VVoice Analysis
Read More
  • 2 minute read

[๋ฒˆ์—ญ] Baidu Deep Voice: Part 2 โ€“ ํ•™์Šต (Training)

Deep Voice ํŒŒํŠธ 1 ๋งํฌ ์ด ๊ธ€์€ ์ €์ž(Dhruv Parthasarathy)์˜ ํ—ˆ๋ฝ์„ ๋ฐ›์•„ ๋ฒˆ์—ญํ•˜์—ฌ ๊ฒŒ์‹œํ•˜๋Š” ๊ธ€์ž…๋‹ˆ๋‹ค. ์›๋ฌธ์€ https://blog.athelas.com/baidu-deep-voice-explained-part-2-training-810e87d20047์—์„œ ํ™•์ธํ•  ์ˆ˜…
๐Ÿ’ป๐Ÿ’ป Programming
Read More
  • 1 minute read

[R] anaconda๋ฅผ ์ด์šฉํ•œ R package ์„ค์น˜

Jupyter notebook์„ ์ด์šฉํ•ด์„œ R์„ ์‚ฌ์šฉํ•  ๋•Œ package ์„ค์น˜๊ฐ€ ์šด์˜์ฒด์ œ๋ณ„๋กœ ๋‹ค๋ฅด๊ฒŒ ์ง„ํ–‰๋  ์ˆ˜ ์žˆ๋Š” ๊ฒƒ ๊ฐ™๋‹ค. ์œˆ๋„์šฐ์—์„œ๋Š” ์•„๋ž˜์™€ ๊ฐ™์ด…
VVoice Analysis
Read More
  • 2 minute read

[๋ฒˆ์—ญ] Baidu Deep Voice: Part 1 – Text-to-speech ํŒŒ์ดํ”„๋ผ์ธ(The Inference Pipeline)

๋“ค์–ด๊ฐ€๋ฉฐ ์ด ๊ธ€์€ 2017๋…„ 3์›”์— ์ž‘์„ฑ๋œ ๋‚ด์šฉ์œผ๋กœ, ๋”ฅ๋Ÿฌ๋‹ ๋ชจ๋ธ, ์•Œ๊ณ ๋ฆฌ์ฆ˜์˜ ๋ฐœ์ „ ์†๋„๋ฅผ ์ƒ๊ฐํ•ด๋ณด๋ฉด 2๋…„๊ฐ„์˜ ์ฐจ์ด๋Š” ์ƒ๋‹นํžˆ ํฌ๋‹ค๊ณ  ๋ณผ…
PPython
Read More
  • 1 minute read

Mask R-CNN์„ ์ด์šฉํ•œ ๊ณ ๋ง‰ ๊ฒ€์ถœ ์—ฐ๊ตฌ (The semantic segmentation approach for normal and pathologic tympanic membrane using deep learning)

bioRxiv ๋งํฌ: https://www.biorxiv.org/content/early/2019/01/10/515007   ์œ„ ์—ฐ๊ตฌ๋Š” 2018๋…„ 10์›” ๋Œ€ํ•œ์ด๋น„์ธํ›„๊ณผํ•™ํšŒ ํ•™์ˆ ๋Œ€ํšŒ์—์„œ ์ฒ˜์Œ ๋ฐœํ‘œํ–ˆ์Šต๋‹ˆ๋‹ค. 11์›” ์ค‘์ˆœ ๋ถ€ํ„ฐ ํˆฌ๊ณ ๋ฅผ ์ง„ํ–‰ํ–ˆ์œผ๋‚˜ ๋‘์–ด์ฐจ๋ก€…
RR Statistics
Read More
  • 2 minute read

[R] ๊ฐ„๋‹จํ•œ ํ†ต๊ณ„ ๊ณผ์ • ๋ชฉ๋ก (์—‘์…€ xlsx ๋ถˆ๋Ÿฌ์˜ค๊ธฐ ๋ถ€ํ„ฐ)

๊ฐ„๋‹จํ•œ ์—ฐ๊ตฌ๋ฅผ ํ•˜๊ธฐ ์œ„ํ•ด ์ง„ํ–‰ํ–ˆ๋˜ flow ์ •๋ฆฌ #1: 1๋ฒˆ ์œ„ํฌ์‹œํŠธ๋ฅผ ๋ถˆ๋Ÿฌ์˜ด #2: 1,2,3์œผ๋กœ ์ ํ˜€์žˆ๋Š” ๋ฐ์ดํ„ฐ์— ๊ฐ๊ฐ์˜ label ๋ถ€์—ฌ…
๐Ÿ’ป๐Ÿ’ป Programming
Read More
  • 3 minute read

[๋ฒˆ์—ญ๊ธ€] Image Segmentation์— ๋Œ€ํ•œ ์งง์€ ์ด์•ผ๊ธฐ: R-CNN ์—์„œ ๋ถ€ํ„ฐ Mask R-CNN ๊นŒ์ง€

  ์ œ ์ฒซ deep learning ์—ฐ๊ตฌ๋ฅผ ์•„์นด์ด๋ธŒ์— ์˜ฌ๋ ธ์Šต๋‹ˆ๋‹ค. Mask R-CNN์„ ์ด์šฉํ•œ ๊ณ ๋ง‰ ๊ฒ€์ถœ ์—ฐ๊ตฌ (The semantic segmentation approach…
PPython

4. Cocoapi setup @ python 3.5

Stackoverflow๋ฅผ ๋ณด๋ฉด python3์—์„œ๋Š” ์ž‘๋™์ด ์•ˆ๋œ๋‹ค๊ณ ๋“ค ํ•˜๋Š”๋ฐ, ์–ด์ฉŒ๋‹ค๋ณด๋‹ˆ ์‹คํ–‰์ด ๋˜๋Š” ๋“ฏํ•ด์„œ ๊ธฐ๋ก์— ๋‚จ๊ธด๋‹ค. 1. pip์—์„œ git๋ฅผ ์ด์šฉํ•˜๊ธฐ ์œ„ํ•ด git๋ฅผ…
PPython
Read More
  • 1 minute read

3. ํ•œ ๋‹จ๊ณ„๋ฅผ ๋„˜์–ด๊ฐ„๋‹ค.

์ฐธ ํ•ด๋ณด๊ณ ๋‚˜๋ฉด ๋ณ„๊ฑฐ ์•„๋‹ˆ์ง€๋งŒ, ์‹œํ–‰์ฐฉ์˜ค๊ฐ€ ๋งŽ๋‹ค. (1) Class activation map (2) ๊ทธ๋ฆฌ๊ณ  ์ด๊ฒƒ์„ ๋ฝ‘์•„๋‚ด๊ธฐ ์œ„ํ•ด ๋งˆ์ง€๋ง‰ convolution layer๋ฅผ…
โœโœ๏ธ Monolog
Read More
  • 1 minute read

On going

ResNet50 ๋ชจ๋ธ์„ ๊ธฐ๋ฐ˜์œผ๋กœ ๋Œ๋ฆฐ ๋‚˜์˜ ๋ฐ์ดํ„ฐ. Overfitting ๋ฌธ์ œ์ธ์ง€ ์•„๋‹ˆ๋ฉด ๊ทธ๋ƒฅ ๋ถ€์ง„ํ•œ ํ•™์Šต์˜ ๋ฌธ์ œ์ธ์ง€, Epoch์„ 50๊นŒ์ง€ ๋Š˜๋ ธ๋Š”๋ฐ, ํ•œ ์—ํญ๋‹น…