TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 7: Structured Prediction 1

Wielkość: px
Rozpocząć pokaz od strony:

Download "TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 7: Structured Prediction 1"

Transkrypt

1 TTIC 31210: Advanced Natural Language Processing Kevin Gimpel Spring 2019 Lecture 7: Structured Prediction 1 1

2 intro (1 lecture) Roadmap deep learning for NLP (5 lectures) structured prediction (4 lectures) introducing/formalizing structured prediction, categories of structures inference: dynamic programming, greedy algorithms, beam search inference with non-local features learning in structured prediction generative models, latent variables, unsupervised learning, variational autoencoders (2 lectures) Bayesian methods in NLP (2 lectures) Bayesian nonparametrics in NLP (2 lectures) review & other topics (1 lecture) 2

3 Assignments we will briefly go over Assignment 1 today Assignment 2 was posted last week, due May 1 st reminder: for those graduating this quarter, Assignment 5 is optional 3

4 What is Structured Prediction? 4

5 <latexit sha1_base64="u0xwf3yzhfhquldgyizfugbiywg=">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 hwqhfefztqvldvyfnx6mg6h3nwq9w8lzihvrs6tom8qw67tt24ghoc7b2vuiah67qetce4rvtcztu5r4ek3b+2ce81ru9ulvmdocp2sp117y9wzv/kfnxze+rybgmm9/4rzmj809k7i8v7u2o9nzhf39/45m/n38+8+bab9b+shz7bbt2p7vp1p6sha89xwtv/fpwv2/959z/3//f/v/f/+3932nog7caml+t9x7uv/n/igns0g==</latexit> <latexit sha1_base64="bn/zsaedvk1udtj0wpzbn/egj8m=">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</latexit> Classifiers a function from inputs x to outputs y one simple type of classifier: for any input x, assign a score to each output y, parameterized by parameters w: classify by choosing highest-scoring output: 5

6 Notation a vector entry i in the vector a matrix entry (i,j) in the matrix a structured object item i in the structured object 6

7 <latexit sha1_base64="u0xwf3yzhfhquldgyizfugbiywg=">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 hwqhfefztqvldvyfnx6mg6h3nwq9w8lzihvrs6tom8qw67tt24ghoc7b2vuiah67qetce4rvtcztu5r4ek3b+2ce81ru9ulvmdocp2sp117y9wzv/kfnxze+rybgmm9/4rzmj809k7i8v7u2o9nzhf39/45m/n38+8+bab9b+shz7bbt2p7vp1p6sha89xwtv/fpwv2/959z/3//f/v/f/+3932nog7caml+t9x7uv/n/igns0g==</latexit> <latexit sha1_base64="lvj5f1yayqj0pimo0ef8s222bnc=">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</latexit> Modeling, Inference, Learning inference: solve _ modeling: define score function learning: choose _ 7

8 Applications of our Classifier Framework task input (x) output (y) output space ( ) size of text classification a sentence gold standard label for x pre-defined, small label set (e.g., {positive, negative}) 2-10 word sense disambiguation instance of a particular word (e.g., bass) with its context gold standard word sense of target word pre-defined sense inventory from WordNet for bass 2-30 learning skipgram word embeddings instance of a word in a corpus a word in the context of x in a corpus vocabulary V part-of-speech tagging a sentence gold standard part-of-speech tags for x all possible part-ofspeech tag sequences with same length as x P x 8

9 Applications of our Classifier Framework task input (x) output (y) output space ( ) size of text classification word sense disambiguation learning skipgram word embeddings part-of-speech tagging a sentence instance of a particular word (e.g., bass) with its context instance of a word in a corpus a sentence gold standard label for x gold standard word sense of target word pre-defined, small label set (e.g., {positive, negative}) pre-defined sense inventory from WordNet for bass exponential in size of input! a word in the structured prediction context of x in vocabulary V a corpus gold standard part-of-speech tags for x all possible part-ofspeech tag sequences with same length as x P x 9

10 Applications of Classifier Framework (continued) task input (x) output (y) output space ( ) size of named entity recognition a sentence gold standard named entity labels for x (BIO tags) all possible BIO label sequences with same length as x P x constituency parsing a sentence gold standard constituent parse (labeled bracketing) of x all possible labeled bracketings of x exponential in length of x (Catalan number) dependency parsing a sentence gold standard dependency parse (labeled directed spanning tree) of x all possible labeled directed spanning trees of x exponential in length of x machine translation a sentence a translation of x all possible translations of x potentially infinite 10

11 <latexit sha1_base64="u0xwf3yzhfhquldgyizfugbiywg=">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 hwqhfefztqvldvyfnx6mg6h3nwq9w8lzihvrs6tom8qw67tt24ghoc7b2vuiah67qetce4rvtcztu5r4ek3b+2ce81ru9ulvmdocp2sp117y9wzv/kfnxze+rybgmm9/4rzmj809k7i8v7u2o9nzhf39/45m/n38+8+bab9b+shz7bbt2p7vp1p6sha89xwtv/fpwv2/959z/3//f/v/f/+3932nog7caml+t9x7uv/n/igns0g==</latexit> <latexit sha1_base64="lvj5f1yayqj0pimo0ef8s222bnc=">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</latexit> Modeling, Inference, Learning inference: solve _ modeling: define score function learning: choose _ Working definition of structured prediction: size of output space is exponential in size of input or is unbounded (e.g., machine translation) (we can t just enumerate all possible outputs) 11

12 What is Structured Prediction? however, just because the output is a structured object does not necessarily mean we are doing structured prediction we can model many structured output spaces with traditional local or unstructured predictors today we will aim to make this more formal in short, we may be predicting structures but we might not necessarily be using a structured predictor 12

13 Example NLP Tasks we ll go through some examples of NLP tasks that involve predicting output structures 13

14 Sequence Labeling (e.g., Part-of-Speech Tagging) determiner verb (past) prep. proper proper poss. adj. noun determiner verb (past) prep. noun noun poss. adj. noun Some questioned if Tim Cook s first product modal verb det. adjective noun prep. proper punc. modal verb det. adjective noun prep. noun punc. would be a breakaway hit for Apple. 14

15 Unlabeled Segmentations (Chinese Word Segmentation) some languages are written without whitespace task: insert spaces to form words 莎拉波娃现在居住在美国东南部的佛罗里达 莎拉波娃现在居住在美国东南部的佛罗里达 Sharapova now lives in US southeastern Florida J&M/SLP3

16 Labeled Segmentations (Named Entity Recognition) Some questioned if Tim Cook s first product would be a breakaway hit for Apple. PERSON ORGANIZATION 16

17 Labeled Segmentations (Entity Linking) Some questioned if Tim Cook s first product would be a breakaway hit for Apple. 17

18 Labeled Segmentation as Sequence Labeling O O O B-PERSON I-PERSON O O O Some questioned if Tim Cook s first product O O O O O O B-ORGANIZATION O would be a breakaway hit for Apple. B = begin I = inside O = outside 18

19 Trees (Constituency Parsing) (S (NP the man) (VP walked (PP to (NP the park)))) S VP NP PP NP DT NN VBD IN DT NN the man walked to the park Key: S = sentence NP = noun phrase VP = verb phrase PP = prepositional phrase DT = determiner NN = noun VBD = verb (past tense) IN = preposition 19

20 Unlabeled Segmentation + Clustering (Coreference Resolution)

21 Generation there are many language generation tasks that involve predicting a phrase, sentence, document, or some other textual sequence 21

22 Answers (Question Answering)

23 Sentences (Machine Translation)

24 Summaries (Summarization) The Apple Watch has drawbacks. There are other smartwatches that offer more capabilities. 24

25 Structured Prediction what is and is not structured prediction? we use the term structured prediction when: we use a structured score function or a structured loss function a structured score/loss function does not decompose across minimal parts of output to apply this definition we need to define parts and minimal parts 25

26 parts: each part is a subcomponent of entire input/output pair e.g., a single word and its associated POS tag for POS tagging or a sequence of two words and their POS tags or a sequence of two POS tags determiner verb (past) prep. proper proper poss. adj. noun determiner verb (past) prep. noun noun poss. adj. noun Some questioned if Tim Cook s first product modal verb det. adjective noun prep. proper punc. modal verb det. adjective noun prep. noun punc. would be a breakaway hit for Apple. 26

27 parts: each part is a subcomponent of entire input/output pair parts function = decomposition of input/output pair into a set of parts parts functions defined for score/loss function, rather than for task (many parts functions possible for a task) parts may overlap determiner verb (past) prep. proper proper poss. adj. noun determiner verb (past) prep. noun noun poss. adj. noun Some questioned if Tim Cook s first product modal verb det. adjective noun prep. proper punc. modal verb det. adjective noun prep. noun punc. would be a breakaway hit for Apple. 27

28 parts: each part is a subcomponent of entire input/output pair parts function = decomposition of input/output pair into a set of parts parts functions defined for score/loss function, rather than for task (many parts functions possible for a task) parts may overlap minimal parts: smallest possible parts for the task minimal parts function defined for task (structured output space), not for structured score/loss function minimal parts are non-overlapping 28

29 determiner verb (past) prep. proper proper poss. adj. noun determiner verb (past) prep. noun noun poss. adj. noun Some questioned if Tim Cook s first product modal verb det. adjective noun prep. proper punc. modal verb det. adjective noun prep. noun punc. would be a breakaway hit for Apple. minimal parts: smallest possible parts for the task minimal parts function defined for task (structured output space), not for structured score/loss function minimal parts are non-overlapping 29

30 Categories of Structured Prediction Problems multi-label classification: each input can be labeled with multiple labels e.g., document classification where each document can have multiple labels 30

31 Categories of Structured Prediction Problems multi-label classification in NLP:

32 Categories of Structured Prediction Problems 1 Coarse genre 1.1 Company Business, Strategy, etc. (elaborate in Section 3 [Topics]) 1.2 Purely multi-label Personal classification in NLP: 1.3 Personal but in professional context (e.g., it was good working with you) 1.4 Logistic Arrangements (meeting scheduling, technical support, etc) 4 Emotional tone (if not neutral) 4.1 jubilation 4.2 hope / anticipation 4.3 humor 4.4 camaraderie 4.5 admiration 4.6 gratitude 4.7 friendship / affection

33 Categories of Structured Prediction Problems multi-label classification: each input can be labeled with multiple labels if there are N possible labels, output space has size? (Q1 on handout) 33

34 Categories of Structured Prediction Problems multi-label classification: each input can be labeled with multiple labels if there are N possible labels, output space has size 2 N what are the minimal parts? (Q2 on handout) 34

35 parts: each part is a subcomponent of entire input/output pair parts function = decomposition of input/output pair into a set of parts parts functions defined for score/loss function, rather than for task (many parts functions possible for a task) parts may overlap minimal parts: smallest possible parts for the task minimal parts function defined for task (structured output space), not for structured score/loss function minimal parts are non-overlapping 35

36 <latexit sha1_base64="cs2rlstvqqvviaakpmv2nv3uyg0=">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 rcmw7/y+vxw862zrtdb0z33d/7tzr/f+y8h//ngvx7894p/mddvfdbr/nxw6ofb//0/vhkhjq==</latexit> Categories of Structured Prediction Problems multi-label classification: each input can be labeled with multiple labels if there are N possible labels, output space has size 2 N what are the minimal parts? individual labels the mp(y) function defines the set of minimal parts of the structured output y 36

37 <latexit sha1_base64="cs2rlstvqqvviaakpmv2nv3uyg0=">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 rcmw7/y+vxw862zrtdb0z33d/7tzr/f+y8h//ngvx7894p/mddvfdbr/nxw6ofb//0/vhkhjq==</latexit> minimal parts: Multi-Label Classification if score & loss functions factor across minimal parts, then we are not doing structured prediction e.g., we could build N binary classifiers, one for each label, and use them to independently predict each label for each input this would not be considered structured prediction 37

TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 8: Structured PredicCon 2

TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 8: Structured PredicCon 2 TTIC 31210: Advanced Natural Language Processing Kevin Gimpel Spring 2019 Lecture 8: Structured PredicCon 2 1 Roadmap intro (1 lecture) deep learning for NLP (5 lectures) structured predic+on (4 lectures)

Bardziej szczegółowo

TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 9: Inference in Structured Prediction

TTIC 31210: Advanced Natural Language Processing. Kevin Gimpel Spring Lecture 9: Inference in Structured Prediction TTIC 31210: Advanced Natural Language Processing Kevin Gimpel Spring 2019 Lecture 9: Inference in Structured Prediction 1 intro (1 lecture) Roadmap deep learning for NLP (5 lectures) structured prediction

Bardziej szczegółowo

deep learning for NLP (5 lectures)

deep learning for NLP (5 lectures) TTIC 31210: Advanced Natural Language Processing Kevin Gimpel Spring 2019 Lecture 6: Finish Transformers; Sequence- to- Sequence Modeling and AJenKon 1 Roadmap intro (1 lecture) deep learning for NLP (5

Bardziej szczegółowo

TTIC 31190: Natural Language Processing

TTIC 31190: Natural Language Processing TTIC 31190: Natural Language Processing Kevin Gimpel Spring 2018 Lecture 17: Machine TranslaDon; SemanDcs Roadmap words, morphology, lexical semandcs text classificadon simple neural methods for NLP language

Bardziej szczegółowo

Previously on CSCI 4622

Previously on CSCI 4622 More Naïve Bayes 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

Bardziej szczegółowo

Linear Classification and Logistic Regression. Pascal Fua IC-CVLab

Linear Classification and Logistic Regression. Pascal Fua IC-CVLab Linear Classification and Logistic Regression Pascal Fua IC-CVLab 1 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

Bardziej szczegółowo

tum.de/fall2018/ in2357

tum.de/fall2018/ in2357 https://piazza.com/ tum.de/fall2018/ in2357 Prof. Daniel Cremers From to Classification Categories of Learning (Rep.) Learning Unsupervised Learning clustering, density estimation Supervised Learning learning

Bardziej szczegółowo

Hard-Margin Support Vector Machines

Hard-Margin Support Vector Machines Hard-Margin Support Vector Machines aaacaxicbzdlssnafiyn9vbjlepk3ay2gicupasvu4iblxuaw2hjmuwn7ddjjmxm1bkcg1/fjqsvt76fo9/gazqfvn8y+pjpozw5vx8zkpvtfxmlhcwl5zxyqrm2vrg5zw3vxmsoezi4ogkr6phieky5crvvjhriqvdom9l2xxftevuwcekj3lktmhghgniauiyutvrwxtvme34a77kbvg73gtygpjsrfati1+xc8c84bvraowbf+uwnipyehcvmkjrdx46vlykhkgykm3ujjdhcyzqkxy0chur6ax5cbg+1m4bbjptjcubuz4kuhvjoql93hkin5hxtav5x6yyqopnsyuneey5ni4keqrxbar5wqaxbik00icyo/iveiyqqvjo1u4fgzj/8f9x67bzmxnurjzmijtlybwfgcdjgfdtajwgcf2dwaj7ac3g1ho1n4814n7wwjgjmf/ys8fenfycuzq==

Bardziej szczegółowo

Machine Learning for Data Science (CS4786) Lecture11. Random Projections & Canonical Correlation Analysis

Machine Learning for Data Science (CS4786) Lecture11. Random Projections & Canonical Correlation Analysis Machine Learning for Data Science (CS4786) Lecture11 5 Random Projections & Canonical Correlation Analysis The Tall, THE FAT AND THE UGLY n X d The Tall, THE FAT AND THE UGLY d X > n X d n = n d d The

Bardziej szczegółowo

Helena Boguta, klasa 8W, rok szkolny 2018/2019

Helena Boguta, klasa 8W, rok szkolny 2018/2019 Poniższy zbiór zadań został wykonany w ramach projektu Mazowiecki program stypendialny dla uczniów szczególnie uzdolnionych - najlepsza inwestycja w człowieka w roku szkolnym 2018/2019. Składają się na

Bardziej szczegółowo

Machine Learning for Data Science (CS4786) Lecture 11. Spectral Embedding + Clustering

Machine Learning for Data Science (CS4786) Lecture 11. Spectral Embedding + Clustering Machine Learning for Data Science (CS4786) Lecture 11 Spectral Embedding + Clustering MOTIVATING EXAMPLE What can you say from this network? MOTIVATING EXAMPLE How about now? THOUGHT EXPERIMENT For each

Bardziej szczegółowo

Analysis of Movie Profitability STAT 469 IN CLASS ANALYSIS #2

Analysis of Movie Profitability STAT 469 IN CLASS ANALYSIS #2 Analysis of Movie Profitability STAT 469 IN CLASS ANALYSIS #2 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

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

Tychy, plan miasta: Skala 1: (Polish Edition)

Tychy, plan miasta: Skala 1: (Polish Edition) Tychy, plan miasta: Skala 1:20 000 (Polish Edition) Poland) Przedsiebiorstwo Geodezyjno-Kartograficzne (Katowice Click here if your download doesn"t start automatically Tychy, plan miasta: Skala 1:20 000

Bardziej szczegółowo

Machine Learning for Data Science (CS4786) Lecture 24. Differential Privacy and Re-useable Holdout

Machine Learning for Data Science (CS4786) Lecture 24. Differential Privacy and Re-useable Holdout Machine Learning for Data Science (CS4786) Lecture 24 Differential Privacy and Re-useable Holdout Defining Privacy Defining Privacy Dataset + Defining Privacy Dataset + Learning Algorithm Distribution

Bardziej szczegółowo

SSW1.1, HFW Fry #20, Zeno #25 Benchmark: Qtr.1. Fry #65, Zeno #67. like

SSW1.1, HFW Fry #20, Zeno #25 Benchmark: Qtr.1. Fry #65, Zeno #67. like SSW1.1, HFW Fry #20, Zeno #25 Benchmark: Qtr.1 I SSW1.1, HFW Fry #65, Zeno #67 Benchmark: Qtr.1 like SSW1.2, HFW Fry #47, Zeno #59 Benchmark: Qtr.1 do SSW1.2, HFW Fry #5, Zeno #4 Benchmark: Qtr.1 to SSW1.2,

Bardziej szczegółowo

www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C, Part II If "Yes," complete Schedule C, Part

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

ERASMUS + : Trail of extinct and active volcanoes, earthquakes through Europe. SURVEY TO STUDENTS.

ERASMUS + : Trail of extinct and active volcanoes, earthquakes through Europe. SURVEY TO STUDENTS. ERASMUS + : Trail of extinct and active volcanoes, earthquakes through Europe. SURVEY TO STUDENTS. Strona 1 1. Please give one answer. I am: Students involved in project 69% 18 Student not involved in

Bardziej szczegółowo

Revenue Maximization. Sept. 25, 2018

Revenue Maximization. Sept. 25, 2018 Revenue Maximization Sept. 25, 2018 Goal So Far: Ideal Auctions Dominant-Strategy Incentive Compatible (DSIC) b i = v i is a dominant strategy u i 0 x is welfare-maximizing x and p run in polynomial time

Bardziej szczegółowo

Zakopane, plan miasta: Skala ok. 1: = City map (Polish Edition)

Zakopane, plan miasta: Skala ok. 1: = City map (Polish Edition) Zakopane, plan miasta: Skala ok. 1:15 000 = City map (Polish Edition) Click here if your download doesn"t start automatically Zakopane, plan miasta: Skala ok. 1:15 000 = City map (Polish Edition) Zakopane,

Bardziej szczegółowo

OpenPoland.net API Documentation

OpenPoland.net API Documentation OpenPoland.net API Documentation Release 1.0 Michał Gryczka July 11, 2014 Contents 1 REST API tokens: 3 1.1 How to get a token............................................ 3 2 REST API : search for assets

Bardziej szczegółowo

ARNOLD. EDUKACJA KULTURYSTY (POLSKA WERSJA JEZYKOWA) BY DOUGLAS KENT HALL

ARNOLD. EDUKACJA KULTURYSTY (POLSKA WERSJA JEZYKOWA) BY DOUGLAS KENT HALL Read Online and Download Ebook ARNOLD. EDUKACJA KULTURYSTY (POLSKA WERSJA JEZYKOWA) BY DOUGLAS KENT HALL DOWNLOAD EBOOK : ARNOLD. EDUKACJA KULTURYSTY (POLSKA WERSJA Click link bellow and free register

Bardziej szczegółowo

www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C, Part II If "Yes," complete Schedule C, Part

Bardziej szczegółowo

Presented by. Dr. Morten Middelfart, CTO

Presented by. Dr. Morten Middelfart, CTO Meeting Big Data challenges in Leadership with Human-Computer Synergy. Presented by Dr. Morten Middelfart, CTO Big Data Data that exists in such large amounts or in such unstructured form that it is difficult

Bardziej szczegółowo

Gradient Coding using the Stochastic Block Model

Gradient Coding using the Stochastic Block Model Gradient Coding using the Stochastic Block Model Zachary Charles (UW-Madison) Joint work with Dimitris Papailiopoulos (UW-Madison) aaacaxicbvdlssnafj3uv62vqbvbzwarxjsqikaboelgzux7gcaeywtsdp1mwsxeaepd+ctuxcji1r9w5984bbpq1gmxdufcy733bcmjutn2t1fawl5zxsuvvzy2t7z3zn29lkwyguktjywrnqbjwigntuuvi51uebqhjlsdwfxebz8qiwnc79uwjv6mepxgfcoljd88uiox0m1hvlnzwzgowymjn7tjyzertmvpareju5aqkndwzs83thawe64wq1j2httvxo6eopirccxnjekrhqae6wrkuuykl08/gmnjryqwsoqurubu/t2ro1jkyrzozhipvpz3juj/xjdt0ywxu55mina8wxrldkoetukairuekzbubgfb9a0q95fawonqkjoez/7lrdi6trzbcm7pqvwrio4yoarh4aq44bzuwq1ogcba4be8g1fwzjwzl8a78tfrlrnfzd74a+pzb2h+lzm=

Bardziej szczegółowo

Emilka szuka swojej gwiazdy / Emily Climbs (Emily, #2)

Emilka szuka swojej gwiazdy / Emily Climbs (Emily, #2) Emilka szuka swojej gwiazdy / Emily Climbs (Emily, #2) Click here if your download doesn"t start automatically Emilka szuka swojej gwiazdy / Emily Climbs (Emily, #2) Emilka szuka swojej gwiazdy / Emily

Bardziej szczegółowo

Katowice, plan miasta: Skala 1: = City map = Stadtplan (Polish Edition)

Katowice, plan miasta: Skala 1: = City map = Stadtplan (Polish Edition) Katowice, plan miasta: Skala 1:20 000 = City map = Stadtplan (Polish Edition) Polskie Przedsiebiorstwo Wydawnictw Kartograficznych im. Eugeniusza Romera Click here if your download doesn"t start automatically

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

Stargard Szczecinski i okolice (Polish Edition)

Stargard Szczecinski i okolice (Polish Edition) Stargard Szczecinski i okolice (Polish Edition) Janusz Leszek Jurkiewicz Click here if your download doesn"t start automatically Stargard Szczecinski i okolice (Polish Edition) Janusz Leszek Jurkiewicz

Bardziej szczegółowo

Realizacja systemów wbudowanych (embeded systems) w strukturach PSoC (Programmable System on Chip)

Realizacja systemów wbudowanych (embeded systems) w strukturach PSoC (Programmable System on Chip) Realizacja systemów wbudowanych (embeded systems) w strukturach PSoC (Programmable System on Chip) Embeded systems Architektura układów PSoC (Cypress) Możliwości bloków cyfrowych i analogowych Narzędzia

Bardziej szczegółowo

Instrukcja konfiguracji usługi Wirtualnej Sieci Prywatnej w systemie Mac OSX

Instrukcja konfiguracji usługi Wirtualnej Sieci Prywatnej w systemie Mac OSX UNIWERSYTETU BIBLIOTEKA IEGO UNIWERSYTETU IEGO Instrukcja konfiguracji usługi Wirtualnej Sieci Prywatnej w systemie Mac OSX 1. Make a new connection Open the System Preferences by going to the Apple menu

Bardziej szczegółowo

MaPlan Sp. z O.O. Click here if your download doesn"t start automatically

MaPlan Sp. z O.O. Click here if your download doesnt start automatically Mierzeja Wislana, mapa turystyczna 1:50 000: Mikoszewo, Jantar, Stegna, Sztutowo, Katy Rybackie, Przebrno, Krynica Morska, Piaski, Frombork =... = Carte touristique (Polish Edition) MaPlan Sp. z O.O Click

Bardziej szczegółowo

Rozpoznawanie twarzy metodą PCA Michał Bereta 1. Testowanie statystycznej istotności różnic między jakością klasyfikatorów

Rozpoznawanie twarzy metodą PCA Michał Bereta   1. Testowanie statystycznej istotności różnic między jakością klasyfikatorów Rozpoznawanie twarzy metodą PCA Michał Bereta www.michalbereta.pl 1. Testowanie statystycznej istotności różnic między jakością klasyfikatorów Wiemy, że możemy porównywad klasyfikatory np. za pomocą kroswalidacji.

Bardziej szczegółowo

Egzamin maturalny z języka angielskiego na poziomie dwujęzycznym Rozmowa wstępna (wyłącznie dla egzaminującego)

Egzamin maturalny z języka angielskiego na poziomie dwujęzycznym Rozmowa wstępna (wyłącznie dla egzaminującego) 112 Informator o egzaminie maturalnym z języka angielskiego od roku szkolnego 2014/2015 2.6.4. Część ustna. Przykładowe zestawy zadań Przykładowe pytania do rozmowy wstępnej Rozmowa wstępna (wyłącznie

Bardziej szczegółowo

Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition)

Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition) Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition) J Krupski Click here if your download doesn"t start automatically Karpacz, plan miasta 1:10 000: Panorama

Bardziej szczegółowo

Pielgrzymka do Ojczyzny: Przemowienia i homilie Ojca Swietego Jana Pawla II (Jan Pawel II-- pierwszy Polak na Stolicy Piotrowej) (Polish Edition)

Pielgrzymka do Ojczyzny: Przemowienia i homilie Ojca Swietego Jana Pawla II (Jan Pawel II-- pierwszy Polak na Stolicy Piotrowej) (Polish Edition) Pielgrzymka do Ojczyzny: Przemowienia i homilie Ojca Swietego Jana Pawla II (Jan Pawel II-- pierwszy Polak na Stolicy Piotrowej) (Polish Edition) Click here if your download doesn"t start automatically

Bardziej szczegółowo

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO POZIOM ROZSZERZONY MAJ 2010 CZĘŚĆ I. Czas pracy: 120 minut. Liczba punktów do uzyskania: 23 WPISUJE ZDAJĄCY

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO POZIOM ROZSZERZONY MAJ 2010 CZĘŚĆ I. Czas pracy: 120 minut. Liczba punktów do uzyskania: 23 WPISUJE ZDAJĄCY Centralna Komisja Egzaminacyjna Arkusz zawiera informacje prawnie chronione do momentu rozpoczęcia egzaminu. Układ graficzny CKE 2010 KOD WPISUJE ZDAJĄCY PESEL Miejsce na naklejkę z kodem dysleksja EGZAMIN

Bardziej szczegółowo

Agnostic Learning and VC dimension

Agnostic Learning and VC dimension Agnostic Learning and VC dimension Machine Learning Spring 2019 The slides are based on Vivek Srikumar s 1 This Lecture Agnostic Learning What if I cannot guarantee zero training error? Can we still get

Bardziej szczegółowo

Lecture 18 Review for Exam 1

Lecture 18 Review for Exam 1 Spring, 2019 ME 323 Mechanics of Materials Lecture 18 Review for Exam 1 Reading assignment: HW1-HW5 News: Ready for the exam? Instructor: Prof. Marcial Gonzalez Announcements Exam 1 - Wednesday February

Bardziej szczegółowo

Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition)

Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition) Karpacz, plan miasta 1:10 000: Panorama Karkonoszy, mapa szlakow turystycznych (Polish Edition) J Krupski Click here if your download doesn"t start automatically Karpacz, plan miasta 1:10 000: Panorama

Bardziej szczegółowo

Jak zasada Pareto może pomóc Ci w nauce języków obcych?

Jak zasada Pareto może pomóc Ci w nauce języków obcych? Jak zasada Pareto może pomóc Ci w nauce języków obcych? Artykuł pobrano ze strony eioba.pl Pokazuje, jak zastosowanie zasady Pareto może usprawnić Twoją naukę angielskiego. Słynna zasada Pareto mówi o

Bardziej szczegółowo

Ankiety Nowe funkcje! Pomoc magda.szewczyk@slo-wroc.pl. magda.szewczyk@slo-wroc.pl. Twoje konto Wyloguj. BIODIVERSITY OF RIVERS: Survey to students

Ankiety Nowe funkcje! Pomoc magda.szewczyk@slo-wroc.pl. magda.szewczyk@slo-wroc.pl. Twoje konto Wyloguj. BIODIVERSITY OF RIVERS: Survey to students Ankiety Nowe funkcje! Pomoc magda.szewczyk@slo-wroc.pl Back Twoje konto Wyloguj magda.szewczyk@slo-wroc.pl BIODIVERSITY OF RIVERS: Survey to students Tworzenie ankiety Udostępnianie Analiza (55) Wyniki

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

Mixed-integer Convex Representability

Mixed-integer Convex Representability Mixed-integer Convex Representability Juan Pablo Vielma Massachuse=s Ins?tute of Technology Joint work with Miles Lubin and Ilias Zadik INFORMS Annual Mee?ng, Phoenix, AZ, November, 2018. Mixed-Integer

Bardziej szczegółowo

Zarządzanie sieciami telekomunikacyjnymi

Zarządzanie sieciami telekomunikacyjnymi SNMP Protocol The Simple Network Management Protocol (SNMP) is an application layer protocol that facilitates the exchange of management information between network devices. It is part of the Transmission

Bardziej szczegółowo

Instrukcja obsługi User s manual

Instrukcja obsługi User s manual Instrukcja obsługi User s manual Konfigurator Lanberg Lanberg Configurator E-mail: support@lanberg.pl support@lanberg.eu www.lanberg.pl www.lanberg.eu Lanberg 2015-2018 WERSJA VERSION: 2018/11 Instrukcja

Bardziej szczegółowo

Dolny Slask 1: , mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition)

Dolny Slask 1: , mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition) Dolny Slask 1:300 000, mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition) Click here if your download doesn"t start automatically Dolny Slask 1:300 000, mapa turystyczno-samochodowa: Plan Wroclawia

Bardziej szczegółowo

Wroclaw, plan nowy: Nowe ulice, 1:22500, sygnalizacja swietlna, wysokosc wiaduktow : Debica = City plan (Polish Edition)

Wroclaw, plan nowy: Nowe ulice, 1:22500, sygnalizacja swietlna, wysokosc wiaduktow : Debica = City plan (Polish Edition) Wroclaw, plan nowy: Nowe ulice, 1:22500, sygnalizacja swietlna, wysokosc wiaduktow : Debica = City plan (Polish Edition) Wydawnictwo "Demart" s.c Click here if your download doesn"t start automatically

Bardziej szczegółowo

Few-fermion thermometry

Few-fermion thermometry Few-fermion thermometry Phys. Rev. A 97, 063619 (2018) Tomasz Sowiński Institute of Physics of the Polish Academy of Sciences Co-authors: Marcin Płodzień Rafał Demkowicz-Dobrzański FEW-BODY PROBLEMS FewBody.ifpan.edu.pl

Bardziej szczegółowo

aforementioned device she also has to estimate the time when the patients need the infusion to be replaced and/or disconnected. Meanwhile, however, she must cope with many other tasks. If the department

Bardziej szczegółowo

POLITYKA PRYWATNOŚCI / PRIVACY POLICY

POLITYKA PRYWATNOŚCI / PRIVACY POLICY POLITYKA PRYWATNOŚCI / PRIVACY POLICY TeleTrade DJ International Consulting Ltd Sierpień 2013 2011-2014 TeleTrade-DJ International Consulting Ltd. 1 Polityka Prywatności Privacy Policy Niniejsza Polityka

Bardziej szczegółowo

Leba, Rowy, Ustka, Slowinski Park Narodowy, plany miast, mapa turystyczna =: Tourist map = Touristenkarte (Polish Edition)

Leba, Rowy, Ustka, Slowinski Park Narodowy, plany miast, mapa turystyczna =: Tourist map = Touristenkarte (Polish Edition) Leba, Rowy, Ustka, Slowinski Park Narodowy, plany miast, mapa turystyczna =: Tourist map = Touristenkarte (Polish Edition) FotKart s.c Click here if your download doesn"t start automatically Leba, Rowy,

Bardziej szczegółowo

Convolution semigroups with linear Jacobi parameters

Convolution semigroups with linear Jacobi parameters Convolution semigroups with linear Jacobi parameters Michael Anshelevich; Wojciech Młotkowski Texas A&M University; University of Wrocław February 14, 2011 Jacobi parameters. µ = measure with finite moments,

Bardziej szczegółowo

Rev Źródło:

Rev Źródło: KamPROG for AVR Rev. 20190119192125 Źródło: http://wiki.kamamilabs.com/index.php/kamprog_for_avr Spis treści Introdcution... 1 Features... 2 Standard equipment... 4 Installation... 5 Software... 6 AVR

Bardziej szczegółowo

!850016! www.irs.gov/form8879eo. e-file www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C,

Bardziej szczegółowo

PLSH1 (JUN14PLSH101) General Certificate of Education Advanced Subsidiary Examination June 2014. Reading and Writing TOTAL

PLSH1 (JUN14PLSH101) General Certificate of Education Advanced Subsidiary Examination June 2014. Reading and Writing TOTAL Centre Number Surname Candidate Number For Examiner s Use Other Names Candidate Signature Examiner s Initials Section Mark Polish Unit 1 Reading and Writing General Certificate of Education Advanced Subsidiary

Bardziej szczegółowo

Surname. Other Names. For Examiner s Use Centre Number. Candidate Number. Candidate Signature

Surname. Other Names. For Examiner s Use Centre Number. Candidate Number. Candidate Signature A Surname _ Other Names For Examiner s Use Centre Number Candidate Number Candidate Signature Polish Unit 1 PLSH1 General Certificate of Education Advanced Subsidiary Examination June 2014 Reading and

Bardziej szczegółowo

Weronika Mysliwiec, klasa 8W, rok szkolny 2018/2019

Weronika Mysliwiec, klasa 8W, rok szkolny 2018/2019 Poniższy zbiór zadań został wykonany w ramach projektu Mazowiecki program stypendialny dla uczniów szczególnie uzdolnionych - najlepsza inwestycja w człowieka w roku szkolnym 2018/2019. Tresci zadań rozwiązanych

Bardziej szczegółowo

SNP SNP Business Partner Data Checker. Prezentacja produktu

SNP SNP Business Partner Data Checker. Prezentacja produktu SNP SNP Business Partner Data Checker Prezentacja produktu Istota rozwiązania SNP SNP Business Partner Data Checker Celem produktu SNP SNP Business Partner Data Checker jest umożliwienie sprawdzania nazwy

Bardziej szczegółowo

Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition)

Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition) Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition) Piotr Maluskiewicz Click here if your download doesn"t start automatically Miedzy

Bardziej szczegółowo

Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition)

Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition) Miedzy legenda a historia: Szlakiem piastowskim z Poznania do Gniezna (Biblioteka Kroniki Wielkopolski) (Polish Edition) Piotr Maluskiewicz Click here if your download doesn"t start automatically Miedzy

Bardziej szczegółowo

Steps to build a business Examples: Qualix Comergent

Steps to build a business Examples: Qualix Comergent How To Start a BUSINESS Agenda Steps to build a business Examples: Qualix Comergent 1 Idea The Idea is a Piece of a Company 4 2 The Idea is a Piece of a Company Investing_in_New_Ideas.wmv Finding_the_Problem_is_the_Hard_Part_Kevin

Bardziej szczegółowo

www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C, Part II If "Yes," complete Schedule C, Part

Bardziej szczegółowo

Learning about Language with Normalizing Flows Graham Neubig Language Technologies Institute, Carnegie Mellon University

Learning about Language with Normalizing Flows Graham Neubig Language Technologies Institute, Carnegie Mellon University Learning about Language with Normalizing Flows Graham Neubig Language Technologies Institute, Carnegie Mellon University Chunting Zhou Junxian He Di Wang, Xuezhe Ma, Daniel Spokoyny, Taylor Berg-Kirkpatrick

Bardziej szczegółowo

www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C, Part II If "Yes," complete Schedule C, Part

Bardziej szczegółowo

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO POZIOM ROZSZERZONY MAJ 2010 CZĘŚĆ I. Czas pracy: 120 minut. Liczba punktów do uzyskania: 23 WPISUJE ZDAJĄCY

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO POZIOM ROZSZERZONY MAJ 2010 CZĘŚĆ I. Czas pracy: 120 minut. Liczba punktów do uzyskania: 23 WPISUJE ZDAJĄCY Centralna Komisja Egzaminacyjna Arkusz zawiera informacje prawnie chronione do momentu rozpoczęcia egzaminu. Układ graficzny CKE 2010 KOD WPISUJE ZDAJĄCY PESEL Miejsce na naklejkę z kodem dysleksja EGZAMIN

Bardziej szczegółowo

JĘZYK ANGIELSKI ĆWICZENIA ORAZ REPETYTORIUM GRAMATYCZNE

JĘZYK ANGIELSKI ĆWICZENIA ORAZ REPETYTORIUM GRAMATYCZNE MACIEJ MATASEK JĘZYK ANGIELSKI ĆWICZENIA ORAZ REPETYTORIUM GRAMATYCZNE 1 Copyright by Wydawnictwo HANDYBOOKS Poznań 2014 Wszelkie prawa zastrzeżone. Każda reprodukcja lub adaptacja całości bądź części

Bardziej szczegółowo

Bardzo formalny, odbiorca posiada specjalny tytuł, który jest używany zamiast nazwiska

Bardzo formalny, odbiorca posiada specjalny tytuł, który jest używany zamiast nazwiska - Wstęp Dear Mr. President, Dear Mr. President, Bardzo formalny, odbiorca posiada specjalny tytuł, który jest używany zamiast nazwiska Dear Sir, Dear Sir, Formalny, odbiorcą jest mężczyzna, którego nazwiska

Bardziej szczegółowo

y = The Chain Rule Show all work. No calculator unless otherwise stated. If asked to Explain your answer, write in complete sentences.

y = The Chain Rule Show all work. No calculator unless otherwise stated. If asked to Explain your answer, write in complete sentences. The Chain Rule Show all work. No calculator unless otherwise stated. If asked to Eplain your answer, write in complete sentences. 1. Find the derivative of the functions y 7 (b) (a) ( ) y t 1 + t 1 (c)

Bardziej szczegółowo

www.irs.gov/form990. If "Yes," complete Schedule A Schedule B, Schedule of Contributors If "Yes," complete Schedule C, Part I If "Yes," complete Schedule C, Part II If "Yes," complete Schedule C, Part

Bardziej szczegółowo

The Overview of Civilian Applications of Airborne SAR Systems

The Overview of Civilian Applications of Airborne SAR Systems The Overview of Civilian Applications of Airborne SAR Systems Maciej Smolarczyk, Piotr Samczyński Andrzej Gadoś, Maj Mordzonek Research and Development Department of PIT S.A. PART I WHAT DOES SAR MEAN?

Bardziej szczegółowo

Supervised Hierarchical Clustering with Exponential Linkage. Nishant Yadav

Supervised Hierarchical Clustering with Exponential Linkage. Nishant Yadav Supervised Hierarchical Clustering with Exponential Linage Nishant Yadav Ari Kobren Nicholas Monath Andrew McCallum At train time, learn A :2 X! Y Supervised Clustering aaab8nicbvdlssnafl2pr1pfvzdugvwvrirdfl147kcfuabymq6aydozslmjvbcp8onc0xc+jxu/bsnbrbaemdgcm69zlntaq36hnftmltfwnzq7xd2dnd2z+ohh61juo1zs2qhnldbgmugqt5chyn9gmxkfgnxbyl/udj6ynv/irpwlyjkspokuojv6/zjgmbkr3cwg1zpx9+zwv4lfbouaa6qx/2homnmjfjbjon5xojbrjryktis08nswidbhrwspjzeyqzspp3dordn1iafsunp190zgymomcwgn84hm2cvf/7xeitf1hgzpmgxxwupcjf5eb3u0ouguuxtyrqzw1wl46jjhrtsxvbgr988ippx9r9yx8ua43boo4ynmapnimpv9cae2hccygoeizxehpqexheny/fampdo7hd5zph3bqvc=

Bardziej szczegółowo

Compatible cameras for NVR-5000 series Main Stream Sub stream Support Firmware ver. 0,2-1Mbit yes yes yes n/d

Compatible cameras for NVR-5000 series Main Stream Sub stream Support Firmware ver. 0,2-1Mbit yes yes yes n/d NOVUS IP CAMERAS CLASSIC CAMERAS Compatible cameras for NVR-5000 series Main Stream Sub stream Support Firmware ver. Resolution Bitrate FPS GOP Resolution Bitrate FPS GOP Audio Motion detection NVIP 5000

Bardziej szczegółowo

Klaps za karę. Wyniki badania dotyczącego postaw i stosowania kar fizycznych. Joanna Włodarczyk

Klaps za karę. Wyniki badania dotyczącego postaw i stosowania kar fizycznych. Joanna Włodarczyk Klaps za karę Wyniki badania dotyczącego postaw i stosowania kar fizycznych Joanna Włodarczyk joanna.wlodarczyk@fdds.pl Warszawa, 1.12.2017 Fundacja Dajemy Dzieciom Siłę, 2017 Informacje o badaniu Badanie

Bardziej szczegółowo

Neural Networks (The Machine-Learning Kind) BCS 247 March 2019

Neural Networks (The Machine-Learning Kind) BCS 247 March 2019 Neural Networks (The Machine-Learning Kind) BCS 247 March 2019 Neurons http://biomedicalengineering.yolasite.com/neurons.php Networks https://en.wikipedia.org/wiki/network_theory#/media/file:social_network_analysis_visualization.png

Bardziej szczegółowo

EPS. Erasmus Policy Statement

EPS. Erasmus Policy Statement Wyższa Szkoła Biznesu i Przedsiębiorczości Ostrowiec Świętokrzyski College of Business and Entrepreneurship EPS Erasmus Policy Statement Deklaracja Polityki Erasmusa 2014-2020 EN The institution is located

Bardziej szczegółowo

Arrays -II. Arrays. Outline ECE Cal Poly Pomona Electrical & Computer Engineering. Introduction

Arrays -II. Arrays. Outline ECE Cal Poly Pomona Electrical & Computer Engineering. Introduction ECE 114-9 Arrays -II Dr. Z. Aliyazicioglu Electrical & Computer Engineering Electrical & Computer Engineering 1 Outline Introduction Arrays Declaring and Allocation Arrays Examples Using Arrays Passing

Bardziej szczegółowo

No matter how much you have, it matters how much you need

No matter how much you have, it matters how much you need CSR STRATEGY KANCELARIA FINANSOWA TRITUM GROUP SP. Z O.O. No matter how much you have, it matters how much you need Kancelaria Finansowa Tritum Group Sp. z o.o. was established in 2007 we build trust among

Bardziej szczegółowo

Rachunek lambda, zima

Rachunek lambda, zima Rachunek lambda, zima 2015-16 Wykład 2 12 października 2015 Tydzień temu: Własność Churcha-Rossera (CR) Jeśli a b i a c, to istnieje takie d, że b d i c d. Tydzień temu: Własność Churcha-Rossera (CR) Jeśli

Bardziej szczegółowo

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO

EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO Miejsce na naklejkę z kodem szkoły dysleksja MJA-R1_1P-072 EGZAMIN MATURALNY Z JĘZYKA ANGIELSKIEGO MAJ ROK 2007 Instrukcja dla zdającego POZIOM ROZSZERZONY CZĘŚĆ I Czas pracy 120 minut 1. Sprawdź, czy

Bardziej szczegółowo

MoA-Net: Self-supervised Motion Segmentation. Pia Bideau, Rakesh R Menon, Erik Learned-Miller

MoA-Net: Self-supervised Motion Segmentation. Pia Bideau, Rakesh R Menon, Erik Learned-Miller MoA-Net: Self-supervised Motion Segmentation Pia Bideau, Rakesh R Menon, Erik Learned-Miller University of Massachusetts Amherst College of Information and Computer Science Motion Segmentation P Bideau,

Bardziej szczegółowo

LEARNING AGREEMENT FOR STUDIES

LEARNING AGREEMENT FOR STUDIES LEARNING AGREEMENT FOR STUDIES The Student First and last name(s) Nationality E-mail Academic year 2014/2015 Study period 1 st semester 2 nd semester Study cycle Bachelor Master Doctoral Subject area,

Bardziej szczegółowo

KONSPEKT DO LEKCJI MATEMATYKI W KLASIE 3 POLO/ A LAYER FOR CLASS 3 POLO MATHEMATICS

KONSPEKT DO LEKCJI MATEMATYKI W KLASIE 3 POLO/ A LAYER FOR CLASS 3 POLO MATHEMATICS KONSPEKT DO LEKCJI MATEMATYKI W KLASIE 3 POLO/ A LAYER FOR CLASS 3 POLO MATHEMATICS Temat: Funkcja logarytmiczna (i wykładnicza)/ Logarithmic (and exponential) function Typ lekcji: Lekcja ćwiczeniowa/training

Bardziej szczegółowo

HAPPY ANIMALS L01 HAPPY ANIMALS L03 HAPPY ANIMALS L05 HAPPY ANIMALS L07

HAPPY ANIMALS L01 HAPPY ANIMALS L03 HAPPY ANIMALS L05 HAPPY ANIMALS L07 HAPPY ANIMALS L0 HAPPY ANIMALS L0 HAPPY ANIMALS L0 HAPPY ANIMALS L07 INSTRUKCJA MONTAŻU ASSEMBLY INSTRUCTIONS Akcesoria / Fittings K ZW W8 W7 Ø x 6 szt. / pcs Ø7 x 70 Narzędzia / Tools DO MONTAŻU POTRZEBNE

Bardziej szczegółowo

HAPPY ANIMALS L02 HAPPY ANIMALS L04 HAPPY ANIMALS L06 HAPPY ANIMALS L08

HAPPY ANIMALS L02 HAPPY ANIMALS L04 HAPPY ANIMALS L06 HAPPY ANIMALS L08 HAPPY ANIMALS L02 HAPPY ANIMALS L04 HAPPY ANIMALS L06 HAPPY ANIMALS L08 INSTRUKCJA MONTAŻU ASSEMBLY INSTRUCTIONS Akcesoria / Fittings K O G ZW W8 W4 20 szt. / pcs 4 szt. / pcs 4 szt. / pcs 4 szt. / pcs

Bardziej szczegółowo

Poland) Wydawnictwo "Gea" (Warsaw. Click here if your download doesn"t start automatically

Poland) Wydawnictwo Gea (Warsaw. Click here if your download doesnt start automatically Suwalski Park Krajobrazowy i okolice 1:50 000, mapa turystyczno-krajoznawcza =: Suwalki Landscape Park, tourist map = Suwalki Naturpark,... narodowe i krajobrazowe) (Polish Edition) Click here if your

Bardziej szczegółowo

SNP Business Partner Data Checker. Prezentacja produktu

SNP Business Partner Data Checker. Prezentacja produktu SNP Business Partner Data Checker Prezentacja produktu Istota rozwiązania SNP Business Partner Data Checker Celem produktu SNP Business Partner Data Checker jest umożliwienie sprawdzania nazwy oraz danych

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

Projekt: Mikro zaprogramowane na sukces!

Projekt: Mikro zaprogramowane na sukces! Projekt współfinansowany przez Unię Europejską w ramach Europejskiego Funduszu Społecznego Projekt: Mikro zaprogramowane na sukces! Opis autoryzowanych szkoleń Oracle planowanych do realizacji w ramach

Bardziej szczegółowo

Dolny Slask 1: , mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition)

Dolny Slask 1: , mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition) Dolny Slask 1:300 000, mapa turystycznosamochodowa: Plan Wroclawia (Polish Edition) Click here if your download doesn"t start automatically Dolny Slask 1:300 000, mapa turystyczno-samochodowa: Plan Wroclawia

Bardziej szczegółowo

MAGNESY KATALOG d e s i g n p r o d u c e d e l i v e r

MAGNESY KATALOG d e s i g n p r o d u c e d e l i v e r MAGNESY KATALOG design produce deliver MAGNET 0,4 / 0,75MM owal, prostokąt, koło, kwadrat od 50 sztuk Flexible magnet 0.4 = strength example: able to hold one A4 sheet. 0.75 = strength example: able to

Bardziej szczegółowo

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition)

Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Wojewodztwo Koszalinskie: Obiekty i walory krajoznawcze (Inwentaryzacja krajoznawcza Polski) (Polish Edition) Robert Respondowski Click here if your download doesn"t start automatically Wojewodztwo Koszalinskie:

Bardziej szczegółowo

SubVersion. Piotr Mikulski. SubVersion. P. Mikulski. Co to jest subversion? Zalety SubVersion. Wady SubVersion. Inne różnice SubVersion i CVS

SubVersion. Piotr Mikulski. SubVersion. P. Mikulski. Co to jest subversion? Zalety SubVersion. Wady SubVersion. Inne różnice SubVersion i CVS Piotr Mikulski 2006 Subversion is a free/open-source version control system. That is, Subversion manages files and directories over time. A tree of files is placed into a central repository. The repository

Bardziej szczegółowo

Proposal of thesis topic for mgr in. (MSE) programme in Telecommunications and Computer Science

Proposal of thesis topic for mgr in. (MSE) programme in Telecommunications and Computer Science Proposal of thesis topic for mgr in (MSE) programme 1 Topic: Monte Carlo Method used for a prognosis of a selected technological process 2 Supervisor: Dr in Małgorzata Langer 3 Auxiliary supervisor: 4

Bardziej szczegółowo

Network Services for Spatial Data in European Geo-Portals and their Compliance with ISO and OGC Standards

Network Services for Spatial Data in European Geo-Portals and their Compliance with ISO and OGC Standards INSPIRE Conference 2010 INSPIRE as a Framework for Cooperation Network Services for Spatial Data in European Geo-Portals and their Compliance with ISO and OGC Standards Elżbieta Bielecka Agnieszka Zwirowicz

Bardziej szczegółowo

ABOUT NEW EASTERN EUROPE BESTmQUARTERLYmJOURNAL

ABOUT NEW EASTERN EUROPE BESTmQUARTERLYmJOURNAL ABOUT NEW EASTERN EUROPE BESTmQUARTERLYmJOURNAL Formanminsidemlookmatmpoliticsxmculturexmsocietymandm economyminmthemregionmofmcentralmandmeasternm EuropexmtheremismnomothermsourcemlikemNew Eastern EuropeImSincemitsmlaunchminmPw--xmthemmagazinemhasm

Bardziej szczegółowo

. Natural Language Processing. Jan Daciuk Department of Intelligent Interactive Systems ETI Faculty, Gdańsk University of Technology.

. Natural Language Processing. Jan Daciuk Department of Intelligent Interactive Systems ETI Faculty, Gdańsk University of Technology. .. Natural Language Processing Jan Daciuk Department of Intelligent Interactive Systems ETI Faculty, Gdańsk University of Technology May 5, 2014 Jan Daciuk, DIIS, ETI, GUT Natural Language Processing 0.

Bardziej szczegółowo