YouTube2Text

Hugging Face Tutorial for Beginners — Transcript

by proflead · 1,909 words · 267 segments · language en · Watch on YouTube

Full transcript

  1. 0:00if you're interested about artificial
  2. 0:02intelligence then I recommend you watch
  3. 0:05this video because in today's video
  4. 0:08we're going to talk about tool that
  5. 0:10could help you run thousand different
  6. 0:13models on your local computer hey guys
  7. 0:16my name is lat and welcome back to my
  8. 0:18channel in this video we're going to
  9. 0:19talk about hugging
  10. 0:21face this will be mini tutorial that
  11. 0:25give you comprehensive understanding
  12. 0:28what is this hame face and what you can
  13. 0:30do with it if you're a person who
  14. 0:33fascinated about artificial intelligence
  15. 0:36I recommend you watch this video because
  16. 0:38we will be able to run different AI
  17. 0:41models on your local computer if you're
  18. 0:44ready then let's get started first thing
  19. 0:48first what exactly is hugging face
  20. 0:51simply put uh it is an open- Source Air
  21. 0:54Platform that specialize in natural
  22. 0:56language processing through Transformers
  23. 0:59models whether you want to analyze text
  24. 1:02uh generate creative content or even
  25. 1:05work with images and audio hugin phas
  26. 1:08have something in store for you to get
  27. 1:10started with it open huggingface website
  28. 1:14and create an account there then you
  29. 1:16will be redirected to your dashboard and
  30. 1:19then from here we can start explore the
  31. 1:22website you should be aware of three
  32. 1:24things models data sets and spaces the
  33. 1:27huging face model Hub is a repos itory
  34. 1:30where you can find pre-training models
  35. 1:32for a wide range of tasks such as
  36. 1:35natural language processing computer
  37. 1:37Vision Audio processing and more these
  38. 1:40models are contributed by the community
  39. 1:42and hen face itself the data sets
  40. 1:45library at hugen phas designed to
  41. 1:47provide a simple and efficient way to
  42. 1:50access a wide array of data sets they
  43. 1:53could be useful for machine learning in
  44. 1:55data driven projects and spaces on hgen
  45. 1:58phas providing easy to use platform for
  46. 2:01users to deploy their machine learning
  47. 2:03models and showcase interactive AI
  48. 2:06applications so if you want to showcase
  49. 2:09your AI model hugging face is the best
  50. 2:12place to do that as you could see here a
  51. 2:15lot of different models from small and
  52. 2:18big companies and also from private
  53. 2:21people let me show you some example I'm
  54. 2:24going to select a random model let's try
  55. 2:27image generation first I will select the
  56. 2:30one that has the most likes okay it's
  57. 2:33stable diffuser let's try the demo all
  58. 2:37you need just write a prompt and then
  59. 2:40press the generate image button and here
  60. 2:43we go this is my result as you can see
  61. 2:46it's really easy to use this web
  62. 2:48interface to try these different models
  63. 2:51but we also can do this in code and I'm
  64. 2:53going to show it later okay let's go
  65. 2:56back to spaces and try some other model
  66. 2:59all I'm going to try background removal
  67. 3:02let's see the interface of this AI
  68. 3:05application I'm going to find some image
  69. 3:08let's use this
  70. 3:10image and let's try to run the
  71. 3:13application all you need just press the
  72. 3:16submit button it could take some time
  73. 3:19because application could wait when GPU
  74. 3:22to become available but usually the
  75. 3:24waiting time is not that long and as you
  76. 3:27could see my picture is ready and here
  77. 3:30we go this is my image without the
  78. 3:33background and seems it works perfectly
  79. 3:37that's probably it about spaces let's go
  80. 3:39to the coding part Hing face
  81. 3:43Transformers Transformers is a core
  82. 3:46library that supports a broad array of
  83. 3:48Transformer models Transformers are a
  84. 3:52type of deep learning model architecture
  85. 3:54that excels and understanding the
  86. 3:57context and nuances of language
  87. 4:00the library provides a lot of pre
  88. 4:03trining models and F tuning tools that
  89. 4:06you can use for different tasks such as
  90. 4:08text classification tokenization
  91. 4:11translation summarization and so on I'm
