[B! python][æ©Ÿæ¢°å¦ç¿’] slay-tã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

slay-t id:slay-t

pythonã¨æ©Ÿæ¢°å¦ç¿’ã«é–¢ã™ã‚‹slay-tã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (64)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

Transformerã‚’ç”¨ã„ã¦ã‚ªãƒŽãƒžãƒˆãƒšã‹ã‚‰ç’°å¢ƒéŸ³ã‚’åˆæˆã™ã‚‹æ‰‹æ³•ã‚’PyTorchã§å®Ÿè£…ã—ãŸï¼ˆTransformerç‰ˆ Onoma-to-Waveï¼‰ - å‚™å¿˜éŒ²
ã¯ã˜ã‚ã« äº‹å‰æº–å‚™ å®Ÿè£…ã®æ¦‚è¦ ç’°å¢ƒéŸ³åˆæˆå®Ÿé¨“ å®Ÿé¨“æ¡ä»¶ å®Ÿé¨“çµæžœ å®Ÿè£…ã®èˆžå°è£ãªã© ãŠã‚ã‚Šã« ã¯ã˜ã‚ã« ä»¥å‰ã€Onoma-to-Waveã‚’å®Ÿè£…ã—ãŸè¨˜äº‹ã‚’æ›¸ã„ãŸã“ã¨ãŒã‚ã£ãŸï¼š tam5917.hatena blog.com Onoma-to-Waveã¨ã¯ã‚ªãƒŽãƒžãƒˆãƒšï¼ˆæ–‡å—åˆ—ï¼‰ã‚’ç’°å¢ƒéŸ³ï¼ˆã‚¹ãƒšã‚¯ãƒˆãƒã‚°ãƒ©ãƒ ï¼‰ã«å¤‰æ›ã™ã‚‹ãƒ¢ãƒ‡ãƒ«ã§ã‚ã‚‹ã€‚RNNã«åŸºã¥ãencoderã¨decoderã‹ã‚‰æ§‹æˆã•ã‚Œã¦ãŠã‚Šã€ã„ã‚ã‚†ã‚‹Sequence-to-Sequenceï¼ˆSeq2Seqï¼‰ã®æ§‹é€ ã‚’æŒã¤ã€‚ ãã‚Œã‚‰ã‚’Transf ormerã«ã‚ˆã£ã¦ç½®ãæ›ãˆãŸãƒ¢ãƒ‡ãƒ«ãŒã€Onoma-to-Waveã®è‘—è€…ã‚‰ã«ã‚ˆã£ã¦å®Ÿã¯ã™ã§ã«ææ¡ˆã•ã‚Œã¦ã„ã‚‹ã€‚ å²¡æœ¬ æ‚ å¸Œï¼Œäº•æœ¬ æ¡‚å³ï¼Œé«˜é“ æ…Žä¹‹ä»‹ï¼Œç¦æ£® éš†å¯›ï¼Œå±±ä¸‹ æ´‹ä¸€ï¼Œ"Transf ormerã‚’ç”¨ã„ãŸã‚ªãƒŽãƒžãƒˆãƒšã‹ã‚‰ã®ç’°å¢ƒéŸ³åˆæˆï¼Œ" æ—¥æœ¬éŸ³éŸ¿å¦ä¼š2021å¹´ç§‹å£ç ”ç©¶ç™ºè¡¨ä¼šï¼Œpp. 943-946ï¼Œ2021. ä¸Š
slay-t 2022/10/11
ç’°å¢ƒ

è¨å®š

æ©Ÿæ¢°å¦ç¿’

python
ãƒªãƒ³ã‚¯
Python ã§å¦ã¶ãƒ™ã‚¤ã‚ºãƒ•ã‚£ãƒ«ã‚¿ã¨ã‚«ãƒ«ãƒžãƒ³ãƒ•ã‚£ãƒ«ã‚¿ (ç¿»è¨³) - inzkyk.xyz
ç¿»è¨³ã«ã¤ã„ã¦ ã“ã‚Œã¯ Roger R. Labbe è‘— Kalman and Bayesian Filters in Python ã®ç¿»è¨³ã§ã™ã€‚è‹±èªžç‰ˆã¯ CC BY 4.0 ãƒ©ã‚¤ã‚»ãƒ³ã‚¹ã§å…¬é–‹ã•ã‚Œã¦ã„ã¾ã™ã€‚ ã“ã®ç¿»è¨³ã¯ CC BY 4.0 ãƒ©ã‚¤ã‚»ãƒ³ã‚¹ã®è¨±è«¾ã«åŸºã¥ã„ã¦å…¬é–‹ã•ã‚Œã¾ã™ã€‚ PDF ç‰ˆã¨ Jupyter Notebook ç‰ˆã«ã¤ã„ã¦ ã“ã®ç¿»è¨³ã® PDF ç‰ˆã¨ Jupyter Notebook ç‰ˆã‚’ BOOTH ã§è²©å£²ã—ã¦ã„ã¾ã™ã€‚ è¬è¾ž è‹±èªžç‰ˆã®è‘—è€… Roger R. Labbe æ°ã«æ„Ÿè¬ã—ã¾ã™ã€‚ èª¤æ¤ã‚’æŒ‡æ‘˜ã—ã¦é ‚ã„ãŸå°å±±æµ©ä¹‹æ° (https://twitter.com/0yama) ã«æ„Ÿè¬ã—ã¾ã™ã€‚
slay-t 2022/07/20
python

ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ

æ©Ÿæ¢°å¦ç¿’

AI
ãƒªãƒ³ã‚¯
PyTorch ã®åŸºç¤Ž - Training
ã“ã®ãƒ–ãƒ©ã‚¦ã‚¶ãƒ¼ã¯ã‚µãƒãƒ¼ãƒˆã•ã‚Œãªããªã‚Šã¾ã—ãŸã€‚ Microsoft Edge ã«ã‚¢ãƒƒãƒ—ã‚°ãƒ¬ãƒ¼ãƒ‰ã™ã‚‹ã¨ã€æœ€æ–°ã®æ©Ÿèƒ½ã€ã‚»ã‚ãƒ¥ãƒªãƒ†ã‚£æ›´æ–°ãƒ—ãƒã‚°ãƒ©ãƒ ã€ãŠã‚ˆã³ãƒ†ã‚¯ãƒ‹ã‚«ãƒ« ã‚µãƒãƒ¼ãƒˆã‚’åˆ©ç”¨ã§ãã¾ã™ã€‚
slay-t 2022/01/11
computer

language

ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿

å¦ç¿’

æ©Ÿæ¢°å¦ç¿’

python

å‹‰å¼·

è‡ªç„¶è¨€èªžå‡¦ç†

machinelearning
ãƒªãƒ³ã‚¯
æ±äº¬å·¥æ¥å¤§å¦ã€æ©Ÿæ¢°å¦ç¿’ã®è¬›ç¾©ãƒŽãƒ¼ãƒˆãŒç„¡æ–™å…¬é–‹ Pythonã®å®Ÿè£…ã‚‚å¦ã¹ã‚‹ | Ledge.ai
ã‚µã‚¤ãƒ³ã‚¤ãƒ³ã—ãŸçŠ¶æ…‹ã§ã€Œã„ã„ãã€ã‚’æŠ¼ã™ã¨ã€ãƒžã‚¤ãƒšãƒ¼ã‚¸ã® ã€Œã„ã„ãå±¥æ´ã€ã«ä¸€è¦§ã¨ã—ã¦ä¿å˜ã•ã‚Œã¦ã„ãã®ã§ã€ å†åº¦èªã¿ãŸããªã£ãŸæ™‚ã‚„ã€ã‚ã¨ã§ã˜ã£ãã‚Šèªã¿ãŸã„ã¨ãã«ä¾¿åˆ©ã§ã™ã€‚
slay-t 2021/12/10
å¦ç¿’

python

ãƒ—ãƒã‚°ãƒ©ãƒ

AI

æ±äº¬

å‹‰å¼·

æ©Ÿæ¢°å¦ç¿’

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

è³‡æ–™
ãƒªãƒ³ã‚¯
ã¯ã˜ã‚ã« â€” æ©Ÿæ¢°å¦ç¿’å¸³
import torch x = torch.tensor([1., -1.]) w = torch.tensor([1.0, 0.5], requires_grad=True) loss = -torch.dot(x, w).sigmoid().log() loss.backward() print(loss.it em()) print(w.grad)
slay-t 2021/11/29
å¦ç¿’

log

4

å‹‰å¼·

æ©Ÿæ¢°å¦ç¿’

python

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

machinelearning

æ•°å¦
ãƒªãƒ³ã‚¯
https://e2eml.school/transformers.html
slay-t 2021/11/26
deeplearning

machinelearning

æ©Ÿæ¢°å¦ç¿’

python

è‡ªç„¶è¨€èªžå‡¦ç†

nlp

math

æ•°å¦
ãƒªãƒ³ã‚¯
å®Ÿè·µ æ™‚ç³»åˆ—è§£æž
æ™‚ç³»åˆ—ãƒ‡ãƒ¼ã‚¿ãŒä½¿ã‚ã‚Œã‚‹ç¯„å›²ã¯åºƒãã€åŒ»ç™‚ãƒ‡ãƒ¼ã‚¿ã€é‡‘èžåˆ†æžã€çµŒæ¸ˆäºˆæ¸¬ã€å¤©æ°—äºˆå ±ãªã©ã€ã•ã¾ã–ã¾ãªåˆ†é‡Žã§ä½¿ã‚ã‚Œã¦ã„ã¾ã™ã€‚æœ¬æ›¸ã¯æ™‚ç³»åˆ—ãƒ‡ãƒ¼ã‚¿ã‚’é€šã—ã¦ãƒ‡ãƒ¼ã‚¿è§£æžæ‰‹æ³•ã‚’å¦ã‚“ã§ã‚†ãã‚¢ãƒ—ãƒãƒ¼ãƒã§ã€ãƒ‡ãƒ¼ã‚¿ã®ã‚¯ãƒªãƒ¼ãƒ‹ãƒ³ã‚°ã€ãƒ—ãƒãƒƒãƒˆã®æ–¹æ³•ã€å…¥å‡ºåŠ›ãªã©åŸºæœ¬çš„ãªãƒˆãƒ”ãƒƒã‚¯ã«ã¤ã„ã¦ã²ã¨ã¨ãŠã‚Šã‚«ãƒãƒ¼ã—ã¦ã‹ã‚‰ã€ã•ã¾ã–ã¾ãªåˆ†é‡Žã®äº‹ä¾‹ã‚’æ•°å¤šãå–ã‚Šä¸Šã’ã€çµ±è¨ˆçš„æ‰‹æ³•ã¨æ©Ÿæ¢°å¦ç¿’æ‰‹æ³•ã®ä¸¡æ–¹ã‚’æ™‚ç³»åˆ—ãƒ‡ãƒ¼ã‚¿ã«é©ç”¨ã—ã€ã¾ãŸäººæ°—ã®ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ãƒ„ãƒ¼ãƒ«ã‚‚ç©æ¥µçš„ã«å–ã‚Šå…¥ã‚ŒãŸæ‰‹æ³•ã‚’ç´¹ä»‹ã—ã¾ã™ã€‚ãƒ—ãƒã‚°ãƒ©ãƒ ã«ã¯Rã¨Pythonã®ä¸¡æ–¹ã‚’åˆ©ç”¨ã€‚ãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã‚„ã‚³ãƒ¼ãƒ‰ã¯GitHubã‹ã‚‰ãƒ€ã‚¦ãƒ³ãƒãƒ¼ãƒ‰å¯èƒ½ã§ã™ã€‚ ã¯ã˜ã‚ã« 1ç« ã€€æ™‚ç³»åˆ—ã®æ¦‚è«–ã¨ç°¡å˜ãªæ´å² 1.1ã€€æ™‚ç³»åˆ—ã®å¤šæ§˜ãªç”¨é€”ã®æ´å² 1.1.1ã€€æ™‚ç³»åˆ—å•é¡Œã¨ã—ã¦ã®åŒ»å¦ 1.1.2ã€€æ°—è±¡äºˆæ¸¬ 1.1.3ã€€çµŒæ¸ˆæˆé•·ã®äºˆæ¸¬ 1.1.4ã€€å¤©æ–‡å¦ 1.2ã€€æ™‚ç³»åˆ—è§£æžã®äººæ°—ã«ç«ãŒã¤ã 1.3ã€€çµ±è¨ˆçš„æ™‚ç³»åˆ—è§£æžã®èµ·æº 1.4ã€€
slay-t 2021/09/03
ãƒ‡ãƒ¼ã‚¿

æ©Ÿæ¢°å¦ç¿’

python

book

æœ¬

data

å‹‰å¼·

æ›¸ç±

machinelearning
ãƒªãƒ³ã‚¯
ãƒ“ãƒ¼ãƒ•ã‚¹ãƒˆãƒã‚¬ãƒŽãƒ•ã¯ã©ã®ãã‚‰ã„å¼·ã„ã®ã‹ - Qiita
# !wget https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.ja.300.vec.gzã§è½ã¨ã›ã¾ã™ model = gensim.models.KeyedVectors.load_word2vec_format('cc.ja.300.vec.gz', binary=False) repat = re.compile(r'^[ã‚-ã‚“\u30A1-\u30F4\u4E00-\u9FD0]+$') vocab_list = [w for w in list(model.vocab.keys())[10000:50000] if len(w) > 2 and repat.fullmatch(w) and w[-1] != 'ã£' and w not in list(ww_df.word) and w not in list(sw
slay-t 2021/09/01
qiita

ãƒ‡ãƒ¼ã‚¿

food

æ©Ÿæ¢°å¦ç¿’

ãƒã‚¿

AI

é£Ÿ

python

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°
ãƒªãƒ³ã‚¯
Google Colab
Sign in
slay-t 2021/07/20
work

browser

google

ãƒ–ãƒ©ã‚¦ã‚¶

ä»•äº‹

æ©Ÿæ¢°å¦ç¿’

python

machinelearning

Google Colaboratory
ãƒªãƒ³ã‚¯
pandas ã§ãƒ¡ãƒ¢ãƒªã«ä¹—ã‚‰ãªã„ å¤§å®¹é‡ãƒ•ã‚¡ã‚¤ãƒ«ã‚’ä¸Šæ‰‹ã«æ‰±ã† - StatsFragments
æ¦‚è¦ åˆ†æžã®ãŸã‚ã«ãƒ‡ãƒ¼ã‚¿é›†ã‚ã—ã¦ã„ã‚‹ã¨ã€ãŸã¾ã« ãƒžã‚¸ã‹ï¼? ã¨æ€ã†ã‚µã‚¤ã‚ºã® CSV ã«å‡ºãã‚ã™ã“ã¨ãŒã‚ã‚‹ã€‚ãªãœã“ã‚“ãªã«è‚²ã¤ã¾ã§æ”¾ã£ã¦ãŠã„ãŸã®ã‹ã€ã€ã€ï¼Ÿ ã“ã®ã‚¨ãƒ³ãƒˆãƒªã§ã¯æ™®é€šã«ã¯é–‹ã‘ãªã„ã‚µã‚¤ã‚ºã® CSV ã‚’ pandas ã‚’ä½¿ã£ã¦ã†ã¾ã„ã“ã¨å‡¦ç†ã™ã‚‹æ–¹æ³•ã‚’ã¾ã¨ã‚ãŸã„ã€‚ ã‚µãƒ³ãƒ—ãƒ«ãƒ‡ãƒ¼ã‚¿ ãŸã¾ã«ã¯å®Ÿãƒ‡ãƒ¼ã‚¿ä½¿ãŠã†ã€ã¨ã„ã†ã“ã¨ã§ WorldBankã‹ã‚‰ GDPãƒ‡ãƒ¼ã‚¿ã‚’è½ã¨ã™ã€‚ä»¥ä¸‹ã®ãƒšãƒ¼ã‚¸å³ä¸Šã® "DOWNLOAD DATA" ãƒœã‚¿ãƒ³ã§ CSV ã‚’é¸æŠžã—ã€ãƒãƒ¼ã‚«ãƒ«ã« zip ã‚’ä¿å˜ã™ã‚‹ã€‚è§£å‡ã—ãŸ "ny.gdp.mktp.cd_Indicator_en_csv_v2.csv" ãƒ•ã‚¡ã‚¤ãƒ«ã‚’ã‚µãƒ³ãƒ—ãƒ«ã¨ã—ã¦ä½¿ã†ã€‚ http://data.worldbank.org/indicator/NY.GDP.MKTP.CD?page=1 è£œè¶³ pandas ã® Remote Data Access ã§ WorldBan
slay-t 2021/07/02
ãƒ¡ãƒ¢ãƒª

CD

code

CSV

python

æ©Ÿæ¢°å¦ç¿’

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

programming

data
ãƒªãƒ³ã‚¯
GitHub - microsoft/ML-For-Beginners: 12 weeks, 26 lessons, 52 quizzes, classic Machine Learning for all
ðŸŒ Travel around the world as we explore Machine Learning by means of world cultures ðŸŒ Cloud Advocates at Microsoft are pleased to offer a 12-week, 26-lesson curriculum all about Machine Learning. In this curriculum, you will learn about what is sometimes called classic machine learning, using primarily Scikit-learn as a library and avoiding deep learning, which is covered in our AI for Beginners
slay-t 2021/07/02
ALL

microsoft

as

github

æ©Ÿæ¢°å¦ç¿’

å¦ç¿’

machinelearning

AI

python
ãƒªãƒ³ã‚¯
Streamlitã¨FastAPIã§éžåŒæœŸæŽ¨è«–MLã‚¢ãƒ—ãƒªã‚’ä½œã‚‹
Streamlitã¯Pythonã ã‘ã§webã‚¢ãƒ—ãƒªã‚’ä½œã‚‹ã“ã¨ãŒã§ãã‚‹ãƒ„ãƒ¼ãƒ«ï¼ˆãƒ©ã‚¤ãƒ–ãƒ©ãƒªï¼‰ã§ã™ã€‚ãƒ•ãƒãƒ³ãƒˆã«é–¢ã™ã‚‹çŸ¥è˜ãŒã»ã¨ã‚“ã©ä¸è¦ãªãŸã‚ã€ç°¡å˜ãªãƒ€ãƒƒã‚·ãƒ¥ãƒœãƒ¼ãƒ‰ã‚„ãƒ‡ãƒ¢ã‚¢ãƒ—ãƒªã‚’ä½œã‚‹ã®ã«é©ã—ã¦ã„ã¾ã™ã€‚å…¬å¼ã®ãƒšãƒ¼ã‚¸ã§ã¯æ§˜ã€…ãªã‚µãƒ³ãƒ—ãƒ«ã‚¢ãƒ—ãƒªãŒå…¬é–‹ã•ã‚Œã¦ã„ã¾ã™ã€‚ ã¨ã“ã‚ã§æ©Ÿæ¢°å¦ç¿’ï¼ˆç‰¹ã«æ·±å±¤å¦ç¿’ï¼‰ãƒ¢ãƒ‡ãƒ«ã§ã¯ã€ä¾‹ãˆã°ç”»åƒ1æžšã‚ãŸã‚Šæ•°ç§’ã®æŽ¨è«–æ™‚é–“ãŒã‹ã‹ã‚‹ã“ã¨ã‚‚ã‚ã‚Šã¾ã™ã€‚Streamlitã¯æ©Ÿæ¢°å¦ç¿’ã®ãƒ‡ãƒ¢ã‚¢ãƒ—ãƒªç”¨é€”ã¨ã—ã¦ã‚‚é©ã—ã¦ã„ã‚‹ã¨æ€ã„ã¾ã™ãŒã€æŽ¨è«–ã«æ™‚é–“ãŒã‹ã‹ã‚‹å ´åˆã«ã„ã¡ã„ã¡æŽ¨è«–å®Œäº†ã‚’å¾…ã¤ã®ã¯é€€å±ˆã‹ã‚‚ã—ã‚Œã¾ã›ã‚“ã€‚ã“ã“ã§ã¯Pythonã®webãƒ•ãƒ¬ãƒ¼ãƒ ãƒ¯ãƒ¼ã‚¯ã§ã‚ã‚‹FastAPIã‚’çµ„ã¿åˆã‚ã›ã‚‹ã“ã¨ã§ã€æŽ¨è«–ã‚’éžåŒæœŸã§è¡Œã†ç”»åƒèªè˜ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚’ä½œã‚Šã¾ã™ã€‚ ã‚³ãƒ¼ãƒ‰ã¯ã“ã¡ã‚‰ã«é…ç½®ã—ã¾ã—ãŸã€‚ ã‚¢ãƒ—ãƒªå†…å®¹ Streamlitã«ã‚ˆã‚‹GUIã¯ä»¥ä¸‹ã®ã‚ˆã†ã«ãªã‚Šã¾ã™ã€‚ç”»åƒã‚’ã‚¢ãƒƒãƒ—ãƒãƒ¼ãƒ‰ã—ã€ã€ŒSubmitã€ãƒœã‚¿ãƒ³ã‚’æŠ¼ã™ã“ã¨ã§ç”»åƒèªè˜
slay-t 2021/06/22
ç”»åƒ

file

ã‚¢ãƒ—ãƒª

æ©Ÿæ¢°å¦ç¿’

python

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

programming

machinelearning
ãƒªãƒ³ã‚¯
CyberAgent AI Labã‚’æ”¯ãˆã‚‹Cloudå®Ÿé¨“ç’°å¢ƒ / ML Experiment Management via Cloud Computing Platform in CyberAgent AI Lab
PyData.Tokyo Meetup #23ã§ã®ç™ºè¡¨è³‡æ–™ã§ã™
slay-t 2021/05/28
cloud

AI

data

ç’°å¢ƒ

ã‚¯ãƒ©ã‚¦ãƒ‰

æ©Ÿæ¢°å¦ç¿’

python

machinelearning
ãƒªãƒ³ã‚¯
ã‚µã‚¤ãƒãƒ¼ã‚¨ãƒ¼ã‚¸ã‚§ãƒ³ãƒˆã«ãŠã‘ã‚‹MLOpsã«é–¢ã™ã‚‹å–ã‚Šçµ„ã¿ at PyDataTokyo 23
PyData.Tokyo Meetup #23 MLOpsã€œAIã‚’ç¤¾ä¼šã«å±Šã‘ã‚‹æŠ€è¡“ã§ã®ç™ºè¡¨è³‡æ–™ https://pydatatokyo.connpass.com/event/210654/Read less
slay-t 2021/05/27
ãƒ¡ãƒ¢ãƒª

å¦ç¿’

websocket

python

C

å‹‰å¼·

æ©Ÿæ¢°å¦ç¿’

machinelearning
ãƒªãƒ³ã‚¯
cuGraph ã§ãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯ã‚’è¨ˆç®—ã—ãŸã‚‰çˆ†é€Ÿã ã£ãŸ - Taste of Tech Topics
æ¦‚è¦ ã“ã‚“ã«ã¡ã¯ã€æ©Ÿæ¢°å¦ç¿’ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã®å¤è³€ã§ã™ã€‚ æœ€è¿‘ã€äººã®å‹•ãã‚’æ™‚ç³»åˆ—ã§è§£æžã™ã‚‹ãŸã‚ã«ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ã‚’æ‰±ã£ãŸã®ã§ã™ãŒã€ãƒ‡ãƒ¼ã‚¿é‡ãŒå¤§ãããªã‚‹ã¨è§£æžã«æ™‚é–“ãŒã‹ã‹ã£ã¦ã—ã¾ã„ã€åŠ¹çŽ‡ãŒæ‚ªã„ã¨æ„Ÿã˜ã‚‹ã“ã¨ãŒã‚ã‚Šã¾ã—ãŸã€‚ ãã‚“ãªä¸ã€cuGraph ã¨ã„ã†é«˜é€Ÿã«ã‚°ãƒ©ãƒ•åˆ†æžãŒã§ãã‚‹ãƒ©ã‚¤ãƒ–ãƒ©ãƒªãŒ ã‚ã‚‹ã“ã¨ã‚’çŸ¥ã£ãŸã®ã§ã€ã©ã‚Œãã‚‰ã„é«˜é€Ÿãªã®ã‹ã€æœ‰åãªãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯ã®è¨ˆç®—ã‚’é¡Œæã«ä»–ã®ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã¨é€Ÿåº¦ã‚’æ¯”è¼ƒã—ã¦ã¿ã¾ã—ãŸã€‚ ç›®æ¬¡ã¯ä»¥ä¸‹ã§ã™ã€‚ æ¦‚è¦ ã‚°ãƒ©ãƒ•ã¨ã¯ Python ã«ã‚ˆã‚‹ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ã®åˆ†æž cuGraphã¨ã¯ ãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯ã¨ã¯ ãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯å€¤ã®å®šç¾© ãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯ã¨ã‚°ãƒ©ãƒ• æ¤œè¨¼ å®Ÿè¡Œç’°å¢ƒ cuGraph ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã®ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ« ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã®ã‚¤ãƒ³ãƒãƒ¼ãƒˆ ãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆ æ¤œè¨¼å†…å®¹ãƒ»çµæžœ 1. NetworkX ã®ã‚°ãƒ©ãƒ•ã€NetworkX ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ç”¨ã„ã¦ãƒšãƒ¼ã‚¸ãƒ©ãƒ³ã‚¯ã‚’è¨ˆç®— 2. NetworkX ã®ã‚°ãƒ©ãƒ•ã€cuGr
slay-t 2021/05/25
ãƒ©ã‚¤ãƒ–ãƒ©ãƒª

ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ

tech

web

GPU

ãƒ‡ãƒ¼ã‚¿

js

python

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
StyleGAN2ã§å±žæ€§ã‚’æŒ‡å®šã—ã¦é¡”ç”»åƒã‚’ç”Ÿæˆã™ã‚‹ - ã™ãŽã‚ƒãƒ¼ã‚“ãƒ¡ãƒ¢
memo.sugyan.com ã®è¨˜äº‹ã®ç¶šã(ï¼Ÿ)ã€‚ ã‚ã‚‹ç¨‹åº¦ã®å¦ç¿’ãƒ‡ãƒ¼ã‚¿ã‚’åŽé›†ã—ã¦å¦ç¿’ã•ã›ãŸãƒ¢ãƒ‡ãƒ«ãŒå‡ºæ¥ãŸã®ã§ã€ãã‚Œã‚’ä½¿ã£ã¦å®Ÿéš›ã«è‰²ã€…ã‚„ã£ã¦ã¿ã‚‹ã€‚ StyleGAN2-ADA å¦ç¿’ mappingå‡ºåŠ›ã¨ç”Ÿæˆç”»åƒ ç”Ÿæˆç”»åƒã®å±žæ€§æŽ¨å®šçµæžœã‹ã‚‰æ½œåœ¨ç©ºé–“ã®åã‚Šã‚’æŠ½å‡º è¡¨æƒ…æŽ¨å®š é¡”å§¿å‹¢æŽ¨å®š é«ªé ˜åŸŸæŽ¨å®š (é¡”è§£æž) å¹´é½¢ (ä¸Šæ‰‹ãã„ã‹ãš) è¤‡åˆ Repository StyleGAN2-ADA å‰å›žã®è¨˜äº‹ã§ã‚‚æ›¸ã„ãŸã‘ã©ã€åŽ³é¸ã—ãŸ16,000æžšã®ç”»åƒã‚’ä½¿ã£ã¦ StyleGAN2-ADA ã‚’ä½¿ã£ã¦ç”Ÿæˆãƒ¢ãƒ‡ãƒ«ã‚’å¦ç¿’ã•ã›ã¦ã¿ãŸã€‚ github.com ã“ã‚Œã¯ StyleGAN2 ã‹ã‚‰é€²åŒ–ã—ãŸã‚‚ã®ã§ã€ã‚ˆã‚Šå°‘ãªã„æžšæ•°ã‹ã‚‰ã§ã‚‚å®‰å®šã—ã¦å¦ç¿’ãŒæˆåŠŸã™ã‚‹ã‚ˆã†ã«ãªã£ã¦ã„ã¦ã€ã•ã‚‰ã«parameteræ•°ãªã©èª¿æ•´ã•ã‚Œã¦å¦ç¿’ã‚„æŽ¨è«–ã‚‚ã‚ˆã‚Šæ—©ããªã£ã¦ã„ã‚‹ã€ã¨ã®ã“ã¨ã€‚ ãã‚Œã¾ã§ã®StyleGANã‚·ãƒªãƒ¼ã‚ºã¯TensorFlowã§å®Ÿè£…ã•ã‚Œ
slay-t 2021/04/02
ç”»åƒ

å¦ç¿’

AI

self

å‹‰å¼·

python

Deep Learning

machinelearning

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
Pythonè‡ªç„¶è¨€èªžå‡¦ç†ãƒ†ã‚¯ãƒ‹ãƒƒã‚¯é›†ã€åŸºç¤Žç·¨ã€‘
è‡ªåˆ†ãŒã‚ˆãä½¿ç”¨ã™ã‚‹æ—¥æœ¬èªžè‡ªç„¶è¨€èªžå‡¦ç†ã®ãƒ†ãƒ³ãƒ—ãƒ¬ã‚’ã¾ã¨ã‚ãŸã‚‚ã®ã§ã™ã€‚ ä¸»ã«è‡ªåˆ†ã§ã‚³ãƒ”ãƒšã—ã¦ä½¿ã†ç”¨ã«ã¾ã¨ã‚ãŸã‚‚ã®ã§ã™ãŒã€ã¿ãªã•ã‚“ã®ãŠå½¹ã«ç«‹ã¦ã‚Œã°å¹¸ã„ã§ã™ã€‚ ç’°å¢ƒã¯Python3ç³»ã€Google Colaboratoryï¼ˆUbuntuï¼‰ã§å‹•ä½œç¢ºèªã—ã¦ã„ã¾ã™ã€‚ Pythonã®æ¨™æº–æ©Ÿèƒ½ã¨pipã§å®¹æ˜“ã«ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ã§ãã‚‹ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã«é™å®šã—ã¦ã„ã¾ã™ã€‚ æ©Ÿæ¢°å¦ç¿’ã€ãƒ‡ã‚£ãƒ¼ãƒ—ãƒ©ãƒ¼ãƒ‹ãƒ³ã‚°ã¯å‡ºã¦ãã¾ã›ã‚“ï¼ãƒ†ã‚ã‚¹ãƒˆãƒ‡ãƒ¼ã‚¿ã®å‰å‡¦ç†ãŒä¸å¿ƒã§ã™ã€‚ å‰å‡¦ç†ç³» å¤§æ–‡å—å°æ–‡å— æ—¥æœ¬èªžã®ãƒ†ã‚ã‚¹ãƒˆã«ã‚‚è‹±èªžãŒå‡ºã¦ãã‚‹ã“ã¨ã¯ã‚ã‚‹ã®ã§ã€‚ s = "Youmou" print(s.upper()) # YOUMOU print(s.lower()) # youmou å…¨è§’åŠè§’ æ—¥æœ¬èªžã ã¨ã“ã¡ã‚‰ã®ã»ã†ãŒå¤§äº‹ã€‚ å…¨è§’åŠè§’å¤‰æ›ã®ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã¯ã„ãã¤ã‹ã‚ã‚Šã¾ã™ãŒã€è‡ªåˆ†ã¯jaconvæ´¾ã€‚ MIT Licenseã§åˆ©ç”¨å¯èƒ½ã§ã™ã€‚ import jaco
slay-t 2021/03/30
è¨€èªž

ãƒ©ã‚¤ãƒ–ãƒ©ãƒª

python

language

js

è‡ªç„¶è¨€èªžå‡¦ç†

æ©Ÿæ¢°å¦ç¿’

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

å‹‰å¼·
ãƒªãƒ³ã‚¯
PyTorchã§GPUã®è¨ˆç®—æ™‚é–“ã‚’æ£ã—ãè¨ˆæ¸¬ã™ã‚‹ - ã¾ã£ãŸã‚Šå‹‰å¼·ãƒŽãƒ¼ãƒˆ
æ¸¬å®šçµæžœ çµæžœã¨ã—ã¦ã€é–“é•ã£ãŸæ¸¬å®šæ–¹æ³•ã ã¨CPUã¨GPUã‚’æ¯”è¼ƒã™ã‚‹ã¨ã€Œ1883å€é€Ÿããªã‚Šã¾ã—ãŸï¼ã€ã¨ã„ã†ä¸»å¼µã‚’ã—ã¦ã—ã¾ã†ã“ã¨ã«ãªã‚Šã¾ã™ã€‚ã¡ãªã¿ã«GPUã§1000å€ãªã‚“ã¦æ•°å—ãŒå‡ºã¦ããŸã‚‰ç¢ºå®Ÿã«ã©ã“ã‹é–“é•ãˆã¦ã„ã¾ã™ã€‚å®Ÿéš›ã€ä»Šå›žã®ã‚±ãƒ¼ã‚¹ã§ã¯æœ¬å½“ã¯ã€Œç´„59å€é€Ÿããªã‚Šã¾ã—ãŸï¼ã€ã¨ã„ã†ã®ãŒæ£ã—ã„çµæžœã«ãªã‚Šã¾ã™ã€‚ torch.cuda.synchronize()ã¨torch.cuda.Eventã‚’ä½¿ã£ãŸå ´åˆã®é•ã„ ä»Šå›žtorch.cuda.synchronize()ã¨torch.cuda.Event ã®2ç¨®é¡žã‚’ç´¹ä»‹ã—ã¾ã—ãŸã€‚å ´åˆã«ã‚ˆã£ã¦ã¯ä½¿ã„åˆ†ã‘ã‚’ã—ãŸã»ã†ãŒã„ã„ã®ã§ã“ã®äºŒã¤ã®é•ã„ã‚’èª¬æ˜Žã—ã¦ã„ãã¾ã™ã€‚ torch.cuda.synchronize() ã‚’åˆ©ç”¨ã—ãŸå ´åˆã€ç°¡å˜ãªã®ã§æ¸¬å®šã—ã‚„ã™ã„ã®ã§ã„ã„ã®ã§ã€ã±ã£ã¨æ¸¬å®šã—ãŸã„å ´åˆã¯ã“ã¡ã‚‰ã®æ–¹æ³•ãŒæ¥½ã§ã‚ˆã„ã‹ã¨æ€ã„ã¾ã™ã€‚ãŸã ã€ã“ã¡ã‚‰ã®æ–¹æ³•ã¯kernelé–¢æ•°ã®ç™ºè¡Œã¨æ¸¬
slay-t 2021/03/22
GPU

kernel

event

CPU

å‹‰å¼·

æ©Ÿæ¢°å¦ç¿’

python

ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿
ãƒªãƒ³ã‚¯
ç„¡æ–™ã§èªã‚ã‚‹çµ±è¨ˆå¦ãƒ»æ©Ÿæ¢°å¦ç¿’å‘¨è¾ºã®ãƒãƒ¥ãƒ¼ãƒˆãƒªã‚¢ãƒ«è«–æ–‡ã‚„è¬›ç¾©ãƒŽãƒ¼ãƒˆ10æœ¬ - Qiita
ã¯ã˜ã‚ã« åƒè‘‰å¤§å¦/Nospareã®ç±³å€‰ã§ã™ï¼Žä»Šå›žã¯ï¼Œçµ±è¨ˆå¦ãƒ»æ©Ÿæ¢°å¦ç¿’å‘¨è¾ºã§ï¼Œåƒ•ãŒè‰¯ã„ã¨æ€ã£ãŸãƒãƒ¥ãƒ¼ãƒˆãƒªã‚¢ãƒ«/ã‚µãƒ¼ãƒ™ã‚¤è«–æ–‡ã¨è¬›ç¾©ãƒŽãƒ¼ãƒˆã‚’ç°¡å˜ãªã‚³ãƒ¡ãƒ³ãƒˆä»˜ãã§ç´¹ä»‹ã—ãŸã„ã¨æ€ã„ã¾ã™ï¼Žãƒãƒ¥ãƒ¼ãƒˆãƒªã‚¢ãƒ«è«–æ–‡ã‚„ã‚µãƒ¼ãƒ™ã‚¤è«–æ–‡ã¯ï¼Œãã®ãƒˆãƒ”ãƒƒã‚¯ã®è§£èª¬ã‚„æ•™è‚²é¢ã«ãƒ•ã‚©ãƒ¼ã‚«ã‚¹ã—ã¦ã„ã¦ï¼Œä½•ã‹å‹‰å¼·ã—ãŸã‚Šï¼Œç¶²ç¾…çš„ã«æŠŠæ¡ã™ã‚‹ã¨ãã«éžå¸¸ã«ä¾¿åˆ©ã ã¨å€‹äººçš„ã«æ€ã„ã¾ã™ï¼Žã¾ãŸå…¬é–‹ã•ã‚Œã¦ã„ã‚‹è¬›ç¾©ãƒŽãƒ¼ãƒˆã®ä¸ã«ã¯éžå¸¸ã«å‹‰å¼·ã«ãªã‚‹ã‚‚ã®ãŒå¤šãã‚ã‚Šã¾ã™ï¼Žå†…å®¹ã¯åƒ•ãŒèˆˆå‘³ãŒã‚ã‚‹ãƒˆãƒ”ãƒƒã‚¯ã«åã£ã¦ã„ã¾ã™ï¼Žã¾ãŸã™ã¹ã¦ç„¡æ–™ã§èªã‚ã¾ã™ï¼Žï¼ˆæ•™ç§‘æ›¸ç‰ã®æµ·è³Šç‰ˆã¿ãŸã„ãªã®ã¯è¼‰ã›ã¦ã„ã¾ã›ã‚“ï¼‰ 10æœ¬ã®ç´¹ä»‹ Nickl "STATISTICAL THEORY" Nicklã®çµ±è¨ˆå¦ã®è¬›ç¾©ãƒŽãƒ¼ãƒˆã§ã™ï¼Žã„ã‚ã‚†ã‚‹MæŽ¨å®šé‡ã®æ¼¸è¿‘çš„æ€§è³ªã«åŠ ãˆï¼Œãƒãƒ¼ãƒ³ã‚¹ã‚¿ã‚¤ãƒ³ãƒ»ãƒ•ã‚©ãƒ³ãƒŸãƒ¼ã‚¼ã‚ºå®šç†ç‰ã‚‚è¨¼æ˜Žä»˜ãã§è§£èª¬ã•ã‚Œã¦ãŠã‚Šï¼Œä¸Šç´šãƒ¬ãƒ™ãƒ«ã®æ•°ç†çµ±è¨ˆå¦ã‚’å¦ã¶ã®ã«é‡å®ã™ã‚‹ã¨æ€ã„ã¾ã™ï¼Ž Doucet and
slay-t 2021/01/25
å¦ç¿’

æœ¬

å‹‰å¼·

qiita

ãƒ‡ãƒ¼ã‚¿

tutorial

æ©Ÿæ¢°å¦ç¿’

æ•°å¦

python
ãƒªãƒ³ã‚¯
GitHub - ml-tooling/best-of-ml-python: ðŸ† A ranked list of awesome machine learning Python libraries. Updated weekly.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
slay-t 2021/01/20
github

python

clone

apache

https

git

install

å¦ç¿’

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
1 2 3 4 æ¬¡ã®ãƒšãƒ¼ã‚¸