrogram.
+
+33
+00:02:38,000 --> 00:02:43,000
+If you will be interested, you can type your questions and comments to this video and I will be happy
+
+34
+00:02:43,000 --> 00:02:44,000
+to answer.
+
+35
+00:02:44,000 --> 00:02:46,000
+Let's learn more about Whisper System.
+
+36
+00:02:47,000 --> 00:02:55,000
+After audio was divided into 32nd segments and processed through an encoder and decoder of Whisper is
+
+37
+00:02:55,000 --> 00:03:02,000
+then trained to predict the corresponding text captions with special tokens to direct the model's tasks
+
+38
+00:03:02,000 --> 00:03:09,000
+such as language identification, timestamps, multilingual transcription and speech translation to
+
+39
+00:03:09,000 --> 00:03:14,000
+English due to its training on a large and diverse dataset.
+
+40
+00:03:14,000 --> 00:03:20,000
+Whisper demonstrates improved robustness compared to other similar models.
+
+41
+00:03:20,000 --> 00:03:28,000
+It makes fewer errors when tested on various datasets, highlighting its adaptability and versatility.
+
+42
+00:03:28,000 --> 00:03:36,000
+Whispers dataset includes a significant portion of non-English audio and it can transcribe in the original
+
+43
+00:03:36,000 --> 00:03:39,000
+language or translate it into English.
+
+44
+00:03:39,000 --> 00:03:47,000
+This approach is found to be effective for speech, to text translation and outperforms supervised state
+
+45
+00:03:47,000 --> 00:03:50,000
+of the art models in certain translation tasks.
+
+46
+00:03:51,000 --> 00:03:54,000
+To understand it better what a whisper model is.
+
+47
+00:03:54,000 --> 00:03:57,000
+Let's review its key features.
+
+48
+00:03:57,000 --> 00:03:59,000
+The first one.
+
+49
+00:03:59,000 --> 00:04:00,000
+Transcription.
+
+50
+00:04:00,000 --> 00:04:06,000
+Whisper can transcribe audio into text written in the original language of the audio.
+
+51
+00:04:06,000 --> 00:04:09,000
+The second one translation.
+
+52
+00:04:09,000 --> 00:04:16,000
+In addition to transcription, Whisper can translate audio from supported languages into English, making
+
+53
+00:04:16,000 --> 00:04:20,000
+it a valuable tool for multilingual applications.
+
+54
+00:04:20,000 --> 00:04:22,000
+File format Support.
+
+55
+00:04:22,000 --> 00:04:27,000
+Whisper supports multiple input and output file formats.
+
+56
+00:04:27,000 --> 00:04:39,000
+The file can be of the following formats Flac, MP3, MP4 and PAC, Mpeg, M4A, Ogg, Vov or Wepm.
+
+57
+00:04:40,000 --> 00:04:43,000
+It can handle various common audio file types.
+
+58
+00:04:44,000 --> 00:04:46,000
+Huge amount of supported languages.
+
+59
+00:04:46,000 --> 00:04:55,000
+Whisper supports a wide range of languages, including Afrikaans, Arabic, Chinese, French, English,
+
+60
+00:04:55,000 --> 00:05:01,000
+German, Japanese, Korean, Russian, Spanish and many more.
+
+61
+00:05:01,000 --> 00:05:06,000
+These languages are supported for both transcriptions and translations.
+
+62
+00:05:07,000 --> 00:05:14,000
+Handling longer inputs, while the default limit for input files is 25MB.
+
+63
+00:05:14,000 --> 00:05:18,000
+Whisper provides guidance on how to handle longer audio files.
+
+64
+00:05:18,000 --> 00:05:25,000
+Users can split longer files into chunks of 25MB or less for processing.
+
+65
+00:05:26,000 --> 00:05:28,000
+Prompting for improved quality.
+
+66
+00:05:28,000 --> 00:05:34,000
+Users can use prompts to improve the quality of transcripts generated by whisper.
+
+67
+00:05:34,000 --> 00:05:40,000
+Prompts allow for some control over the style and content of the generated text.
+
+68
+00:05:41,000 --> 00:05:43,000
+Reliability improvement techniques.
+
+69
+00:05:43,000 --> 00:05:51,000
+Whisper offers techniques to address common challenges, such as recognizing uncommon words or acronyms.
+
+70
+00:05:51,000 --> 00:05:57,000
+Prompting and specific prompt styles can be used to enhance the reliability of transcriptions.
+
+71
+00:05:57,000 --> 00:06:04,000
+As you can see, Whisper is designed to be a versatile tool for converting audio into text and support
+
+72
+00:06:04,000 --> 00:06:11,000
+applications in various fields, including transcription services, language translation and more.
+
+73
+00:06:11,000 --> 00:06:17,000
+And before we jump to learn the details, let's understand business use cases.
+
+74
+00:06:17,000 --> 00:06:25,000
+Or in other words, why Whisper model might be interested to you and when you can use it and in which
+
+75
+00:06:25,000 --> 00:06:26,000
+cases?
