The latest articles on DEV Community by Sadikov Dev (@iravshan). https://dev.to/iravshan https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F839256%2F31b6d780-ff4d-4834-9021-412f4b19eca0.jpeg https://dev.to/iravshan en Sadikov Dev Tue, 21 Feb 2023 08:59:14 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-5-dars-58fh https://dev.to/iravshan/computer-vision-from-0-to-hero-5-dars-58fh <h2> Rasm tasnifi, ma'lumotlarga asoslangan yondashuv </h2> <p><strong>Surat tasnifi</strong></p> discuss Sadikov Dev Mon, 20 Feb 2023 14:31:47 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-4-dars-4op8 https://dev.to/iravshan/computer-vision-from-0-to-hero-4-dars-4op8 <h2> SciPy </h2> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fra59rz1qr5d5hxgxt5ly.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fra59rz1qr5d5hxgxt5ly.png" alt="Scipy" width="318" height="158"></a></p> <p>Numpy yuqori samarali ko'p o'lchovli massivlar va ushbu massivlar bilan hisoblash va ularni boshqarish uchun asosiy vositalarni taqdim etadi. SciPy numpy massivlariga asoslanadi va turli xil ilmiy va muhandislik ilovalari uchun foydali bo'lgan ko'plab funksiyalarni taqdim etadi.</p> <p>SciPy bilan tanishishning eng yaxshi usuli <a href="https://docs.scipy.org/doc/scipy/reference/index.html" rel="noopener noreferrer">hujjatlarni ko'rib chiqishdir</a>. Hozirda esa biz uchun foydali bo'lishi mumkin bo'lgan SciPy'ning ba'zi qismlarini qarab chiqamiz.</p> <p><strong>Rasm operatsiyalari</strong></p> <p>SciPy tasvirlar bilan ishlash uchun ba'zi asosiy funksiyalarni taqdim etadi. Masalan, tasvirlarni numpy massivlarga o'qish, numpy massivlarni tasvir sifatida yozish va tasvir hajmini o'zgartirish funksiyalariga ega. Mana bu funksiyalarni ko'rsatadigan oddiy misol:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>from scipy.misc import imread, imsave, imresize # JPEG suratni numpy massivga o'qib olish img = imread('assets/cat.jpg') print(img.dtype, img.shape) # Prints "uint8 (400, 248, 3)" # Rang kanallarining har birini boshqa skalyar konstanta bilan masshtablash orqali tasvirni ranglashimiz mumkin; # Tasvirning (400, 248, 3) shakli bor; # Uni (3,) shakldagi [1, 0,95, 0,9] massivga ko'paytiramiz; # numpy broadcasting - bu qizil kanalni o'zgarishsiz qoldirishini anglatadi va yashil va ko'k kanallarni mos ravishda 0,95 va 0,9 ga ko'paytiradi. img_tinted = img * [1, 0.95, 0.9] # Rangli tasvirni 300x300 px holatiga keltirish img_tinted = imresize(img_tinted, (300, 300)) # Rangli tasvirni xotiraga saqlash imsave('assets/cat_tinted.jpg', img_tinted) </code></pre> </div> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fxg8d17u6tnfrkt276m8v.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fxg8d17u6tnfrkt276m8v.png" alt="Test" width="248" height="400"></a></p> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fp88m4cynxpd16m7au8yt.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fp88m4cynxpd16m7au8yt.png" alt="test" width="300" height="300"></a></p> <p>Tepada: asl rasm. Pastda: ranglangan va o'lchami o'zgartirilgan rasm.</p> <h2> MATLAB fayllari </h2> <p><code>scipy.io.loadmat</code> va <code>scipy.io.savemat</code> funktsiyalari sizga MATLAB fayllarini o'qish va yozish imkonini beradi. Ular haqida batafsil <a href="https://docs.scipy.org/doc/scipy/reference/io.html" rel="noopener noreferrer">hujjatlarda o'qishingiz mumkin</a>.</p> <p><strong>Nuqtalar orasidagi masofa</strong></p> <p>SciPy nuqtalar to'plami orasidagi masofani hisoblash uchun ba'zi foydali funktsiyalarni taqdim etadi.</p> <p><code>scipy.spatial.distance.pdist</code> funksiyasi berilgan to'plamdagi barcha (x, y) nuqtalar orasidagi masofani hisoblaydi:</p> <p>Ushbu funksiya haqida barcha tafsilotlarni <a href="https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.distance.pdist.html" rel="noopener noreferrer">hujjatlarda</a> o'qishingiz mumkin .</p> <p>Shunga o'xshash funksiya (<code>scipy.spatial.distance.cdist</code>) ikkita nuqta to'plamidagi barcha juftliklar orasidagi masofani hisoblaydi, bu haqda <a href="https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.distance.cdist.html" rel="noopener noreferrer">hujjatlarda</a> o'qishingiz mumkin .</p> <h2> Matplotlib </h2> <p><strong>Matplotlib</strong> - bu chizmachilik kutubxonasi. Ushbu bo'limda <code>matplotlib.pyplot</code> MATLAB ga o'xshash chizma tizimini taqdim etadigan modul haqida qisqacha ma'lumot beriladi.</p> <p><strong>Syujet tuzish</strong></p> <p>Matplotlib'dagi eng muhim funksiya bu <code>plot2D</code> ma'lumotlarni chizish imkonini beradi. Mana oddiy misol:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>import numpy as np import matplotlib.pyplot as plt # Sinus egri chiziqdagi nuqtalar uchun x va y koordinatalarini hisoblash x = np.arange(0, 3 * np.pi, 0.1) y = np.sin(x) # Matplotlib yordamida nuqtalarni ifodalash plt.plot(x, y) plt.show() # Grafiklar paydo bo'lishi uchun plt.show() ga murojaat qilishingiz kerak. </code></pre> </div> <p>Ushbu kodni ishga tushirish quyidagi chizma hosil qiladi:</p> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fr47emcv9cvvx3ecpczmo.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fr47emcv9cvvx3ecpczmo.png" alt="Graph" width="381" height="256"></a></p> <p>Bir oz qo'shimcha ish bilan biz bir vaqtning o'zida bir nechta satrlarni osongina chizishimiz, sarlavha va koordinata o'qlarining nomlarini qo'shishimiz mumkin:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>import numpy as np import matplotlib.pyplot as plt # Sinus va kosinus egri chiziqlaridagi nuqtalar uchun x va y koordinatalarini hisoblash x = np.arange(0, 3 * np.pi, 0.1) y_sin = np.sin(x) y_cos = np.cos(x) # Matplotlib yordamida nuqtalarni chizish plt.plot(x, y_sin) plt.plot(x, y_cos) plt.xlabel('x axis label') plt.ylabel('y axis label') plt.title('Sine and Cosine') plt.legend(['Sine', 'Cosine']) plt.show() </code></pre> </div> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fzr4tvbih9vptwuah8fu9.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fzr4tvbih9vptwuah8fu9.png" alt="gr" width="396" height="281"></a></p> <p><strong>Tasvirlar</strong></p> <p>Rasmlarni ko'rsatish uchun <code>imshow</code> funksiyasidan foydalanishingiz mumkin . Mana bir misol:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>import numpy as np from scipy.misc import imread, imresize import matplotlib.pyplot as plt img = imread('assets/cat.jpg') img_tinted = img * [1, 0.95, 0.9] # Original suratni chiqarish plt.subplot(1, 2, 1) plt.imshow(img) # Ranglangan suratni chiqarish plt.subplot(1, 2, 2) # tasvirni ko'rsatishdan oldin uni uint8 ga konvert qilamiz plt.imshow(np.uint8(img_tinted)) plt.show() </code></pre> </div> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F1bzhb30ubfr2d2l3cekn.