profile
๐Ÿ‘‹ ์ธ๊ณต์ง€๋Šฅ์„ ํ†ตํ•ด ๋‹ค์Œ ์„ธ๋Œ€๊ฐ€ ๋” ๋‚˜์€ ์‚ถ์„ ์‚ด๋„๋ก
post-thumbnail

ํ™•๋ฅ ๊ณผ ์ƒ์„ฑํ˜• ๋ชจ๋ธ์— ๋Œ€ํ•œ ํšŒ๊ณ 

ํ™•๋ฅ ์— ๋Œ€ํ•œ ์ธ์‹์˜ ๋ณ€ํ™”์™€ ์ƒ์„ฑํ˜• ๋ชจ๋ธ๊ณผ์˜ ์—ฐ๊ด€์„ฑ.

2024๋…„ 7์›” 25์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

Colab ์žฅ๊ธฐ๊ฐ„ ํ•™์Šต ๋ฌธ์ œ

Colab ์—์„œ ์ธ๊ณต์ง€๋Šฅ ๋ชจ๋ธ์„ ํ•™์Šตํ•  ๋•Œ ์žฅ๊ธฐ๊ฐ„ ํ•™์Šต์‹œ ๋ฐœ์ƒํ•˜๋Š” ์˜ค๋ฅ˜์— ๋Œ€ํ•ด์„œ ์•Œ์•„๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 5์›” 24์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

[PyTorch] ๊ฐ€์ค‘์น˜ ๊ณ ์ •

์ด๋ฒˆ ๊ธ€์—์„œ๋Š” PyTorch์—์„œ ์ฃผ์š”ํ•˜๊ฒŒ ์‚ฌ์šฉ๋˜๋Š” ๊ฐ€์ค‘์น˜ ๊ณ ์ • ๋ฐฉ๋ฒ•์ธ requires_grad, torch.no_grad(), detach(), Optimizer์— ๋ณ€์ˆ˜ ์ œ๊ณต ๋ฐฉ๋ฒ• 4๊ฐ€์ง€์— ๋Œ€ํ•ด์„œ ์„ค๋ช…ํ•˜๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 5์›” 6์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

[PyTorch] AutoGrad๋ž€ ๋ฌด์—‡์ธ๊ฐ€? (2)

์ด๋ฒˆ ๊ธ€์—์„œ๋Š” AutoGrad๋Š” ์–ด๋–ป๊ฒŒ ์ž‘๋™ํ•˜๋Š”์ง€์— ๋Œ€ํ•ด ์ˆœ์ฐจ์ ์œผ๋กœ ์•Œ์•„๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 5์›” 1์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

[PyTorch] AutoGrad๋ž€ ๋ฌด์—‡์ธ๊ฐ€? (1)

์ด๋ฒˆ ๊ธ€์—์„œ๋Š” PyTorch์—์„œ Gradient๋ฅผ ๊ณ„์‚ฐํ•ด์ฃผ๋Š” AutoGrad๊ฐ€ ๋ฌด์—‡์ด๊ณ  ์–ด๋–ค ์‹œ์Šคํ…œ์„ ๋”ฐ๋ฅด๋Š”์ง€ ์•Œ์•„๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 4์›” 20์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

SKT FLY AI Challenger 4๊ธฐ ์ˆ˜๋ฃŒ ํ›„๊ธฐ

์ข‹์€ ํŒ€์„ ๋งŒ๋‚˜ ์ข‹์€ ํ™˜๊ฒฝ์—์„œ ์ˆ˜์ƒ๊นŒ์ง€!

