Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Char-rnn aurkezpena
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
Manex Agirrezabal
March 14, 2016
Research
110
0
Share
Char-rnn aurkezpena
Manex Agirrezabal
March 14, 2016
More Decks by Manex Agirrezabal
See All by Manex Agirrezabal
The Flipped Classroom model for teaching Conditional Random Fields in an NLP course
manexagirrezabal
0
46
NLP for poetry generation and analysis
manexagirrezabal
0
85
Institut seminar 2020
manexagirrezabal
0
43
Automatic Scansion of Poetry (KU)
manexagirrezabal
0
680
RANLP talk
manexagirrezabal
0
81
Defense (Final version)
manexagirrezabal
0
82
Poesiaren eskantsio automatikoa: Bi hizkuntzen azterketa
manexagirrezabal
0
82
CodeFEST literature presentation
manexagirrezabal
0
68
Ongoing work (in mid 2016)
manexagirrezabal
0
30
Other Decks in Research
See All in Research
はじまりの クエスチョンブック —余暇と豊かさにあふれた社会とは?
culturaltransition
PRO
0
500
Dual Quadric表現を用いた動的物体追跡とRGB-D・IMU制約の密結合によるオドメトリ推定
nanoshimarobot
0
400
ローテーション別のサイドアウト戦略 ~なぜあのローテは回らないのか?~
vball_panda
0
340
2026年3月1日(日)福島「除染土」の公共利用をかんがえる
atsukomasano2026
0
620
LLM の Attention 機構まとめ — 数式・計算量・メモリ
puwaer
7
2k
都市交通マスタープランとその後への期待@熊本商工会議所・熊本経済同友会
trafficbrain
0
220
「行ける・行けない表」による地域公共交通の性能評価
bansousha
0
160
Scalable dynamic origin-destination demand estimation enhanced by high-resolution satellite imagery data
satai
2
250
PGDM: Physically Guided Diffusion Model for L Downscaling
satai
0
240
[BlackHatAsia2026] Hidden Telemetry: Uncovering TraceLogging ETW Providers You're Not Using (Yet)
asuna_jp
1
490
2026 東京科学大 情報通信系 研究室紹介 (すずかけ台)
icttitech
0
3.7k
SoftMatcha 2: 1兆語規模コーパスの超高速かつ柔らかい検索
e869120_sub
6
3.4k
Featured
See All Featured
Building an army of robots
kneath
306
46k
Faster Mobile Websites
deanohume
310
31k
AI in Enterprises - Java and Open Source to the Rescue
ivargrimstad
0
1.3k
Jamie Indigo - Trashchat’s Guide to Black Boxes: Technical SEO Tactics for LLMs
techseoconnect
PRO
0
160
DevOps and Value Stream Thinking: Enabling flow, efficiency and business value
helenjbeal
1
220
Bootstrapping a Software Product
garrettdimon
PRO
307
120k
Gemini Prompt Engineering: Practical Techniques for Tangible AI Outcomes
mfonobong
2
420
Imperfection Machines: The Place of Print at Facebook
scottboms
270
14k
Agile Actions for Facilitating Distributed Teams - ADO2019
mkilby
0
200
Technical Leadership for Architectural Decision Making
baasie
3
400
The Illustrated Guide to Node.js - THAT Conference 2024
reverentgeek
1
370
The Curse of the Amulet
leimatthew05
1
13k
Transcript
Poesiaren metrika DL bidez Manex Agirrezabal https://github.com/manexagirrezabal/char-rnn/
Proba ezberdinak TensorFlow: Sequence-to-sequence models https://www.tensorflow.org/versions/master/tutorials/seq2seq/index.html Torch: char-rnn (Andrew Karpathy)
https://github.com/karpathy/char-rnn/
Char-rnn http://karpathy.github.io/2015/05/21/rnn-effectiveness/ Karaktere mailako hizkuntz-ereduak sortzeko balio du. Sarrera gisa
testu hutsa.
Char-rnn Gure beharretarako moldatu behar: to swell the gourd and
plump the ha zel shells - ' - ' - ' - ' - ' wo man much missed how you call to me call to me ' - - ' - - ' - - ' - -
Char-rnn Dataset-a testu soil gisa: To_= swell_+ the_= gourd_+ and_=
plump_+ the_= ha_+ zel_= shells_+ To_= swell_+ the_= gourd_+ and_= plump_+ the_= hazel_+= shells_+ Wo_+ man_= much_= missed_+ how_= you_= call_+ to_= me_= call_+ to_= me_= Woman_+= much_= missed_+ how_= you_= call_+ to_= me_= call_+ to_= me_=
Char-rnn (training) $ th train.lua Parametroak: Model: [RNN, LSTM edo
GRU] rnn_size: LSTMaren (zelda) barruko tamaina num_layers: LSTMaren kapa kopurua seq_length: sekuentzian ikasteko karaktere kopurua
Char-rnn (prediction) $ th sample(mod).lua Parametroak: Model: eredu entrenatua Primetext:
sarrera testua (_ karakterearekin amaituta)
Char-rnn (prediction) Python programa bat (callSampleMod.py) aurreko programari deitzeko pausuz
pausu: $ th sampleMod.lua model M1 primetext “to_” = $ th sampleMod.lua model M1 primetext “to_= swell_” + $ th sampleMod.lua model M1 primetext “to_= swell_+ the_” = ...
Char-rnn (prediction) Arazoa: Hasieran, informazio gutxi duenez, batzuetan hanka sartzen
(+ propagatzen) du predikzioan. Adibidez, “to_” sarrerarekin Horrentzako soluzioa, predikzioa bi aldetara egitea.
Char-rnn (FW) Parametroak optimizatu nahi ditugu (seq_length, batch_size, rnn_size, ...)
Embedding-ak erabili nahi ditugu, baina gure hipotesia da ez dutela asko lagunduko.