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F1bzhb30ubfr2d2l3cekn.png" alt="img1" width="357" height="256"></a></p> career backenddevelopment uiux community Sadikov Dev Mon, 20 Feb 2023 14:01:09 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-3-dars-1oed https://dev.to/iravshan/computer-vision-from-0-to-hero-3-dars-1oed <p><strong>Python versiyalari</strong></p> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--cQQ6NvIn--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/lltgnffpahfq59wik5je.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--cQQ6NvIn--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/lltgnffpahfq59wik5je.png" alt="Versions" width="780" height="350"></a></p> <p>2020-yil 1-yanvardan boshlab Pythonning <code>python2</code> versiyasi qoʻllab-quvvatlash rasman toʻxtatildi. Ushbu sinf uchun barcha kodlar Python 3.7 dan foydalanadi . Ushbu qoʻllanmani davom ettirishdan oldin oʻrnatish koʻrsatmalaridan oʻtganingizga va virtual muhitni toʻgʻri oʻrnatganingizga ishonch hosil qiling. <code>python3</code> ni ishga tushirish orqali muhitni faollashtirgandan so'ng, Python versiyangizni buyruq satrida tekshirishingiz mumkin <code>python --version</code>.</p> <p>Pythonning barcha versiyalari va ular haqida ko'proq ma'lumot olish uchun pythonning rasmiy saytida o'qishingiz mumkin:</p> <p><a href="https://www.python.org/doc/versions/">Bu yerda</a></p> uzbek ai machinelearning computerscience Sadikov Dev Mon, 20 Feb 2023 13:28:20 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-2-dars-4g19 https://dev.to/iravshan/computer-vision-from-0-to-hero-2-dars-4g19 <h2> Python </h2> <p>Python so'nggi yillarda dunyodagi eng mashhur dasturlash tillaridan biriga aylandi. U mashinani o'rganishdan tortib vebsaytlar yaratish va dasturiy ta'minotni sinovdan o'tkazishgacha qo'llaniladi. </p> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--cFSDU46x--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/0yl5zj2r999mkua0i4hb.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--cFSDU46x--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/0yl5zj2r999mkua0i4hb.png" alt="Coder" width="800" height="450"></a></p> <p>Python - veb ilovalar, mobil ilovalar, sun'iy intellekt modellari, katta ma'lumotlar tahlili va dasturiy ta'minotlar yaratish uchun keng qo'llanilmoqda. </p> <p><strong>Python nima?</strong></p> <p>Python - bu dasturlash tili bo'lib, ko'pincha vebsaytlar va dasturlarni yaratish, vazifalarni avtomatlashtirish va ma'lumotlar tahlilini o'tkazish uchun ishlatiladi. Python - bu umumiy maqsadli til, ya'ni, u turli xil dasturlarni yaratish uchun ishlatilishi mumkin va hech qanday maxsus muammolar uchun ixtisoslanmagan. Ushbu ko'p qirralilik, yangi boshlovchilar uchun qulayligi bilan birga, uni bugungi kunda eng ko'p ishlatiladigan dasturlash tillaridan biriga aylantirdi. RedMonk sanoat tahliliy firmasi tomonidan o'tkazilgan so'rov shuni ko'rsatdiki, u 2021-yilda ishlab chiquvchilar orasida ikkinchi eng mashhur dasturlash tili deb topildi.</p> <p><strong>Python nima uchun ishlatiladi?</strong></p> <p>Python odatda vebsaytlar va dasturiy ta'minotni ishlab chiqish, vazifalarni avtomatlashtirish, ma'lumotlarni tahlil qilish va ma'lumotlarni vizuallashtirish uchun ishlatiladi. O'rganish nisbatan oson bo'lgani uchun Python ko'plab dasturchi bo'lmaganlar, masalan, buxgalterlar va olimlar tomonidan moliyani tashkil qilish kabi kundalik vazifalar uchun qo'llaniladi.</p> <p><strong>Python bilan nima qila olasiz?</strong></p> <ul> <li><p>Ma'lumotlarni tahlil qilish va mashinani o'rganish</p></li> <li><p>Veb ishlab chiqish</p></li> <li><p>Avtomatlashtirish yoki skript yaratish</p></li> <li><p>Dasturiy ta'minotni sinovdan o'tkazish va prototiplash</p></li> <li><p>Kundalik vazifalar</p></li> </ul> <p><strong>Ma'lumotlarni tahlil qilish va mashinani o'rganish</strong></p> <p>Python ma'lumotlar tahlilida kuchli "qurolga" aylandi, bu ma'lumotlar tahlilchilari va boshqa mutaxassislarga murakkab statistik hisob-kitoblarni amalga oshirish, ma'lumotlar vizualizatsiyasini yaratish, mashinani o'rganish algoritmlarini yaratish, ma'lumotlarni manipulyatsiya qilish va tahlil qilish va boshqa ma'lumotlar bilan bog'liq vazifalarni bajarishga imkon yaratdi.</p> <p>Python turli xil ma'lumotlar vizualizatsiyasini yaratishi mumkin, masalan, chiziqli grafiklar, doiraviy diagrammalar, gistogrammalar va 3D chizmalar. Python shuningdek, TensorFlow va Keras kabi koderlarga ma'lumotlarni tahlil qilish va mashinani o'rganish uchun dasturlarni tezroq va samaraliroq yozish imkonini beruvchi bir qator kutubxonalarga ega.</p> <p><strong>Veb ishlab chiqish</strong></p> <p>Python ko'pincha veb-sayt yoki ilovaning orqa tomoni (backend) ni - foydalanuvchi ko'rmaydigan qismlarni ishlab chiqish uchun ishlatiladi. Pythonning veb-ishlab chiqishdagi roli serverlarga ma'lumotlarni yuborish, ma'lumotlarni qayta ishlash va ma'lumotlar bazalari bilan aloqa qilish, URL marshrutlash va xavfsizlikni ta'minlashni o'z ichiga olishi mumkin. Python veb ishlab chiqish uchun bir nechta ramka (framework) larni taklif qiladi. Ko'p ishlatiladiganlarga Django va Flask kiradi.</p> <p>Pythondan foydalanadigan ba'zi veb-ishlab chiqarish ishlariga backend muhandislari, fullstack muhandislari, Python dasturchilari, dasturiy ta'minot muhandislari va DevOps muhandislari kiradi.</p> <p><strong>Avtomatlashtirish yoki skript yaratish</strong></p> <p>Agar biror vazifani qayta-qayta bajarayotganingizni sezsangiz, uni Python bilan avtomatlashtirish orqali samaraliroq ishlashingiz mumkin. Ushbu avtomatlashtirilgan jarayonlarni yaratish uchun ishlatiladigan kod yozish skript deb ataladi. Kodlash dunyosida avtomatlashtirish bir nechta fayllardagi xatolarni tekshirish, fayllarni aylantirish, oddiy matematikani bajarish va ma'lumotlardagi dublikatlarni olib tashlash uchun ishlatilishi mumkin.</p> <p>Python hatto yangi boshlovchilar tomonidan ham kompyuterdagi oddiy vazifalarni avtomatlashtirish uchun ishlatilishi mumkin, masalan, fayllar nomini o'zgartirish, onlayn kontentni topish va yuklab olish yoki elektron pochta yoki matnlarni kerakli vaqt oralig'ida yuborish.</p> <p><strong>Dasturiy ta'minotni sinovdan o'tkazish va prototiplash</strong></p> <p>Dasturiy ta'minotni ishlab chiqishda Python qurilishni boshqarish, xatolarni kuzatish va sinovdan o'tkazish kabi vazifalarni bajarishda yordam berishi mumkin. Python yordamida dasturiy ta'minot ishlab chiquvchilar yangi mahsulotlar yoki xususiyatlar uchun testlarni avtomatlashtirishlari mumkin. Dasturiy ta'minotni sinab ko'rish uchun ishlatiladigan ba'zi Python vositalariga Green va Requesium kiradi.