2024๋…„ 3์›” 1์ผ
ยท
10๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

[๋”ฅ๋Ÿฌ๋‹] ๊ฒฝ์‚ฌํ•˜๊ฐ•๋ฒ• ๊ตฌํ˜„๋ถ€ํ„ฐ ํ•™์Šต๊นŒ์ง€ (2)

์ด๋ฒˆ ๊ธ€์—์„œ๋Š” ์ˆœ์ „ํŒŒ์™€ ์—ญ์ „ํŒŒ์— ๋Œ€ํ•ด ์•Œ์•„๋ณด๊ณ , ์ดํ•ด๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ์ง์ ‘ ๊ตฌํ˜„ํ•˜์—ฌ ๊ฐ„๋‹จํ•œ ๋ชจ๋ธ์„ ํ•™์Šตํ•˜๋Š” ์ฝ”๋“œ๋ฅผ ์ž‘์„ฑํ•ด๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 2์›” 4์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

[๋”ฅ๋Ÿฌ๋‹] ๊ฒฝ์‚ฌํ•˜๊ฐ•๋ฒ• ๊ตฌํ˜„๋ถ€ํ„ฐ ํ•™์Šต๊นŒ์ง€ (1)

์ด๋ฒˆ ๊ธ€์—์„œ๋Š” ๊ฒฝ์‚ฌํ•˜๊ฐ•๋ฒ•์ด ๋ฌด์—‡์ด๊ณ  ์–ด๋–ป๊ฒŒ ์ž‘๋™ํ•˜๋Š”์ง€ ์•Œ์•„๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.

2024๋…„ 1์›” 27์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

BERTSUM ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” Text Summarization with Pretrained Encoders ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2024๋…„ 1์›” 14์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

SKT FLY AI Challenger 4๊ธฐ ํ•ฉ๊ฒฉ ํ›„๊ธฐ

์ฒ˜์Œ์œผ๋กœ ์ตœ์ข… ํ•ฉ๊ฒฉ์„?

2023๋…„ 11์›” 29์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

MIC ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2023๋…„ 11์›” 25์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

HRDA ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2023๋…„ 11์›” 25์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

DAFormer ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2023๋…„ 11์›” 24์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

DACS ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” DACS: Domain Adaptation via Cross-domain Mixed Sampling ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2023๋…„ 11์›” 24์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

SegFormer ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ

๋ณธ ํŽ˜์ด์ง€์—์„œ๋Š” SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers ๋…ผ๋ฌธ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.

2023๋…„ 11์›” 24์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

2023 Samsung AI Challenge : Camera-Invariant Domain Adaptation ํšŒ๊ณ 

ํฌ๊ธฐํ•˜์ง€ ๋ง๊ณ  ๋๊นŒ์ง€!

2023๋…„ 11์›” 16์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

LG Aimers 3๊ธฐ Phase 2 ์˜จ๋ผ์ธ ํ•ด์ปคํ†ค ํšŒ๊ณ 

๋ฐ์ดํ„ฐ ๋ถ„์„๊ณผ ์†Œํ†ต์˜ ์ค‘์š”์„ฑ

2023๋…„ 11์›” 12์ผ
ยท
4๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

LG Aimers 3๊ธฐ Phase 1 ์˜จ๋ผ์ธ ๊ต์œก ํ›„๊ธฐ

๊ธฐ์ดˆ์ง€์‹์˜ ๋ถ€์žฌ

2023๋…„ 11์›” 12์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

Segmentation๊ณผ ๋ณด๊ฐ„๋ฒ•

Segmentation ์ž‘์—…์„ ํ•  ๋•Œ Mask์— ๋Œ€ํ•ด์„œ Resize๋ฅผ ํ•ด์ฃผ๋Š” ๊ฒฝ์šฐ ์ฃผ์˜ํ•  ์ ์— ๋Œ€ํ•ด์„œ ์•Œ์•„๋ณด์ž.

2023๋…„ 10์›” 26์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท
post-thumbnail

๋”ฅ๋Ÿฌ๋‹์—์„œ Bottleneck์— ๋Œ€ํ•œ ์‹คํ—˜

Bottleneck์€ ์–ด๋–ค ์—ญํ• ์„ ํ• ๊นŒ?

2023๋…„ 8์›” 4์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท