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
Manex Agirrezabal
March 14, 2016
Research
0
83
Char-rnn aurkezpena
Manex Agirrezabal
March 14, 2016
Tweet
Share
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
26
NLP for poetry generation and analysis
manexagirrezabal
0
66
Institut seminar 2020
manexagirrezabal
0
28
Automatic Scansion of Poetry (KU)
manexagirrezabal
0
500
RANLP talk
manexagirrezabal
0
70
Defense (Final version)
manexagirrezabal
0
68
Poesiaren eskantsio automatikoa: Bi hizkuntzen azterketa
manexagirrezabal
0
67
CodeFEST literature presentation
manexagirrezabal
0
58
Ongoing work (in mid 2016)
manexagirrezabal
0
21
Other Decks in Research
See All in Research
Poster: Feasibility of Runtime-Neutral Wasm Instrumentation for Edge-Cloud Workload Handover
chikuwait
0
290
Weekly AI Agents News! 9月号 プロダクト/ニュースのアーカイブ
masatoto
2
190
ベイズ的方法に基づく統計的因果推論の基礎
holyshun
0
730
EBPMにおける生成AI活用について
daimoriwaki
0
260
The many faces of AI and the role of mathematics
gpeyre
1
1.6k
データサイエンティストをめぐる環境の違い 2024年版〈一般ビジネスパーソン調査の国際比較〉
datascientistsociety
PRO
0
980
「熊本県内バス・電車無料デー」の振り返りとその後の展開@土木計画学SS:成功失敗事例に学ぶ公共交通運賃設定
trafficbrain
0
140
Zipf 白色化:タイプとトークンの区別がもたらす良質な埋め込み空間と損失関数
eumesy
PRO
8
1.2k
Weekly AI Agents News! 11月号 論文のアーカイブ
masatoto
0
270
第79回 産総研人工知能セミナー 発表資料
agiats
3
190
한국어 오픈소스 거대 언어 모델의 가능성: 새로운 시대의 언어 이해와 생성
inureyes
PRO
0
170
Weekly AI Agents News! 9月号 論文のアーカイブ
masatoto
1
170
Featured
See All Featured
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
47
5.1k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
30
2.1k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
27
1.9k
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
28
2.2k
Unsuck your backbone
ammeep
669
57k
A designer walks into a library…
pauljervisheath
205
24k
How STYLIGHT went responsive
nonsquared
96
5.3k
Chrome DevTools: State of the Union 2024 - Debugging React & Beyond
addyosmani
3
260
Bootstrapping a Software Product
garrettdimon
PRO
305
110k
Product Roadmaps are Hard
iamctodd
PRO
50
11k
Distributed Sagas: A Protocol for Coordinating Microservices
caitiem20
330
21k
The Pragmatic Product Professional
lauravandoore
32
6.4k
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.