</p> <p><strong>Kundalik vazifalar</strong></p> <p>Python nafaqat dasturchilar va ma'lumotlar tahlilchilari uchun, balki, uni o'rganish jurnalistlar, kichik biznes egalari yoki ijtimoiy media sotuvchilari kabi ma'lumot talab qiladigan kasblar uchun yangi imkoniyatlar ochishi mumkin. Python, shuningdek, dasturchi bo'lmaganlarga hayotlaridagi muayyan vazifalarni soddalashtirishga imkon beradi. Python bilan avtomatlashtirishingiz mumkin bo'lgan bir nechta vazifalar:</p> <ul> <li><p>Birja yoki kripto narxlarini kuzatib borish</p></li> <li><p>Yomg'ir yog'ayotganda soyabonni olib yurish uchun o'zingizga matnli eslatma yuborish</p></li> <li><p>Xarid qilish ro'yxatini yangilash</p></li> <li><p>Katta hajmdagi fayllar nomini o'zgartirish</p></li> <li><p>Matnli fayllarni elektron jadvallarga aylantirish</p></li> </ul> <p><strong>Nima uchun Python shunchalik mashhur?</strong></p> <p>Python bir necha sabablarga ko'ra mashhur. Kodlovchilar uchun uni juda ko'p qirrali va qulay qilishini chuqurroq ko'rib chiqamiz.</p> <ul> <li><p>U tabiiy tilga taqlid qiluvchi oddiy sintaksisga ega , shuning uchun uni o'qish va tushunish osonroq. Bu loyihalarni tezroq qurish va ularni yaxshilashni tezlashtiradi.</p></li> <li><p>Python veb ishlab chiqishdan tortib, mashinani o'rganishgacha bo'lgan turli xil vazifalar uchun ishlatilishi mumkin.</p></li> <li><p>Bu yangi boshlovchilar uchun qulay bo'lib, u bilan dasturlashni o'rganish juda oson</p></li> <li><p>Bu ochiq manba, ya'ni undan foydalanish va tarqatish, hatto tijorat maqsadlarida ham bepul.</p></li> <li><p>Pythonning modullar va kutubxonalar arxivi - uchinchi tomon foydalanuvchilari Python imkoniyatlarini kengaytirish uchun yaratgan kodlar to'plami - keng va o'sib bormoqda.</p></li> <li><p>Python katta va faol hamjamiyatga ega bo'lib, u Python modullari va kutubxonalari to'plamiga hissa qo'shadi va boshqa dasturchilar uchun foydali manba bo'lib xizmat qiladi. Keng qo'llab-quvvatlash hamjamiyati shuni anglatadiki, agar kodlovchilar qoqilish to'sig'iga duch kelsa, yechim topish nisbatan oson bo'ladi.</p></li> </ul> Sadikov Dev Mon, 20 Feb 2023 12:02:17 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-1-dars-2gd5 https://dev.to/iravshan/computer-vision-from-0-to-hero-1-dars-2gd5 <h2> 1-dars: Jupyter va Google Colab noutbuklari </h2> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--cWTpbko9--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/rglpfd6y6g6v7g22krmz.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--cWTpbko9--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/rglpfd6y6g6v7g22krmz.png" alt="Jupyter Notebook" width="800" height="450"></a></p> <p>Pythonga kirishdan oldin, biz notebooklar haqida qisqacha so'z yuritamiz. <strong>Jupyter noutbuki</strong> veb brauzeringizda localhostda Python kodini yozish va bajarish imkonini beradi. Jupyter noutbuklari kod bilan ishlashni va uni qismlarga bo'lib bajarishni juda osonlashtiradi. Shuning uchun ular ilmiy hisoblashda keng qo'llaniladi. </p> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--pKQF7AYc--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/gwe8tluirfdwev6ctzm2.