+
+76
+00:06:26,000 --> 00:06:32,000
+Let's brainstorm and let me suggest a few ideas for application and start up.
+
+77
+00:06:32,000 --> 00:06:39,000
+The whisper model can be applied in various scenarios depending on your specific needs and use cases.
+
+78
+00:06:39,000 --> 00:06:43,000
+Here are some situations where you might find the whisper model useful.
+
+79
+00:06:43,000 --> 00:06:47,000
+Automatic speech recognition, for example.
+
+80
+00:06:47,000 --> 00:06:48,000
+Transcription Services.
+
+81
+00:06:49,000 --> 00:06:52,000
+If you need to convert recordings into text.
+
+82
+00:06:52,000 --> 00:06:57,000
+Whisper can be used to transcribe these audio inputs accurately.
+
+83
+00:06:57,000 --> 00:07:00,000
+This can be helpful in messaging apps.
+
+84
+00:07:00,000 --> 00:07:06,000
+For example, such messenger as telegram has premium features that allows to convert audio messages
+
+85
+00:07:06,000 --> 00:07:07,000
+into text.
+
+86
+00:07:07,000 --> 00:07:15,000
+This is helpful in case person who received audio message don't want or can't listen to an audio message
+
+87
+00:07:15,000 --> 00:07:16,000
+at the moment.
+
+88
+00:07:17,000 --> 00:07:18,000
+Voice assistance.
+
+89
+00:07:18,000 --> 00:07:25,000
+Whisper can power the speech recognition component of voice assistance or chat bots, enabling them
+
+90
+00:07:25,000 --> 00:07:30,000
+to understand and respond to spoken commands or questions.
+
+91
+00:07:30,000 --> 00:07:32,000
+Or imagine another case.
+
+92
+00:07:32,000 --> 00:07:39,000
+You developed an application similar to the one we created in our course application for automation
+
+93
+00:07:39,000 --> 00:07:41,000
+of project management activities.
+
+94
+00:07:41,000 --> 00:07:44,000
+And you don't want to give commands via a chat.
+
+95
+00:07:44,000 --> 00:07:48,000
+You want to send audio message, for example, like this.
+
+96
+00:07:48,000 --> 00:07:52,000
+Create a user story to implement chat bot and assign it to Andre.
+
+97
+00:07:52,000 --> 00:07:54,000
+And that's it.
+
+98
+00:07:54,000 --> 00:08:00,000
+Our application will use Whisper to recognize audio and convert speech to text.
+
+99
+00:08:00,000 --> 00:08:05,000
+Then speech will be sent to the chat GPT, then chat.
+
+100
+00:08:05,000 --> 00:08:10,000
+GPT will understand that we want to make a function call and then we'll be back to the scenarios that
+
+101
+00:08:10,000 --> 00:08:12,000
+we already implemented in the previous lesson.
+
+102
+00:08:13,000 --> 00:08:15,000
+The second business use case.
+
+103
+00:08:15,000 --> 00:08:21,000
+Multilingual applications like language translation applications, for example.
+
+104
+00:08:22,000 --> 00:08:28,000
+If you want to create applications that translate spoken content from multiple languages into English,
+
+105
+00:08:28,000 --> 00:08:35,000
+whisper can be a valuable component for handling the transcription and translation tasks.
+
+106
+00:08:35,000 --> 00:08:41,000
+For example, you have multilingual team you can use Whisper to boost your communication.
+
+107
+00:08:42,000 --> 00:08:47,000
+Everyone can speak in their original language and it will be translated to English.
+
+108
+00:08:47,000 --> 00:08:50,000
+Another use case Multilingual transcription.
+
+109
+00:08:50,000 --> 00:08:57,000
+Whisper can be used to transcribe audio content in various languages, making it suitable for applications
+
+110
+00:08:57,000 --> 00:09:00,000
+involving multilingual audio data.
+
+111
+00:09:01,000 --> 00:09:05,000
+The group of accessibility and captioning use cases.
+
+112
+00:09:05,000 --> 00:09:10,000
+Imagine closed captioning for video content.
+
+113
+00:09:10,000 --> 00:09:17,000
+Whisper can be used to generate closed captions making your videos accessible to individuals with hearing
+
+114
+00:09:17,000 --> 00:09:20,000
+disabilities or for content localization.
+
+115
+00:09:21,000 --> 00:09:22,000
+Accessibility tools.
+
+116
+00:09:23,000 --> 00:09:29,000
+It can be integrated into accessibility tools that converts spoken content in real time into text,
+
+117
+00:09:29,000 --> 00:09:33,000
+helping individuals with hearing disabilities.
+
+118
+00:09:33,000 --> 00:09:36,000
+Content indexing and search.
+
+119
+00:09:36,000 --> 00:09:43,000
+Whisper can assist in indexing and searching audio or video content by transcribing the audio.
+
+120
+00:09:43,000 --> 00:09:48,000
+You can make the content searchable based on its spoken words and phrases.
+
+121
+00:09:49,000 --> 00:09:50,000
+Voice Data analysis.
+
+122
+00:09:51,000 --> 00:09:52,000
+Market Research.
+
+123
+00:09:52,000 --> 00:10:00,000
+Use Case Analyzing recorded phone interviews, customer service calls or user feedback can provide valuable
+
+124
+00:10:00,000 --> 00:10:01,000
+insights.
+
+125
+00:10:01,000 --> 00:10:06,000
+Whisper can transcribe and then using chat GPT post-processing.
+
+126
+00:10:06,000 --> 00:10:09,000
+You can analyze these audio records.
+
+127
+00:10:09,000 --> 00:10:11,000
+Voice Data Mining.
+
+128
+00:10:11,000 --> 00:10:18,000
+Whisper can help organizations extract useful information from voice data for business intelligence,
+
+129
+00:10:18,000 --> 00:10:20,000
+quality control and trend analysis.
+
+130
+00:10:21,000 --> 00:10:22,000
+Meeting minutes.
+
+131
+00:10:23,000 --> 00:10:28,000
+Imagine that you have productive meeting and you don't want to write meeting minutes in case you have
+
+132
+00:10:28,000 --> 00:10:30,000
+audio recording of the meeting.
+
+133
+00:10:30,000 --> 00:10:37,000
+You can use Whisper model to convert audio to text and then ask ChatGPT to structure meeting minutes
+
+134
+00:10:37,000 --> 00:10:38,000
+on your request.
+
+135
+00:10:39,000 --> 00:10:45,000
+Assign action items to responsible team members and even to send an email to accountable team members.
+
+136
+00:10:46,000 --> 00:10:53,000
+And we already had the lesson in the course about how to generate and send email using ChatGPT and custom
+
+137
+00:10:53,000 --> 00:10:53,000
+application.
+
+138
+00:10:54,000 --> 00:10:55,000
+Language learning.
+
+139
+00:10:56,000 --> 00:11:02,000
+If you develop in language learning platforms or applications, whisper can be used to transcribe and
+
+140
+00:11:02,000 --> 00:11:10,000
+translate spoken content, facilitating language acquisition, podcasts and media production.
+
+141
+00:11:10,000 --> 00:11:17,000
+Whisper can be used to generate transcripts for podcasts or spoken content, making it easier to create
+
+142
+00:11:17,000 --> 00:11:21,000
+written summaries, articles, or searchable show notes.
+
+143
+00:11:22,000 --> 00:11:23,000
+Content Moderation.
+
+144
+00:11:23,000 --> 00:11:30,000
+Whisper can be part of content moderation systems helping identify and flag inappropriate or offensive
+
+145
+00:11:30,000 --> 00:11:32,000
+audio content.
+
+146
+00:11:32,000 --> 00:11:34,000
+Research and data collection.
+
+147
+00:11:35,000 --> 00:11:42,000
+Researchers can use Whisper to transcribe interviews, field recordings or oral history, making it
+
+148
+00:11:42,000 --> 00:11:45,000
+easier to analyze and store data.
+
+149
+00:11:45,000 --> 00:11:52,000
+Custom applications You can integrate, whisper into custom applications or workflows tailored to your
+
+150
+00:11:52,000 --> 00:11:54,000
+specific needs.
+
+151
+00:11:54,000 --> 00:11:59,000
+For example, creating a custom voice assistant for industry specific tasks.
+
+152
+00:11:59,000 --> 00:12:05,000
+Keep in mind that whisper, like any tool, has its strengths and limitations.
+
+153
+00:12:05,000 --> 00:12:14,000
+It excels in handling diverse accents, languages and audio conditions, but may not be perfect in all
+
+154
+00:12:14,000 --> 00:12:15,000
+situations.
+
+155
+00:12:15,000 --> 00:12:22,000
+As you can see, Whisper can be a valuable tool for converting spoken language into text and opening
+
+156
+00:12:22,000 --> 00:12:27,000
+up a wide range of possibilities for audio data processing and analysis.
+
+157
+00:12:27,000 --> 00:12:34,000
+Even despite all advantages and features, it is always worse to know the existing limitations of the
+
+158
+00:12:34,000 --> 00:12:34,000
+model.
+
+159
+00:12:34,000 --> 00:12:38,000
+So let me share with you limitations that you need to consider.
+
+160
+00:12:38,000 --> 00:12:46,000
+File Size Limitation Whispers API supports file uploads of up to 25MB.
+
+161
+00:12:46,000 --> 00:12:53,000
+Larger audio files need to be split into smaller chunks or compressed, potentially leading to context
+
+162
+00:12:53,000 --> 00:12:55,000
+loss if not handled carefully.
+
+163
+00:12:56,000 --> 00:13:03,000
+I'm talking about cases when you split audio file in the middle of sentence or in the middle of a word.
+
+164
+00:13:03,000 --> 00:13:08,000
+To avoid such cases, you can use special libraries to divide audio properly.
+
+165
+00:13:09,000 --> 00:13:17,000
+English only translation is a translation feature of whisper is limited to translating audio into English.
+
+166
+00:13:17,000 --> 00:13:20,000
+It doesn't support translation into other languages.
+
+167
+00:13:20,000 --> 00:13:24,000
+In my opinion, this is probably one of the most critical limitation.
+
+168
+00:13:25,000 --> 00:13:27,000
+Limited language support.
+
+169
+00:13:27,000 --> 00:13:35,000
+While the whisper model was trained on 98 languages, it may not achieve high accuracy for all of them.
+
+170
+00:13:35,000 --> 00:13:42,000
+Only languages with a word error rate of less than 50% are officially supported.
+
+171
+00:13:43,000 --> 00:13:49,000
+After this slide, I'm going to explain you what a word error rate is in more details.
+
+172
+00:13:50,000 --> 00:13:52,000
+Limited audio format support.
+
+173
+00:13:53,000 --> 00:14:05,000
+Whisper supports specific audio file formats Flac, MP3, MP4 and pack Mpeg, M4A, Ogg, Vov or WebP
+
+174
+00:14:05,000 --> 00:14:05,000
+Em.
+
+175
+00:14:06,000 --> 00:14:09,000
+Using unsupported formats can be a limitation.
+
+176
+00:14:10,000 --> 00:14:12,000
+Segmented Audio Context.
+
+177
+00:14:12,000 --> 00:14:20,000
+Whisper considers only the final 224 tokens of a prompt for context when dealing with segmented audio.
+
+178
+00:14:20,000 --> 00:14:28,000
+This limitation can impact the handling of long and complex segments, prompting system limitations
+
+179
+00:14:28,000 --> 00:14:31,000
+while prompts can improve transcription quality.
+
+180
+00:14:31,000 --> 00:14:37,000
+The whisper model's prompting system is more limited compared to other language models.
+
+181
+00:14:37,000 --> 00:14:40,000
+It provides less control over generated audio.
+
+182
+00:14:41,000 --> 00:14:43,000
+How to use prompts in whisper.
+
+183
+00:14:43,000 --> 00:14:47,000
+We We're going to learn later in this lesson.
+
+184
+00:14:47,000 --> 00:14:54,000
+Style marching while the model adapts to the style of the prompt, it may not always match the desired
+
+185
+00:14:54,000 --> 00:15:01,000
+writing style for languages with variations such as simplified and traditional Chinese.
+
+186
+00:15:01,000 --> 00:15:06,000
+For example, recognition of uncommon words and acronyms.
+
+187
+00:15:06,000 --> 00:15:13,000
+Whisper may not always recognize uncommon words or acronyms accurately in the audio prompts can help
+
+188
+00:15:13,000 --> 00:15:15,000
+address this limitation.
+
+189
+00:15:16,000 --> 00:15:18,000
+His ability of third party software.
+
+190
+00:15:19,000 --> 00:15:26,000
+Whisper recommends using third party software like Pi DOP for handling longer audio inputs.
+
+191
+00:15:27,000 --> 00:15:34,000
+We already talked about risk in case you would divide audio file into multiple chunks, you can interrupt
+
+192
+00:15:34,000 --> 00:15:38,000
+the audience the middle of the sentence and even in the middle of the word.
+
+193
+00:15:38,000 --> 00:15:43,000
+And this may impact consistency of the transcription and translation.
+
+194
+00:15:43,000 --> 00:15:52,000
+However, OpenAI makes no guarantees about usability or security of third party libraries, so these
+
+195
+00:15:52,000 --> 00:15:56,000
+are important limitations to consider when you are dealing with whisper model.
+
+196
+00:15:56,000 --> 00:16:00,000
+And I hope with the new versions of Whisper, they will be addressed.
+
+197
+00:16:00,000 --> 00:16:01,000
+So stay tuned.
+
+198
+00:16:02,000 --> 00:16:07,000
+I also promised to explain to you such concepts as word error rate.
+
+199
+00:16:07,000 --> 00:16:12,000
+I believe it will be useful for your general education of speech recognition.
+
+200
+00:16:12,000 --> 00:16:19,000
+Word error rate is a measure used to evaluate the accuracy of automatic speech recognition systems.
+
+201
+00:16:19,000 --> 00:16:27,000
+In simple terms, it calculates the percentage of words that are incorrect or have errors in the transcribed
+
+202
+00:16:27,000 --> 00:16:31,000
+output compared to the reference or original text.
+
+203
+00:16:32,000 --> 00:16:42,000
+A lower word error rate indicates higher accuracy, while a higher word error rate suggests more errors
+
+204
+00:16:42,000 --> 00:16:43,000
+in the transcription.
+
+205
+00:16:43,000 --> 00:16:51,000
+For example, if an automatic speech recognition system transcribes a sentence with three errors out
+
+206
+00:16:51,000 --> 00:17:00,000
+of ten words, the word error rate would be 30% because three out of ten words are incorrect or mismatched.
+
+207
+00:17:00,000 --> 00:17:08,000
+Word error rate provides a straightforward way to assess how well an automatic speech recognition system
+
+208
+00:17:08,000 --> 00:17:12,000
+performs in converting spoken language into written text.
+
+209
+00:17:12,000 --> 00:17:20,000
+And now let's learn API of whisper model and implement simple scenarios in order to understand how to
+
+210
+00:17:20,000 --> 00:17:22,000
+interact with whisper model.
+
+211
+00:17:22,000 --> 00:17:27,000
+As you already learned from the slides, there are two main scenarios for whisper model.
+
+212
+00:17:27,000 --> 00:17:31,000
+They are transcription and translation.
+
+213
+00:17:32,000 --> 00:17:37,000
+I'm going to use Postman and send request from Postman to demo Whisper API.
+
+214
+00:17:38,000 --> 00:17:40,000
+Let's start from the transcription.
+
+215
+00:17:41,000 --> 00:17:45,000
+I created test audio file in the MP3 format as it want to.
+
+216
+00:17:45,000 --> 00:17:46,000
+Transcript.
+
+217
+00:17:46,000 --> 00:17:52,000
+With the help of Whisper model in attachments to the lesson, you can find all files to proceed with
+
+218
+00:17:52,000 --> 00:17:53,000
+this scenario.
+
+219
+00:17:54,000 --> 00:17:56,000
+I added necessary audio files.
+
+220
+00:17:56,000 --> 00:18:00,000
+I also added export of my postman collection.
+
+221
+00:18:01,000 --> 00:18:07,000
+So for this example use file with the name zero one underscore demo underscore file.
+
+222
+00:18:08,000 --> 00:18:12,000
+You can download it and put it in any location on your computer.
+
+223
+00:18:13,000 --> 00:18:16,000
+Then you can import Postman collection.
+
+224
+00:18:16,000 --> 00:18:20,000
+I had a separate lesson about how to work with Postman.
+
+225
+00:18:20,000 --> 00:18:23,000
+That's why I wouldn't stop on this too much.
+
+226
+00:18:23,000 --> 00:18:29,000
+In case you would need advice regarding Postman, let me know in comments below the video and I will
+
+227
+00:18:29,000 --> 00:18:30,000
+be happy to answer.
+
+228
+00:18:31,000 --> 00:18:33,000
+After you imported postman collection.
+
+229
+00:18:33,000 --> 00:18:42,000
+Do not forget to configure API key because in postman collection I use OpenAI key variable and you need
+
+230
+00:18:42,000 --> 00:18:47,000
+to configure OpenAI API key in order to be able to execute API queries.
+
+231
+00:18:48,000 --> 00:18:51,000
+Also another important thing in attachments.
+
+232
+00:18:51,000 --> 00:18:55,000
+You will be able to find the link to the official API reference.
+
+233
+00:18:55,000 --> 00:19:03,000
+OpenAI is famous for dynamic pace of development and sometimes ignorance of backward compatibility with
+
+234
+00:19:03,000 --> 00:19:04,000
+new releases.
+
+235
+00:19:04,000 --> 00:19:11,000
+That's why I recommend you always to check the content in the lesson with the latest updates in the
+
+236
+00:19:11,000 --> 00:19:19,000
+official API documentation in case there will be some changes, use latest API from the documentation.
+
+237
+00:19:20,000 --> 00:19:24,000
+Here we can see that this is a post request to this URL.
+
+238
+00:19:25,000 --> 00:19:32,000
+As you can see here in API reference, it is specified that we have to mandatory and required attributes.
+
+239
+00:19:32,000 --> 00:19:36,000
+They are file and model in the file.
+
+240
+00:19:36,000 --> 00:19:41,000
+You need to pass the real file object, not just the name of the file.
+
+241
+00:19:41,000 --> 00:19:48,000
+In this video I'm going to show you how to pass the file object in the postman today and how to pass
+
+242
+00:19:48,000 --> 00:19:49,000
+it from the code.
+
+243
+00:19:49,000 --> 00:19:52,000
+I will show you in another lesson.
+
+244
+00:19:52,000 --> 00:19:59,000
+As you can see, file can be of different formats like we already discussed when we talked about limitations.
+
+245
+00:19:59,000 --> 00:20:11,000
+The file can be of the following formats flac, mp3, mp4, Mpeg, M4A, ogg Vov or WebP.
+
+246
+00:20:11,000 --> 00:20:12,000
+M.
+
+247
+00:20:12,000 --> 00:20:14,000
+For the model attribute.
+
+248
+00:20:14,000 --> 00:20:15,000
+We just need to pass value.
+
+249
+00:20:15,000 --> 00:20:17,000
+Whisper Dash one.
+
+250
+00:20:17,000 --> 00:20:23,000
+Currently, and based on today, OpenAI has just one model for speech to text.
+
+251
+00:20:23,000 --> 00:20:26,000
+Probably in the nearest future there will be updates and there will be new.
+
+252
+00:20:26,000 --> 00:20:27,000
+Models.
+
+253
+00:20:27,000 --> 00:20:34,000
+But so far you can just put whisper dash one as a attributes are optional and we are going to use them
+
+254
+00:20:34,000 --> 00:20:37,000
+a little bit later in this lesson.
+
+255
+00:20:37,000 --> 00:20:42,000
+Let's just try to get a transcription with just required attributes.
+
+256
+00:20:43,000 --> 00:20:49,000
+Here in Postman, you can import the collection that I shared in attachments to the video or just create
+
+257
+00:20:49,000 --> 00:20:51,000
+this request from scratch.
+
+258
+00:20:51,000 --> 00:20:59,000
+It wouldn't be super complex, so we need to create a post request URL you can take from the API reference
+
+259
+00:20:59,000 --> 00:21:02,000
+page that we just reviewed in headers.
+
+260
+00:21:02,000 --> 00:21:10,000
+You need to specify authorization header and insert your OpenAI API key here and content type.
+
+261
+00:21:10,000 --> 00:21:13,000
+Content type is multi-part form data.
+
+262
+00:21:14,000 --> 00:21:18,000
+Now we need to configure body of our request.
+
+263
+00:21:18,000 --> 00:21:22,000
+Open Body tap select form data.
+
+264
+00:21:22,000 --> 00:21:25,000
+Here we have file and model attributes.
+
+265
+00:21:26,000 --> 00:21:30,000
+Pay attention that here you can specify type of the value.
+
+266
+00:21:30,000 --> 00:21:35,000
+If you hover mouse in this field, you can change text to file type.
+
+267
+00:21:35,000 --> 00:21:37,000
+That is exactly what we need.
+
+268
+00:21:37,000 --> 00:21:44,000
+And in the value field you can select audio file that you can also download from attachments to the
+
+269
+00:21:44,000 --> 00:21:47,000
+video or create a new file if you want.
+
+270
+00:21:47,000 --> 00:21:49,000
+That's it.
+
+271
+00:21:49,000 --> 00:21:50,000
+Pretty simple, don't you think?
+
+272
+00:21:50,000 --> 00:21:54,000
+So Now we are ready to send the request.
+
+273
+00:21:54,000 --> 00:21:58,000
+Let me click Send button and here it is.
+
+274
+00:21:58,000 --> 00:22:05,000
+We received a response By default, the data is returned in the Json format.
+
+275
+00:22:05,000 --> 00:22:13,000
+If you want, you can select any other supported format like for example, text SRT, verbose Json or
+
+276
+00:22:13,000 --> 00:22:14,000
+V double T.
+
+277
+00:22:15,000 --> 00:22:18,000
+Let me show you an example of verbose Json.
+
+278
+00:22:19,000 --> 00:22:27,000
+I just add one more attribute that is called response format, and I write verbose Json as value here
+
+279
+00:22:27,000 --> 00:22:29,000
+and I send request one more time.
+
+280
+00:22:30,000 --> 00:22:37,000
+Verbose Json gives you a lot of meta information like for example, task language, duration of the
+
+281
+00:22:37,000 --> 00:22:40,000
+audio array of segments and tokens.
+
+282
+00:22:40,000 --> 00:22:48,000
+You know that OpenAI language models operate on tokens rather than on the whole words or characters.
+
+283
+00:22:48,000 --> 00:22:53,000
+And in the previous lesson we discussed this tokenized model and other details.
+
+284
+00:22:53,000 --> 00:22:57,000
+So if you are interested you can use verbose mode.
+
+285
+00:22:57,000 --> 00:23:02,000
+For example, this can be useful if you want to gather some analytics.
+
+286
+00:23:03,000 --> 00:23:08,000
+Like I said, it is up to you which format works the best for your business needs.
+
+287
+00:23:08,000 --> 00:23:11,000
+You can make transcription on various languages.
+
+288
+00:23:12,000 --> 00:23:18,000
+The demo file number two that you can find in attachments to the lesson is recorded in the Russian language
+
+289
+00:23:19,000 --> 00:23:26,000
+and if you would select it for transcription and send request, then you would see that language is
+
+290
+00:23:26,000 --> 00:23:31,000
+auto detected and transcript is in the original language.
+
+291
+00:23:31,000 --> 00:23:33,000
+That's amazing, don't you think so?
+
+292
+00:23:34,000 --> 00:23:37,000
+And the quality of transcription is really high.
+
+293
+00:23:38,000 --> 00:23:42,000
+Let's review other attributes that we have available for this endpoint.
+
+294
+00:23:42,000 --> 00:23:43,000
+Prompt.
+
+295
+00:23:43,000 --> 00:23:47,000
+You can use prompt to improve the quality of transcripts.
+
+296
+00:23:47,000 --> 00:23:53,000
+Important thing to mention here is that prompt should be written in the same language as audio.