  92. 4:14going to show you how to use it on Linux
  93. 4:17but it also available for Windows and
  94. 4:19Mac if you using terminal as me then the
  95. 4:23steps will be almost the same before we
  96. 4:25dive in into the specific applications
  97. 4:28ensure that your your development
  98. 4:30environment set up correctly you will
  99. 4:32need python Transformer library and few
  100. 4:35other libraries simply copy this command
  101. 4:38and paste it inside your terminal if you
  102. 4:41see this error it means you are trying
  103. 4:44to run the installation package globally
  104. 4:46but we need to follow the recommendation
  105. 4:49so we have to create a virtual
  106. 4:51environment first simply use this
  107. 4:54command and then activate your virtual
  108. 4:56environment and after that you can run
  109. 4:58the previous command and and everything
  110. 5:01should be okay after we install the
  111. 5:04Transformer and other libraries we need
  112. 5:06to install machine learning framework by
  113. 5:09Toro T are Flor these are two the most
  114. 5:13popular open source Frameworks by torch
  115. 5:16was developed by Facebook and tender
  116. 5:18flow was developed by Google for this
  117. 5:21tutorial I'm going to use torch the next
  118. 5:24step is optional if you are using Nvidia
  119. 5:26graphic card you can follow this step or
  120. 5:29you can skip it in this step we are
  121. 5:31going to install Cuda it is a tool for
  122. 5:34par Computing it allows developers to
  123. 5:37use Nvidia gpus basically this tool
  124. 5:41helps you accelerate of computation
  125. 5:43tasks because it's much faster than CPU
  126. 5:47to install Coda toolkit you need to open
  127. 5:49Nvidia website then go to the Cuda
  128. 5:52download select your operation system
  129. 5:55and then follow the instruction for
  130. 5:57Linux I have to run these eight commands
  131. 6:00in my terminal window for Windows and
  132. 6:02Mac users steps uh could be different so
  133. 6:06just follow the instruction okay and now
  134. 6:09we are ready to explore Transformer in
  135. 6:11action let's go back to huging faed
  136. 6:14website and open the models link there
  137. 6:17are so many different models available
  138. 6:20some of the models will require the
  139. 6:22stronger computer power so keep in mind
  140. 6:25that maybe you can trun them but if you
  141. 6:28have powerful computer
  142. 6:30then it won't be a problem on the model
  143. 6:32page you can apply the filter and select
  144. 6:34different models from the list there
  145. 6:37models that helps you work with the
  146. 6:39computer vision images voices and so on
  147. 6:43if you want to know a bit more details
  148. 6:45about specific model simply click on it
  149. 6:48and then you will see the details
  150. 6:50sometimes you would need to accept the
  151. 6:52terms and condition before you are going
  152. 6:54to use the model and some model don't
  153. 6:57have it okay let's try one model in next
  154. 7:00I'm going to use text classification
  155. 7:02model let's select the most popular one
  156. 7:05okay this is BG ranker let's try it
  157. 7:09first simply click at the model and then
  158. 7:12follow the instructions inside as you
  159. 7:15could see in this model there's no terms
  160. 7:17and conditions so I have to just copy
  161. 7:19and paste the code and run it on my
  162. 7:22computer keep in mind that the
  163. 7:24instruction could be for different
  164. 7:25libraries as well so you don't have to
  165. 7:27follow everything find the one that
  166. 7:30related to Transformer and follow the
  167. 7:32instruction for the Transformer all you
  168. 7:35would need uh just copy the code into
  169. 7:38your ID and then run it this particular
  170. 7:41example of text classification I showed
  171. 7:44the relevant score for the answer on the
  172. 7:47specific question and as you could see
  173. 7:49in the example the first question is
  174. 7:51what is Panda and the answer is high so
  175. 7:54the score is negative and it means the
  176. 7:58answer is a relevant however for the
  177. 8:01second question we get high score and
  178. 8:04seems the answer is relevant this is
  179. 8:06really simple example but it showcase
  180. 8:09how it's easy to use huging face let me
  181. 8:12demonstrate for you one more example
  182. 8:14this time let's try Tex generation model
  183. 8:17as you could see in the list we have DC
  184. 8:18car1 Mistral L Lama and many other
  185. 8:22models I'm going to try P4 mini from
  186. 8:25Microsoft because it's lightwe and
  187. 8:27doesn't require a lot of memory Okay
  188. 8:30click at the model then you can read the
  189. 8:32instruction or you can click this black
  190. 8:34button use this model and then copy and
  191. 8:38paste the code from here keep in mind
  192. 8:40that sometimes under the button the code
  193. 8:44is not full you would need to modify it
  194. 8:46a bit I would recommend you to use the
  195. 8:48instruction instead but if you have
  196. 8:50experience you definitely can use this
  197. 8:53method okay I paste the code and let's
  198. 8:55try to run it if you will encounter this
  199. 8:58error like me have on the the screen now
  200. 9:00it is a probably due to the fact that
  201. 9:02after you install cro you haven't
  202. 9:04restart your computer simply restart
  203. 9:06your computer and the error should gone
  204. 9:09okay I restarted my computer let's go
  205. 9:11back to the huging face website I'm
  206. 9:13going to copy the code from the
  207. 9:15instruction I will slightly modify the
  208. 9:17code I will remove first three questions
  209. 9:20and leave only the last one okay let's
  210. 9:22try to run the code and see the result
  211. 9:24and as you could see the model can solve
  212. 9:28this simple math problem and it gives me
  213. 9:30correct answer okay let me show you one
  214. 9:33more example probably the final one this
  215. 9:35time we will try to work with images I'm
  216. 9:38going to select image to text model we
  217. 9:40are going to try a blip from Salesforce
  218. 9:42this model will help us to understand
  219. 9:44what is on the picture or basically you
  220. 9:47can use it to create captions for the
  221. 9:49pictures for example I'm going to copy
  222. 9:51the example so simply copy this code and
  223. 9:54paste it inside your ID first I'm going
  224. 9:57to use default code and then I will will
  225. 9:59adjust it to use my picture in this
  226. 10:02example we see dog beach and the lady
  227. 10:05let's try to run the code and see what
  228. 10:07caption we get and here we go this is
  229. 10:09the result and as you could see the
  230. 10:12result is correct let's try to modify
  231. 10:14code a bit this time I'm going to use my
  232. 10:16picture and let's see if the model can
  233. 10:19understand it I'm going to use this
  234. 10:20obsidian sync image let me quickly fix
  235. 10:23the code and run it and here we go this
  236. 10:26is the result and as you could see the
  237. 10:29capture is correct and I think this is
  238. 10:31the last example I don't think you need
  239. 10:34any extra details please give it a try
  240. 10:36try different models and see what you
  241. 10:38can get in terms of pricing hgen face
  242. 10:42actually is free but you can get Pro
  243. 10:44count that gives you additional features
  244. 10:47I would recommend you to try free first
  245. 10:49and then decide it if you need the paid
  246. 10:52one and that's probably it if you will
  247. 10:54like this video and you will like
  248. 10:56hugging face I'm going to create a video
  249. 10:58about how you can train your llm and
  250. 11:01upload it into huging face if you're
  251. 11:04interested please let me know in
  252. 11:06comments below and that's it as you can
  253. 11:08see it's really easy to use and it's
  254. 11:11really powerful if you like this video
  255. 11:13please click like if you don't like this
  256. 11:15video it's okay please click like but
  257. 11:19share with me your feedback in comments
  258. 11:20below watch my other videos because they
  259. 11:23are really useful and if you haven't
  260. 11:25subscribed to my channel yet please
  261. 11:28subscribe and don't forget to click Bell
  262. 11:32button to get notification about my new
  263. 11:35videos because they will be really
  264. 11:38useful okay thank you so much for
  265. 11:40watching this video till the end thank
  266. 11:42you so much for your support and I hope
  267. 11:45to see you in my next video bye

About this transcript

This page contains the full transcript of Hugging Face Tutorial for Beginners by proflead, generated from the public captions YouTube serves with the video. The transcript has 1,909 words across 267 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.