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--pKQF7AYc--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/gwe8tluirfdwev6ctzm2.png" alt="Google Colab" width="800" height="303"></a></p> <p><strong>Colab</strong> - bu Google kompaniyasining Jupyter noutbuklari bo'lib, u mashinani o'rganish (machine learning) va ma'lumotlarni tahlil qilish (data analysis) uchun juda mos keladi va butunlay bulutda ishlaydi. Colab asosan hech qanday sozlashlarni talab qilmaydi, ko'plab kerakli paketlar bilan birga o'rnatiladi va undagi yozuvlarni boshqalar bilan baham ko'rish ham juda qulay.</p> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--Yu_1KH8V--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/abvk3fm319fuk2cfzg4p.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--Yu_1KH8V--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_800/https://dev-to-uploads.s3.amazonaws.com/uploads/articles/abvk3fm319fuk2cfzg4p.png" alt="Colab Example" width="640" height="400"></a></p> <p>Yakuniy xulosa sifatida shuni aytib o'tishimiz mumkinki, ushbu notebook'lar bir vaqtning o'zida pythonda yozilgan kodlarni bajarib berish, shu bilan birgalikda izohlar va matnlar bilan boyitish imkonini beradi</p> uzbek ai machinelearning computerscience Sadikov Dev Mon, 20 Feb 2023 11:31:31 +0000 https://dev.to/iravshan/computer-vision-from-0-to-hero-intro-170m https://dev.to/iravshan/computer-vision-from-0-to-hero-intro-170m <p>Computer Vision haqida o'zbekcha ma'lumotlar haddan tashqari kamligi sababli Stanford Universiteti talabalariga 2015, 2016, 2017-yillar mobaynida o'rgatilib, o'qitilib kelingan ingliz tilidagi ma'lumotlarni o'zbek tiliga tarjima qilishga qaror qildik.</p> <p>** Haq Taoloning izni va marhamati ila ushbu seriyamizni boshladik**</p> <h2> 1-dars. Computer Visionga kirish </h2> <p><strong>Python Numpy qo'llanmasi (Jupyter va Colab bilan)</strong></p> <p>Biz ushbu kursdagi barcha topshiriqlar uchun Python dasturlash tilidan foydalanamiz. Python ajoyib dasturlash tilidir va u bir nechta mashhur kutubxonalar (numpy, scipy, matplotlib) yordamida u ilmiy hisoblash uchun kuchli muhitga aylanadi.</p> <p>Ko'pchiligingiz Python va numpy bilan tajribaga ega bo'lishingiz hech gap emas, qolganlar uchun esa ushbu bo'lim Python dasturlash tili va undan ilmiy hisoblash uchun foydalanish bo'yicha tezkor kurs bo'lib xizmat qiladi. Shuningdek, biz Python kodi bilan ishlashning juda qulay usuli bo'lgan daftarlarni (Jupyter Notebooklar) ham taqdim etamiz. </p> <p><strong>Mundarija</strong></p> <ul> <li>Jupyter va Colab notebooklari</li> <li>Python <ul> <li>Python versiyalari</li> <li>Ma'lumotlarning asosiy turlari</li> <li>Konteynerlar <ul> <li>Ro'yxatlar</li> <li>Lug'atlar</li> <li>Setlar</li> <li>Tupllar</li> </ul> </li> <li>Funksiyalar</li> <li>Sinflar (Class'lar)</li> </ul> </li> <li>Numpy <ul> <li>Massivlar</li> <li>Massiv elementlarini indekslash</li> <li>Ma'lumotlar turlari</li> <li>Matematik masalalar uchun massivlar</li> <li>Massivlar ustida amallar (broadcasting)</li> <li>Numpy haqida (Numpy documentation)</li> </ul> </li> <li>SciPy <ul> <li>Rasm operatsiyalari</li> <li>MATLAB fayllari</li> <li>Nuqtalar orasidagi masofa</li> </ul> </li> <li>Matplotlib <ul> <li>Syujet tuzish</li> <li>Subplots</li> <li>Tasvirlar</li> </ul> </li> </ul> <p>Computer Vision ga bag'ishlangan ushbu kichik o'ziga xos kursimiz yuqoridagi mundarija asosida davom etadi. Kursimiz davomida foydalanadigan barcha turdagi ma'lumotlarimiz uchun Stanford University ilmiy xodimlari va talabalariga o'z minnatdorchiligimizni bildirib o'tamiz.</p> computerscience ai machinelearning uzbek