GitHub - chenfei-wu/TaskMatrix
Now TaskMatrix supports GroundingDINO and segment-anything! Thanks @jordddan for his efforts. For the image editing case, GroundingDINO is first used to locate bounding boxes guided by given text, then segment-anything is used to generate the related mask, and finally stable diffusion inpainting is used to edit image based on the mask. Firstly, run python visual_chatgpt.py --load "Text2Box_cuda:0,
データã®æ°‘主化ã¨ã“ã‚Œã‹ã‚‰ã®AI組織｜ã°ã‚“ãã—
ã¯ã˜ã‚ã«Stable Diffusionã ã¨ã‹ChatGPTã€LLMã¿ãŸã„ãªã€Œå¤§è¦æ¨¡ãƒ¢ãƒ‡ãƒ«ã€ã£ã¦è€ƒãˆæ–¹ãŒæ©Ÿæ¢°å¦ç¿’æ¥ç•Œã‹ã‚‰å‡ºã¦ã€ã‚¹ã‚±ãƒ¼ãƒªãƒ³ã‚°å‰‡ã«åŸºã¥ã„ã¦ã¾ã ã¾ã 精度ãŒä¸ŠãŒã‚‹ã¨ã•ã‚Œã¦ã„る昨今。 （スケーリング則ã¯ã©ã†ã®ã“ã†ã®è«¸èª¬ã‚ã‚‹ãŒï¼‰ã•ã¦ãŠãã€ã€Œãƒžãƒ«ãƒãƒ¢ãƒ¼ãƒ€ãƒ«ã«ã€ã€Œã‚らゆるデータをå¦ç¿’ã—ãŸã€ã€Œå¤§è¦æ¨¡ãªãƒ¢ãƒ‡ãƒ«ã€ãŒä»Šå¾Œæ•°å¹´ãƒªãƒ¼ãƒ‰ã—ã¦ã„ã事ã¯é–“é•ã„ãªã„ã¨æ€ã†ã€‚ ãã‚“ãªä¸ã§ã€æˆ‘々機械å¦ç¿’エンジニアやデータサイエンティストã€ã‚¢ãƒŠãƒªã‚¹ãƒˆã€ãƒ‡ãƒ¼ã‚¿ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã€MLOpsエンジニアã¿ãŸã„ãªã€ã„ã‚ゆるAI屋ã¨ã—ã¦åƒã„ã¦ã„る人ãŸã¡ã€çš†ãŒæ‰€å±žã™ã‚‹AI組織ã£ã¦ã©ã†ãªã£ã¦ã„ãã®ã‹ãªã€ã¨ã„ã†è©±ã‚’書ã„ã¦ã¿ã‚‹ã€‚ データã®æ°‘主化AIã®æ°‘主化ã¨ãƒ‡ãƒ¼ã‚¿ã®æ°‘主化AIæ¥ç•Œã§ã¯ã€ŒAIã®æ°‘主化ã€ã¨ã„ã†ãƒ¯ãƒ¼ãƒ‰ãŒã‚る。 便宜的ã«AIã¨ã„ã†ãƒ¯ãƒ¼ãƒ‰ãŒåºƒã使ã‚れるよã†ã«ãªã£ãŸè¾ºã‚Šã§å‡ºã¦ããŸãƒ¯ãƒ¼ãƒ‰ã§ã€OSSやプラットフォームã€ãƒãƒ¼ãƒ‰ã®ç™ºå±•ã«ã‚ˆã£ã¦ã€ŒA
LangChain Python API Reference Documentation.
You will be automatically redirected to the new location of this page.
LLMãŒãªãœå¤§äº‹ãªã®ã‹?経営者ã®è¦–点ã§è€ƒãˆã‚‹æ³¢ã®å¾…ã¡å—ã‘方｜ç¦å³¶è‰¯å…¸ | LayerX
ã¯ã˜ã‚ã«LayerXã®ä»£è¡¨ã‚’ã—ã¦ã„ã¾ã™ç¦å³¶ã¨ç”³ã—ã¾ã™ã€‚本日ã¯LLM(Large Language Model, 大è¦æ¨¡è¨€èªžãƒ¢ãƒ‡ãƒ«)ã«ã¤ã„ã¦ã€ãªãœå¤§äº‹ãªã®ã‹?経営者ã®è¦–点ã§ã©ã†ã“ã®æ³¢ã‚’考ãˆã‚Œã°ã„ã„ã®ã‹ã«ã¤ã„ã¦ã§ã™ã€‚ LLMãŒä»Šéžå¸¸ã«è©±é¡Œã«ãªã£ã¦ã„ã¾ã™ã€‚ChatGPTã®è£å´ã‚‚ã“ã®LLMã‹ã‚‰ã§ãã¦ã„ã‚‹ã¨è¨€ãˆã°ã‚ã‹ã‚Šã‚„ã™ã„ã§ã—ょã†ã‹ã€‚ ç§ã¯ç¾åœ¨ã€LayerXã¨ã„ã†ä¼šç¤¾ã‚’経営ã—ã¦ãŠã‚Šã¾ã™ã€‚LayerXã§ã‚‚多分ã«æ¼ã‚Œãšã€LLMã«å¯¾ã™ã‚‹ã‚‚ã®ã™ã”ã„ワクワクã¨ã€ã“ã®æ³¢ã«å¯¾å¿œã—ãªã„ã¨ä¼šç¤¾ãŒæ¶ˆãˆã¦ãªããªã‚‹ã¨ã„ã†å¼·ã„å±æ©Ÿæ„Ÿã‚’抱ã„ã¦ã„ã¾ã™ã€‚ ç§è‡ªèº«ãŒå…ƒæ©Ÿæ¢°å¦ç¿’エンジニアã€ç¾åœ¨ç¾å½¹ã®çµŒå–¶è€…ã¨ã„ã†ã‚ャリアをæ©ã‚“ã§ãã¾ã—ãŸã€‚ãã®ç«‹å ´ã‹ã‚‰ã€ãªã‚‹ã¹ãã‚ã‹ã‚Šã‚„ã™ãã€LLMã®æ³¢ã¨ã„ã†ã‚‚ã®ã‚’経営者ãŒã©ã†æ‰ãˆã‚‹ã¹ãã‹ã€ã©ã†æ³¢ã‚’ä¹—ã‚Šã“ãªã™ã¹ãã‹ã®ä¸€åŠ©ã¨ãªã‚Œã°ã¨æ€ã„ç†ã‚’å–ã£ã¦ã„ã¾ã™ã€‚(機械å¦ç¿’ã®ãƒ—ãƒã®æ–¹ã‹ã‚‰ã™ã‚‹ã¨ã€ãŠã„ãŠã„ãã‚Œã¯å˜ç´”化ã—
