Spaces:

Djacon
/

text2feature

Running

App Files Files Community

Djacon commited on Dec 1, 2023

Commit

fbf745c

1 Parent(s): 2498dd5

Update metric, fix youtube.py

Browse files

Files changed (9) hide show

extract.py +0 -1
main.py +0 -1
static/404.html +15 -8
static/analytics.html +15 -8
static/emotion_detection.html +15 -8
static/grammar_checker.html +15 -8
static/index.html +140 -8
static/text_summarizer.html +15 -8
youtube.py +4 -4

extract.py CHANGED Viewed

@@ -1,7 +1,6 @@
 import os
 import nltk
 nltk.data.path.append(os.path.join(os.getcwd(), 'nltk-data'))
-print(nltk.data.path)
 from nltk.cluster.util import cosine_distance
 from nltk.tokenize import sent_tokenize, word_tokenize

 import os
 import nltk
 nltk.data.path.append(os.path.join(os.getcwd(), 'nltk-data'))
 from nltk.cluster.util import cosine_distance
 from nltk.tokenize import sent_tokenize, word_tokenize

main.py CHANGED Viewed

@@ -8,7 +8,6 @@ from fastapi.templating import Jinja2Templates
 from youtube import get_youtube_caption
 from inference import predict_emotions, predict_summarization, predict_acceptance
-MAX_ITER_SIZE = 3000
 app = FastAPI(docs_url=None, redoc_url=None)
 app.mount("/files", StaticFiles(directory="files"), name="files")

 from youtube import get_youtube_caption
 from inference import predict_emotions, predict_summarization, predict_acceptance
 app = FastAPI(docs_url=None, redoc_url=None)
 app.mount("/files", StaticFiles(directory="files"), name="files")

static/404.html CHANGED Viewed

@@ -12,15 +12,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

static/analytics.html CHANGED Viewed

@@ -15,15 +15,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

static/emotion_detection.html CHANGED Viewed

@@ -12,15 +12,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

static/grammar_checker.html CHANGED Viewed

@@ -12,15 +12,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

static/index.html CHANGED Viewed

@@ -12,15 +12,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"
@@ -335,6 +342,131 @@
                                 </p>
                             </div>
                         </section>
                     </div>
                 </div>
             </main>