+
+297
+00:23:53,000 --> 00:24:01,000
+The model adapts its style based on the provided prompt, including capitalization and punctuation.
+
+298
+00:24:01,000 --> 00:24:08,000
+When you provide a prompt, the model adjusts its style to match, including factors like capitalization
+
+299
+00:24:08,000 --> 00:24:09,000
+and punctuation.
+
+300
+00:24:10,000 --> 00:24:16,000
+Note that the prompting system in whisper is more limited compared to other language models.
+
+301
+00:24:16,000 --> 00:24:20,000
+It is not just GPT model and you need to consider this limitation.
+
+302
+00:24:20,000 --> 00:24:27,000
+Despite its limitations, prompts offer valuable advantages for refining transcriptions.
+
+303
+00:24:27,000 --> 00:24:32,000
+Let's review some examples when you may need to use, prompting for transcription.
+
+304
+00:24:32,000 --> 00:24:35,000
+Correcting Misrecognized words.
+
+305
+00:24:36,000 --> 00:24:39,000
+Use prompts to address specific mis recognitions.
+
+306
+00:24:39,000 --> 00:24:48,000
+For example, the Li or GPT four Just specify to help the model understand the sound better and transcript
+
+307
+00:24:48,000 --> 00:24:52,000
+it with higher quality context preservation.
+
+308
+00:24:52,000 --> 00:24:57,000
+Maintain context when dealing with segmented audio files.
+
+309
+00:24:57,000 --> 00:25:04,000
+We already discussed limitations and I believe you can understand that in case that single audio file
+
+310
+00:25:04,000 --> 00:25:12,000
+was segmented into multiple files, there is a risk of losing context of what was said in the previous
+
+311
+00:25:12,000 --> 00:25:12,000
+transcription.
+
+312
+00:25:13,000 --> 00:25:18,000
+To improve this, we can pass prompt with a transcript of the preceding segment.
+
+313
+00:25:19,000 --> 00:25:23,000
+Model considers the final 224 tokens of the prompt.
+
+314
+00:25:24,000 --> 00:25:25,000
+Punctuation.
+
+315
+00:25:25,000 --> 00:25:29,000
+Sometimes during the transcription model can skip punctuation.
+
+316
+00:25:29,000 --> 00:25:37,000
+To ensure proper punctuation, you need just submit simple prompt with punctuation marks as simple as
+
+317
+00:25:37,000 --> 00:25:39,000
+Hello comma.
+
+318
+00:25:39,000 --> 00:25:41,000
+Welcome to my lecture point.
+
+319
+00:25:43,000 --> 00:25:50,000
+Retained filler words model can skip some common filler words, but if you want to keep them in the
+
+320
+00:25:50,000 --> 00:25:55,000
+transcript, you need just to use prompt that contains these filler words.
+
+321
+00:25:55,000 --> 00:25:57,000
+For example.
+
+322
+00:25:57,000 --> 00:26:00,000
+Um, let me think like, Hmm.
+
+323
+00:26:00,000 --> 00:26:05,000
+Okay, here is what I'm like thinking.
+
+324
+00:26:06,000 --> 00:26:07,000
+Language style.
+
+325
+00:26:07,000 --> 00:26:14,000
+Certain languages can have variations in writing styles like simplified and traditional Chinese.
+
+326
+00:26:14,000 --> 00:26:20,000
+The model may not consistently adapt the desired writing style for your transcript by default.
+
+327
+00:26:20,000 --> 00:26:25,000
+You can improve this by using a prompt in your preferred writing style.
+
+328
+00:26:25,000 --> 00:26:28,000
+Let's continue review of other attributes.
+
+329
+00:26:28,000 --> 00:26:35,000
+So far we learned and reviewed response format and prompt temperature attribute is a similar one to
+
+330
+00:26:35,000 --> 00:26:37,000
+other language models.
+
+331
+00:26:37,000 --> 00:26:42,000
+In simple words, it impacts randomness of transcription.
+
+332
+00:26:42,000 --> 00:26:50,000
+When you usually do transcription, randomness or creativity, it is not something what you are looking
+
+333
+00:26:50,000 --> 00:26:50,000
+for.
+
+334
+00:26:50,000 --> 00:26:51,000
+Agree.
+
+335
+00:26:51,000 --> 00:26:58,000
+That's why I personally myself don't use this attribute too often for transcription cases, but you
+
+336
+00:26:58,000 --> 00:27:01,000
+can consider it for your business case if you wish.
+
+337
+00:27:02,000 --> 00:27:06,000
+And the last attribute here in the list is the language attribute.
+
+338
+00:27:06,000 --> 00:27:11,000
+This attribute is used to tell model what language is used in audio.
+
+339
+00:27:11,000 --> 00:27:15,000
+You can describe language via two letter code.
+
+340
+00:27:15,000 --> 00:27:19,000
+According to the ISO standard 639 one.
+
+341
+00:27:19,000 --> 00:27:21,000
+Basically that's it.
+
+342
+00:27:21,000 --> 00:27:22,000
+Regarding transcription.
+
+343
+00:27:22,000 --> 00:27:30,000
+Now, when you know API, you can use this API with any programming language and built in this functionality
+
+344
+00:27:30,000 --> 00:27:33,000
+into your business case scenario and business flow.
+
+345
+00:27:34,000 --> 00:27:39,000
+Let's now review translation feature and API for translation.
+
+346
+00:27:39,000 --> 00:27:47,000
+As you already learned, translation API has its own limitations, and one of its limitations is that
+
+347
+00:27:47,000 --> 00:27:50,000
+it translates only into English.
+
+348
+00:27:50,000 --> 00:27:55,000
+So I need to have audio file with the speech in another language.
+
+349
+00:27:55,000 --> 00:28:01,000
+So that's why you need to download another audio file for the following example.
+
+350
+00:28:01,000 --> 00:28:08,000
+The second audio file that you can find in attachments contains my voice speaking in Russian, and we
+
+351
+00:28:08,000 --> 00:28:15,000
+want to use Whisper model in order to get translation of what was said in the audio in the Postman collection
+
+352
+00:28:15,000 --> 00:28:16,000
+that I have shared with you.
+
+353
+00:28:16,000 --> 00:28:20,000
+You can find another request that is called translation.
+
+354
+00:28:21,000 --> 00:28:24,000
+You can open it or create a new one from scratch.
+
+355
+00:28:25,000 --> 00:28:27,000
+The URL is different.
+
+356
+00:28:27,000 --> 00:28:30,000
+I send request to translations resource.
+
+357
+00:28:31,000 --> 00:28:36,000
+I also change file in order to select file where I speak in Russian.
+
+358
+00:28:36,000 --> 00:28:43,000
+In the model attribute I leave Whisper one because OpenAI has just one model for speech to text so far
+
+359
+00:28:43,000 --> 00:28:45,000
+and for response format.
+
+360
+00:28:45,000 --> 00:28:51,000
+I would add text value just to remind you that the response format is optional.
+
+361
+00:28:51,000 --> 00:28:54,000
+All other configurations are similar.
+
+362
+00:28:54,000 --> 00:29:00,000
+I mean, you need also to specify authorization header and content type header in the similar way like
+
+363
+00:29:00,000 --> 00:29:02,000
+we configured in the previous example.
+
+364
+00:29:03,000 --> 00:29:09,000
+So we are ready to send the request and in response I receive a translated text.
+
+365
+00:29:09,000 --> 00:29:10,000
+Amazing.
+
+366
+00:29:10,000 --> 00:29:12,000
+Looks really cool.
+
+367
+00:29:12,000 --> 00:29:16,000
+I can confirm that translation is absolutely correct.
+
+368
+00:29:16,000 --> 00:29:24,000
+So with Whisper model we receive an amazing helper in audio translation and you can use this API inside
+
+369
+00:29:24,000 --> 00:29:26,000
+your applications.
+
+370
+00:29:27,000 --> 00:29:32,000
+Regarding additional attributes and customization, I can say that there are no specific attributes
+
+371
+00:29:32,000 --> 00:29:35,000
+besides the ones we already reviewed.
+
+372
+00:29:35,000 --> 00:29:39,000
+Response format prompt that should be in English.
+
+373
+00:29:39,000 --> 00:29:42,000
+And this is the difference attention.
+
+374
+00:29:42,000 --> 00:29:50,000
+Because if you remember in transcription API prompt should match the audio language and in translation
+
+375
+00:29:50,000 --> 00:29:52,000
+API it should be in English.
+
+376
+00:29:53,000 --> 00:29:56,000
+And temperature attribute that is also already familiar to you.
+
+377
+00:29:57,000 --> 00:30:00,000
+Basically, that's all what I wanted to share with you in this lesson.
+
+378
+00:30:00,000 --> 00:30:07,000
+And just to remind you that in case you have any questions, please let me know below this video.
+
+379
+00:30:07,000 --> 00:30:11,000
+Now let's recap what we have learned in this video.
+
+380
+00:30:12,000 --> 00:30:15,000
+We learned what an OpenAI whisper model is.
+
+381
+00:30:15,000 --> 00:30:20,000
+I explained its key features, reviewed business use cases.
+
+382
+00:30:20,000 --> 00:30:27,000
+Also, I explained you limitations of the models that you need to be aware about and consider when you
+
+383
+00:30:27,000 --> 00:30:28,000
+work with Whisper.
+
+384
+00:30:28,000 --> 00:30:35,000
+Together we learned API documentation of Whisper model and we saw practical examples of transcription
+
+385
+00:30:35,000 --> 00:30:37,000
+and translation.
+
+386
+00:30:37,000 --> 00:30:42,000
+I hope you liked this lesson and that it was interesting for you.
+
+387
+00:30:42,000 --> 00:30:44,000
+Thanks a lot for your attention.
+
+388
+00:30:44,000 --> 00:30:47,000
+I wish you a great day and see you in the next lesson.
+