Azure OpenAI Service モデル - Azure OpenAI
ã“ã®ãƒ–ラウザーã¯ã‚µãƒãƒ¼ãƒˆã•ã‚Œãªããªã‚Šã¾ã—ãŸã€‚ Microsoft Edge ã«ã‚¢ãƒƒãƒ—グレードã™ã‚‹ã¨ã€æœ€æ–°ã®æ©Ÿèƒ½ã€ã‚»ã‚ュリティ更新プãƒã‚°ãƒ©ãƒ ã€ãŠã‚ˆã³ãƒ†ã‚¯ãƒ‹ã‚«ãƒ« サãƒãƒ¼ãƒˆã‚’利用ã§ãã¾ã™ã€‚
Customize a model with Azure OpenAI Service - Azure OpenAI
Azure OpenAI Service lets you tailor our models to your personal datasets by using a process known as fine-tuning. This customization step lets you get more out of the service by providing: Higher quality results than what you can get just from prompt engineering The ability to train on more examples than can fit into a model's max request context limit. Token savings due to shorter prompts Lower-
OpenAI Whisper ã«è¿½åŠ å¦ç¿’ã‚’ã•ã›ã‚‹è©¦ã¿ | IIJ Engineers Blog
åœ°æ–¹æ‹ ç‚¹ã®ä¸€ã¤ã€ä¹å·žæ”¯ç¤¾ã«æ‰€å±žã—ã¦ã„ã¾ã™ã€‚サーãƒãƒ»ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚’ä¸å¿ƒã¨ã—ãŸSIæ¥å‹™ã«æºã‚ã£ã¦ã¾ã—ãŸãŒã€ç¾åœ¨ã¯æŠ€è¡“æŽ¢ç´¢ãƒ»æ·±å €æ¥å‹™ã‚’ä¸å¿ƒã«å¯¾å¿œã—ã¦ã„ã¾ã™ã€‚ 2018å¹´ã«é›£ç—…ã‚’æ‚£ã£ãŸã“ã¨ã«ã‚ˆã‚Šã€å®šæœŸçš„ã«å…¥é€€é™¢ã‚’ç¹°ã‚Šè¿”ã—ã¤ã¤ã€2023å¹´ã«ã¯ç”·æ€§æ›´å¹´æœŸéšœå®³ã®ç™ºç—‡ã‚’ãã£ã‹ã‘ã«ã€ãƒˆãƒ©ãƒ³ã‚¹ã‚¸ã‚§ãƒ³ãƒ€ãƒ¼ã¨ã—ã¦ã®é“ã‚’æ©ã¿å§‹ã‚ã¦ã¾ã™ã€‚ ã€IIJ 2022 TECHアドベントカレンダー 12/8（木）ã®è¨˜äº‹ã§ã™ã€‘ Whisperã®è¿½åŠ å¦ç¿’ã«æŒ‘ã‚€2022å¹´ã®å†¬ 2022年アドベントカレンダーä¼ç”»ã ãã†ã§ã™ã€‚ ã„ã‹ãŒãŠéŽã”ã—ã§ã—ょã†ã‹ã€‚ ç§ã¯ã‚‚ã†èˆˆå‘³ã‚’引ãã‚‚ã®ã«æ²¡é ã™ã‚‹ã—ã‹æ¥½ã—ã¿ãŒãªã„ã‚‚ã‚“ã§ã€PCã«å‘ã‹ã£ã¦ã¼ã‚“ã‚„ã‚Šé¢ç™½ã„ãƒã‚¿ã¯ãªã„ã‹ãªãƒ¼ã¨æŽ¢ã™æ—¥ã€…ã§ã™ã€‚ 最近ã¯ã™ã£ã‹ã‚Šãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã«ã‚ºãƒ–ズブã§ã€ã¨ã†ã¨ã†æ•°å¼ã‹ãã®ãŒé¢å€’ãªã‚ã¾ã‚Šæ‰‹æ›¸ã入力ã®ãŸã‚ã®ãƒšãƒ³ã‚¿ãƒ–レットを買ã„ã¾ã—ãŸã€‚ã¦ã¸ã€‚ 今回ã¯9月ã‹ã‚‰10月ã«ã‹ã‘ã¦Stabl
1
ランã‚ング
ランã‚ング
ãŠçŸ¥ã‚‰ã›
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
VectorStores — 🦜🔗 LangChain 0.0.146
GitHub - openai/whisper: Robust Speech Recognition via Large-Scale Weak Supervision
GitHub - rakyll/openai-go: Go client libraries for OpenAI
Go · ggerganov/whisper.cpp · Discussion #312
{{#tags}}- {{label}}
{{/tags}}