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"
                                 </p>
                             </div>
                         </section>
+                        <!-- <section class="dark-block p-4 mt-8 bg-white rounded-lg shadow-sm xl:p-8">
+                            <h2 class="text-lg font-medium text-gray-700 capitalize sm:text-xl md:text-2xl">Projects</h2>
+                            <p class="flex items-center mt-2 text-gray-500">
+                                <svg xmlns="http://www.w3.org/2000/svg" class="no-invert w-5 h-5 text-green-500" viewBox="0 0 20 20" fill="currentColor">
+                                    <path fill-rule="evenodd" d="M6.267 3.455a3.066 3.066 0 001.745-.723 3.066 3.066 0 013.976 0 3.066 3.066 0 001.745.723 3.066 3.066 0 012.812 2.812c.051.643.304 1.254.723 1.745a3.066 3.066 0 010 3.976 3.066 3.066 0 00-.723 1.745 3.066 3.066 0 01-2.812 2.812 3.066 3.066 0 00-1.745.723 3.066 3.066 0 01-3.976 0 3.066 3.066 0 00-1.745-.723 3.066 3.066 0 01-2.812-2.812 3.066 3.066 0 00-.723-1.745 3.066 3.066 0 010-3.976 3.066 3.066 0 00.723-1.745 3.066 3.066 0 012.812-2.812zm7.44 5.252a1 1 0 00-1.414-1.414L9 10.586 7.707 9.293a1 1 0 00-1.414 1.414l2 2a1 1 0 001.414 0l4-4z" clip-rule="evenodd"></path>
+                                </svg>
+                                <span class="mx-2">30 done this month</span>
+                            </p>
+                            <div class="dark-block flex flex-col mt-8">
+                                <div class="overflow-x-auto">
+                                    <div class="inline-block min-w-full align-middle">
+                                        <div class="overflow-hidden">
+                                            <table class="min-w-full divide-y divide-gray-200">
+                                                <thead>
+                                                    <tr>
+                                                        <th scope="col" class="px-6 py-3 text-sm font-medium tracking-wider text-left text-gray-500 uppercase rtl:text-right">
+                                                            Project Name
+                                                        </th>
+                                                        <th scope="col" class="px-6 py-3 text-sm font-medium tracking-wider text-left text-gray-500 uppercase rtl:text-right">
+                                                            Project Stack
+                                                        </th>
+                                                        <th scope="col" class="px-6 py-3 text-sm font-medium tracking-wider text-left text-gray-500 uppercase rtl:text-right">
+                                                            Date
+                                                        </th>
+                                                        <th scope="col" class="px-6 py-3 text-sm font-medium tracking-wider text-left text-gray-500 uppercase rtl:text-right">
+                                                            COMPLETION
+                                                        </th>
+                                                    </tr>
+                                                </thead>
+                                                <tbody class="bg-white divide-y divide-gray-200">
+                                                    <tr>
+                                                        <td class="px-6 py-4 font-medium text-gray-800 whitespace-nowrap">
+                                                            Text Summarizer
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            NLTK, Numpy, Pandas
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            29.09.23 - 06.10.23
+                                                        </td>
+                                                        <td class="no-invert px-6 py-4 text-gray-700 capitalize whitespace-nowrap">
+                                                            <span class="font-medium text-indigo-500">66%</span>
+                                                            <div class="w-full h-2 overflow-hidden bg-gray-200 rounded-full">
+                                                                <div class="w-8/12 h-full bg-indigo-500"></div>
+                                                            </div>
+                                                        </td>
+                                                    </tr>
+                                                    <tr>
+                                                        <td class="px-6 py-4 font-medium text-gray-800 whitespace-nowrap">
+                                                            Emotion Detection
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            Transformers, Torch
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            29.09.23 - 03.10.23
+                                                        </td>
+                                                        <td class="no-invert px-6 py-4 text-gray-700 capitalize whitespace-nowrap">
+                                                            <span class="font-medium text-indigo-500">25%</span>
+                                                            <div class="w-full h-2 overflow-hidden bg-gray-200 rounded-full">
+                                                                <div class="w-3/12 h-full bg-indigo-500 "></div>
+                                                            </div>
+                                                        </td>
+                                                    </tr>
+                                                    <tr>
+                                                        <td class="px-6 py-4 font-medium text-gray-800 whitespace-nowrap">
+                                                            Analytics
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            JS, Numpy, Pandas
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            29.09.23 - 03.10.23
+                                                        </td>
+                                                        <td class="no-invert px-6 py-4 text-gray-700 capitalize whitespace-nowrap">
+                                                            <span class="font-medium text-indigo-500">100%</span>
+                                                            <div class="w-full h-2 overflow-hidden bg-gray-200 rounded-full">
+                                                                <div class="w-full h-full bg-indigo-500"></div>
+                                                            </div>
+                                                        </td>
+                                                    </tr>
+                                                    <tr>
+                                                        <td class="px-6 py-4 font-medium text-gray-800 whitespace-nowrap">
+                                                            Grammar Checker
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            Transformers, Torch
+                                                        </td>
+                                                        <td class="px-6 py-4 font-medium text-gray-700 whitespace-nowrap">
+                                                            29.09.23 - 06.10.23
+                                                        </td>
+                                                        <td class="no-invert px-6 py-4 text-gray-700 capitalize whitespace-nowrap">
+                                                            <span class="font-medium text-indigo-500">75%</span>
+                                                            <div class="w-full h-2 overflow-hidden bg-gray-200 rounded-full">
+                                                                <div class="w-9/12 h-full bg-indigo-500 "></div>
+                                                            </div>
+                                                        </td>
+                                                    </tr>
+                                                </tbody>
+                                            </table>
+                                        </div>
+                                    </div>
+                                </div>
+                            </div>
+                        </section> -->
                     </div>
                 </div>
             </main>

static/text_summarizer.html CHANGED Viewed

@@ -12,15 +12,22 @@
 </head>
 <body class="overflow-hidden">
-    <!-- Google tag (gtag.js) -->
-    <script async src="https://www.googletagmanager.com/gtag/js?id=G-B751Q3XBFC"></script>
-    <script>
-        window.dataLayer = window.dataLayer || [];
-        function gtag(){dataLayer.push(arguments);}
-        gtag('js', new Date());
-        gtag('config', 'G-B751Q3XBFC');
     </script>
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

 </head>
 <body class="overflow-hidden">
+    <!-- Yandex.Metrika counter -->
+    <script type="text/javascript" >
+        (function(m,e,t,r,i,k,a){m[i]=m[i]||function(){(m[i].a=m[i].a||[]).push(arguments)};
+        m[i].l=1*new Date();
+        for (var j = 0; j < document.scripts.length; j++) {if (document.scripts[j].src === r) { return; }}
+        k=e.createElement(t),a=e.getElementsByTagName(t)[0],k.async=1,k.src=r,a.parentNode.insertBefore(k,a)})
+        (window, document, "script", "https://mc.yandex.ru/metrika/tag.js", "ym");
+        ym(95744490, "init", {
+            clickmap:true,
+            trackLinks:true,
+            accurateTrackBounce:true
+        });
     </script>
+    <noscript><div><img src="https://mc.yandex.ru/watch/95744490" style="position:absolute; left:-9999px;" alt="" /></div></noscript>
+    <!-- /Yandex.Metrika counter -->
     <div x-data="{ sidebarOpen: false }" class="relative flex h-screen text-gray-800 bg-white font-roboto">
         <div x-cloak :class="sidebarOpen ? 'block' : 'hidden'" @click="sidebarOpen = false"

youtube.py CHANGED Viewed

@@ -2,7 +2,7 @@ import re
 from youtube_transcript_api import YouTubeTranscriptApi
 from youtube_transcript_api._errors import TranscriptsDisabled
-MAX_SIZE = 20_000
 YT_REGEX = r'^((http)s?:\/\/)?((www\.)|(m\.))?youtube.com\/watch\?([^\?]*&)?v=.+$'  # noqa
 YT_REGEX_SHORT = r'^((http)s?:\/\/)?youtu.be\/([^\?=]+)(\?[^?]+)?$'
@@ -30,11 +30,11 @@ def get_youtube_caption(url: str) -> str:
         res, size = [], 0
         for transcript in YouTubeTranscriptApi.get_transcript(video_id):
-            res.append(transcript['text'])
             size += len(transcript['text'])
             if size >= MAX_SIZE:
-                return '\n'.join(res)[:MAX_SIZE]
-        return '\n'.join(res)[:MAX_SIZE]
     except TranscriptsDisabled:
         return 'no-cap'
     except Exception:

 from youtube_transcript_api import YouTubeTranscriptApi
 from youtube_transcript_api._errors import TranscriptsDisabled
+MAX_SIZE = 50_000
 YT_REGEX = r'^((http)s?:\/\/)?((www\.)|(m\.))?youtube.com\/watch\?([^\?]*&)?v=.+$'  # noqa
 YT_REGEX_SHORT = r'^((http)s?:\/\/)?youtu.be\/([^\?=]+)(\?[^?]+)?$'
         res, size = [], 0
         for transcript in YouTubeTranscriptApi.get_transcript(video_id):
+            res.append(transcript['text'].replace('\n', ' '))
             size += len(transcript['text'])
             if size >= MAX_SIZE:
+                return ' '.join(res)[:MAX_SIZE]
+        return ' '.join(res)[:MAX_SIZE]
     except TranscriptsDisabled:
         return 'no-cap'
     except Exception: