a.Structure data (tabular data,etc...)

b.Unstructured data (images,text,audio,etc...)

c.semi structured data (XML,JSON,etc...)

a.qualitative (nominal,ordinal,binary)

b.quantitative(discrete,continuous)

1.Beautifulsoup

2.Scrapy

3.Selenium

4.Request to access data 

5.AUTOSCRAPER - https://github.com/alirezamika/autoscraper

6.Twitter scraping tool (𝚝𝚠𝚒𝚗𝚝 or tweepy)-https://github.com/twintproject/twint

  https://analyticsindiamag.com/complete-tutorial-on-twint-twitter-scraping-without-twitters-api/
  
  Scraping Instagram -instaloader

7.urllib

8.pattern

9.Octoparse Easy Web Scraping  https://www.octoparse.com/ 

 ParseHub https://www.parsehub.com/  https://analyticsindiamag.com/parsehub-no-code-gui-based-web-scraping-tool/
 
 Diffbot  https://analyticsindiamag.com/diffbot/
 
 Trustpilot

1)kaggle-https://www.kaggle.com/datasets , 𝚙𝚒𝚙 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 𝚔𝚊𝚐𝚐𝚕𝚎𝚍𝚊𝚝𝚊𝚜𝚎𝚝𝚜

2)movielens-https://grouplens.org/datasets/movielens/latest/

3)data.gov-https://data.gov.in/

4)uci-https://archive.ics.uci.edu/ml/datasets.php

5)Group Lens dataset https://grouplens.org/

6)world3bank  https://data.world/ , worldbank

7)Google Cloud BigQuery public datasets

  Google Public Datasets-cloud.google.com/bigquery/public-data/
  
  Google Cloud Data Catalog  https://cloud.google.com/data-catalog
  
  Academic Torrents-https://academictorrents.com/check.htm?returnto=%2Fbrowse.php

8)online hacktons

9)image data from google_images_download

https://www.visualdata.io/discovery

http://xviewdataset.org/#dataset

https://ai.googleblog.com/2016/09/introducing-open-images-dataset.html

10)image data from Bing_Search

11)https://www.columnfivemedia.com/100-best-free-data-sources-infographic

12)Reddit:https://lnkd.in/dv5UCD4

13)https://datasets.bifrost.ai/?ref=producthunt

14)data.world:https://lnkd.in/gEK897K

15)https://data.world/datasets/open-data

   https://tinyletter.com/data-is-plural

16)FiveThirtyEight :-  https://lnkd.in/gyh-HDj , https://data.fivethirtyeight.com/

17)BuzzFeed :- https://lnkd.in/gzPWyHj

   Buzzfeed News -github.com/BuzzFeedNews
   
   Socrata - https://opendata.socrata.com/

18)Google public datasets :- https://lnkd.in/g5dH8qE

19)Quandl :- https://www.quandl.com  stock data

   statista : https://www.statista.com/ stock data

20)socorateopendata :- https://lnkd.in/gea7JMz

21)AcedemicTorrents :- https://lnkd.in/g-Ur9Xy

22)labelimage:- https://github.com/wkentaro/labelme  ,  https://github.com/tzutalin/labelImg 

Labelbox-https://labelbox.com/

Playment-https://playment.io/

SuperAnnotate -https://www.superannotate.com/

CVAT-https://github.com/openvinotoolkit/cvat

Lionbridge- https://lionbridge.ai/

LinkedAI: A No-code Data Annotations- https://analyticsindiamag.com/linkedai/

Dataturks

V7 Darwin The Rapid Image Annotator   https://docs.v7labs.com/docs/loading-a-dataset-in-python   https://github.com/v7labs/darwin-py#usage-as-a-python-library

https://github.com/heartexlabs/awesome-data-labeling

23)tensorflow_datasets as tfds  https://www.tensorflow.org/datasets  (import tensorflow_datasets as tfds)

24)https://datasets.bifrost.ai/?ref=producthunt

25)https://ourworldindata.org/

26)https://data.worldbank.org/

27)google open images:https://storage.googleapis.com/openimages/web/download.html

28)https://data.gov.in/

29)imagenet dataset-http://www.image-net.org/

30)https://parulpandey.com/2020/08/09/getting-datasets-for-data-analysis-tasks%e2%80%8a-%e2%80%8aadvanced-google-search/

31)https://storage.googleapis.com/openimages/web/index.html  , 

   https://storage.googleapis.com/openimages/web/visualizer/index.html?set=train&type=segmentation&r=false&c=%2Fm%2F09qck
   
   https://console.cloud.google.com/marketplace/browse?filter=solution-type:dataset&_ga=2.35328417.1459465882.1589693499-869920574.1589693499
   
   https://catalog.data.gov/dataset?groups=education2168#topic=education_navigation
   
   https://vincentarelbundock.github.io/Rdatasets/datasets.html
 
32)coco dataset https://cocodataset.org/#explore

33)huggingface datasets-https://github.com/huggingface/datasets  https://huggingface.co/datasets

34)Big Bad NLP Database-https://datasets.quantumstat.com/

https://github.com/niderhoff/nlp-datasets

35)https://www.edureka.co/blog/25-best-free-datasets-machine-learning/

36)bigquery public dataset ,Google Public Data Explorer

37)inbuilt library data eg:iris dataset,mnist dataset,etc...

tf.data.Datasets for TensorFlow Datasets 

38)data.gov.be ,data.egov.bg/ ,data.gov.cz/english ,portal.opendata.dk,govdata.de,opendata.riik.ee,data.gov.ie,data.gov.gr,datos.gob.es,data.gouv.fr,data.gov.hr

dati.gov.it,data.gov.cy,opendata.gov.lt,data.gov.lv,data.public.lu,data.gov.mt,data.overheid.nl,data.gv.at,danepubliczne.gov.pl,dados.gov.pt,data.gov.ro,podatki.gov.si

data.gov.sk,avoindata.fi,oppnadata.se,https://data.adb.org/ ,https://data.iadb.org/ ,https://www.weforum.org/agenda/2018/03/latin-america-smart-cities-big-data/

https://data.fivethirtyeight.com/ , https://wiki.dbpedia.org/ ,https://www.europeandataportal.eu/en ,https://data.europa.eu/ ,https://www.census.gov/,

https://www.who.int/data/gho ,https://data.unicef.org/open-data/ ,http://data.un.org/ ,https://data.oecd.org/ ,https://data.worldbank.org/  

39.Awesome Public Dataset- https://github.com/awesomedata/awesome-public-datasets

40.Datasets for Machine Learning on Graphs-https://ogb.stanford.edu/

41.Big Bad NLP Database-https://datasets.quantumstat.com/

42.30 largest tensorflow datasets-https://lionbridge.ai/datasets/tensorflow-datasets-machine-learning/

43. coco dataset-https://cocodataset.org/#home

Google Open images-https://opensource.google/projects/open-images-dataset  https://storage.googleapis.com/openimages/web/index.html

50+ Object Detection Datasets-https://medium.com/towards-artificial-intelligence/50-object-detection-datasets-from-different-industry-domains-1a53342ae13d

   70+ Image Classification Datasets from different Industry domains-https://medium.com/towards-artificial-intelligence/70-image-classification-datasets-from-different-industry-domains-part-2-cd1af6e48eda
   
tensorflow_datasets.object_detection - https://storage.googleapis.com/openimages/web/index.html

https://github.com/google-research-datasets/Objectron/  https://ai.googleblog.com/2020/11/announcing-objectron-dataset.html?m=1 

http://idd.insaan.iiit.ac.in/   http://database.mmsp-kn.de/koniq-10k-database.html

https://ai.googleblog.com/2020/11/announcing-objectron-dataset.html

ImageNet data -http://image-net.org/

ApolloScape Dataset-http://apolloscape.auto/

44.https://github.com/fivethirtyeight/data

45.Recommender Systems Datasets-https://cseweb.ucsd.edu/~jmcauley/datasets.html

46.indiadataportal-https://indiadataportal.com/

47.US Government Open Dataset: https://www.data.gov/

https://censusreporter.org/   https://data.census.gov/cedsci/

48.AWS Public Data Sets:https://registry.opendata.aws/    https://aws.amazon.com/opendata/?wwps-cards.sort-by=item.additionalFields.sortDate&wwps-cards.sort-order=desc

49.https://the-eye.eu/public/AI/pile_preliminary_components/

  Reddit -https://www.reddit.com/r/datasets/
  
  wikipedia-https://en.wikipedia.org/wiki/List_of_datasets_for_machine-learning_research
  
  http://opendata.cern.ch/  ,  https://www.imf.org/en/Data
  
  Global Health Observatory data repository-https://apps.who.int/gho/data/node.main
  
  CERN Open Data Portal-http://opendata.cern.ch/
  
50.openblender- https://www.openblender.io/#/welcome

51.Top 10 Datasets For Cybersecurity Projects- https://analyticsindiamag.com/top-10-datasets-for-cybersecurity-projects/

52.Datasets from Web Crawl Data (nlp)-http://data.statmt.org/cc-100/

53.https://www.springboard.com/blog/free-public-data-sets-data-science-project/

54.NASA - https://nasa.github.io/data-nasa-gov-frontpage/ace 

55.Academic Torrents,GitHub Datasets,CERN Open Data Portal,Global Health Observatory Data Repository

56.32 Data Sets to Uplift your Skills in Data Science-https://blog.datasciencedojo.com/data-sets-data-science-skills/?utm_content=144243072&utm_medium=social&utm_source=linkedin&hss_channel=lcp-3740012

57.OpenDaL-https://opendatalibrary.com/

Data Is Plural-https://docs.google.com/spreadsheets/d/1wZhPLMCHKJvwOkP4juclhjFgqIY8fQFMemwKL2c64vk/edit#gid=0

VisualData-https://www.visualdata.io/discovery

https://medium.com/towards-artificial-intelligence/best-datasets-for-machine-learning-data-science-computer-vision-nlp-ai-c9541058cf4f
 
58.Pandas Data Reader-https://pandas-datareader.readthedocs.io/en/latest/remote_data.html

59.ieee-dataport-https://ieee-dataport.org/datasets

https://medium.com/towards-artificial-intelligence/best-datasets-for-machine-learning-data-science-computer-vision-nlp-ai-c9541058cf4f

https://github.com/neomatrix369/awesome-ai-ml-dl/blob/master/data/datasets.md#datasets-and-sources-of-raw-data

60.Faker is a Python package that generates fake data for you-https://github.com/joke2k/faker

61.Text Data Annotator Tool - Datasaur  https://datasaur.ai/

62.Google Analytics cost data import https://segmentstream.com/google-analytics?utm_source=twitter&utm_medium=cpc&utm_campaign=ga_costs_import_en&utm_content=guide

63.https://lionbridge.ai/services/crowdsourcing/    https://lionbridge.ai/     https://www.clickworker.com/  https://appen.com/  https://www.globalme.net/

64.Azure Open Datasets https://azure.microsoft.com/en-us/services/open-datasets/       https://azure.microsoft.com/en-in/services/open-datasets/catalog/
  
Yelp Open Dataset  https://www.yelp.com/dataset

https://data.world/

ODK Open Data Kit- https://getodk.org/

World Bank Open Data https://data.worldbank.org/

https://analyticsindiamag.com/10-biggest-data-breaches-that-made-headlines-in-2020/

https://data.mendeley.com/

https://github.com/iamtekson/geospatial-data-download-sites

https://eugeneyan.com/writing/data-discovery-platforms/

65.https://medium.com/towards-artificial-intelligence/best-datasets-for-machine-learning-data-science-computer-vision-nlp-ai-c9541058cf4f

https://github.com/MTG/freesound-datasets

 Types of missing value 
 
 i.missing completely at random(no correlation b/w missing and observed data) we can delete no disturbance of data distribution
 
 ii.missing at random (randomness in missing data, missing value have correlation by data) we can't delete because disturbance of data distribution
 
 iii.missing not at random  (there is reason for missing value and directly related to value)

 1.if missing data too small then delete it a.row deletion b.column deletion c.pairwise deletion
 
 2.replace by statistical method mean(influenced by outiler),median(not influenced by outiler),mode
 
 3.apply classifier algorithm to predict missing value
 
 4.knn imputer
 
 5.apply unsupervised 
 
 6.Random Sample Imputation
 
 7.Adding a variable to capture NAN(missing term)
 
 8.Arbitrary Value Imputation
 
 9.hot deck Imputation
 
 10.regression Imputation
 
 11.End of Distribution Imputation
 
 12.Arbitrary Value Imputation
 
 13.Frequent Category Imputation
 
 14.MICE Imputation
 
 https://stefvanbuuren.name/fimd/want-the-hardcopy.html

 1.Under Sampling - mostly not prefer because lost of data
 
 2.Over Sampling  (RandomOverSampler (here new points create by same dot)) ,  SMOTETomek(new points create by nearest point so take long time),BorderLine Smote,KMeans Smote,SVM Smote,ADASYN,Smote-NC
 
 https://towardsdatascience.com/7-over-sampling-techniques-to-handle-imbalanced-data-ec51c8db349f
 
 3.class_weight give more importance(weight) to that small class
 
 4.use Stratified kfold to keep the ratio of classess constantly
 
 5.Weighted Neural Network

 1.One Hot Encoding
 
 2.Count Or Frequency Encoding
 
 3.Target Guided Ordinal Encoding
 
 4.Mean Encoding
 
 5.Probability Ratio Encoding
 
 6.label encoding
 
 7.probability ratio encoding
 
 8.woe(Weight_of_evidence)
 
 9.one hot encoding with multi category (keep most frequently repeated only)
 
 10.feature hashing 
 
 11.sparse csr matrix
 
 12.entity embeddings
 
 13.binary encoding
 
 14.Rare label encoding
 
 15.Leave-one-out(Loo) encoding

   1.Normalisation  

   2.Standardization
 
   3.Robust Scaler not influenced by outliers because using of median,IQR
   
   4. Min Max Scaling
   
   5.Mean normalization
   
   6.maximum absolute scaling

       a.Guassian Transformation
    
       b.Logarithmic Transformation
    
       c.Reciprocal Trnasformation
    
       d.Square Root Transformation
    
       e.Exponential Transdormation
    
       f.BoxCOx Transformation
    
       g.log(1+x) Transformation
       
       h.johnson

  2 type of outilers available: Global outiler, Local outiler

  eg: incase of fraud detection outilers are very important
  
  methods to find outiler: Standard Deviation,zscore,boxplot,scatter plot,IQR,TensorFlow_Data_Validation
  
  Automatic Outlier Detection:Isolation Forest,Local Outlier Factor,Minimum Covariance Determinant,Robust Random Cut Forest,DBScan Clustering
  
  outiler treatment: mean/median/random imputation,drop,discretization (binning)
  
  if outiler present then use robust scaling
  
  https://medium.com/towards-artificial-intelligence/outlier-detection-and-treatment-a-beginners-guide-c44af0699754

 clustering techniques to find it
 
 Isolation Forest(for Big Data)

 a.biased sampling
 
 b.unbiased sampling

Explore the dataset by using  python or microsoft excel or tableau or powerbi, etc...

Data visualization (Matplotlib,Seaborn,Plotly,pyqtgraph,Bokeh,Pygal,Dash,Pydot,Geoplotlib,ggplot,visualizer,etc...)

Scatterplot,multi line plot,bubble chart,bar chart,histogram,boxplot,distplot,bubble charts,area plot,heat map,index plot,violin plot,time series plot,density plot,dot plot,strip plot,plotly,Choropleth Map,Kepler,PDF,Kernel density function,networkx,Scatter_matrix,Bootstrap_plot,functionvis,Higher-Dimensional Plots,3-D Plots,Word Clouds

𝗞𝗲𝗿𝗮𝘀 𝗠𝗼𝗱𝗲𝗹 𝘃𝗶𝘀𝘂𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗼𝗿(ann-visualizer)- 𝗽𝗶𝗽𝟯 𝗶𝗻𝘀𝘁𝗮𝗹𝗹 𝗴𝗿𝗮𝗽𝗵𝘃𝗶𝘇

univariate and bivariate and multivariate analysis

model visualization Tensorboard,netron,playground tensorflow,plotly,TensorDash,Dash,Microscope,Lucid

distributions(discerte,continous)

data distributions-normal distribution,Standard Normal Distribution,Student's t-Distribution,Bernoulli Distribution,Binomial Distribution,Poisson Distribution,Uniform Distribution,F Distribution,Covariance and Correlation

Types of Statistics  

1.Descriptive

2.Inferential

Types of data

1) Categorical (nomial,ordinal)
 
2) Numerical   (discerte,continous)

random variable(discerte random variable ,continous random variable)

Central Limit Theorem,Bayes Theorem,Confidence Interval,Hypothesis Testing,z test, t test,f test,Confidence Interval,1 tail test, 2 tail test,chisquare test,anova test,A/B testing

1.Filter methods (correleation,chisquare  test,Ttest,anova test,hypothesis test,information gain etc...)

2.Wrapper methods (forward selection,backwaed elimination,stepwise selection etc...)

3.Embedded method (lasso,ridge regression,elasticnet etc...)

4.Feature Importance

   a.ExtraTreesClassifier,ExtraTreesregressor

   b.SelectKBest

   c.Logistic Regression

   d.Random_forest_importance
   
   e.decision tree
   
   f.Linear Regression
   
   g.xgboost

5.curse of dimensionality (as dimension increases performance decreases)

6.highly correleated features then can take any 1 feature (multicollinearity)

7.dimension reduction

8.lasso regression to penalise unimportant features

9.threshold based method 

10.model based selection

11.Mutual Information Feature Selection

12.remove features with very low variance (quasi constant feature dropping)

13.Univariate  feature selection

14.importance of feature (random forest importance)

15.feature importance with decision trees

16.PyImpetus

17.drop constant features (variance=0)

18.variance inflation factor(vif)

19.Recursive Feature Elimination

https://www.analyticsvidhya.com/blog/2020/10/a-comprehensive-guide-to-feature-selection-using-wrapper-methods-in-python/

 Splitting ratio of data deponds on size of dataset available

 Training data,Validation data,Testing data

 1.Regression (output feature in continous data form)
 
   linear regression,polynomial regression,Robust Regression,support vector regression,Decision Tree Regression,Random Forest Regression,
   
   least square method,Random Forest Regression,xgboost,ridge(L2 Regularization),lasso(L1 Regularization (more sparse)),catboost,gradientboosting,adaboost,
   
   elsatic net,light gbm,ordinary least squares,cart
   
   use cases:

 2.Classification (output feature in categorical data form)
 
    Logistic Regression,K-Nearest Neighbors,Support Vector Machine,Kernel SVM,Naive Bayes,Decision Tree Classification,
    
    Random Forest Classification,xgboost,adaboost,Gradient Boost,catboost,gaussian NB,LGBMClassifier,LinearDiscriminantAnalysis, Extreme Gradient Boosting Machine, passive aggressive classifier algorithm,cart,c4.5,c5.0

 1.Dimensionality reduction - PCA,SVD,LDA,som,tsne,plsr,pcr,autoencoders,kpca,lsa

 2.Clustering :https://scikit-learn.org/stable/modules/clustering.html

 3.Association Rule Learning - support,lift,confidence,aprior,elcat,Fp-growth,Fp-tree construction, association_rules

 4.Recommendation system -
 
     a.collaborative Recommendation system (model based, memory based(item based,user based))  user-item interaction matrix
    
     b.content based Recommendation system 
     
     similarity based(user-user similarity,item-item similarity)
     
     matrix factorization
     
     c.utility based Recommendation system 
     
     d.knowledge based Recommendation system 
     
     e.demographic based Recommendation system 
     
     f.hybrid based Recommendation system 
     
     g.Average Weighted Recommendation
     
     h.using K Nearest Neighbor
     
     i.cosine distance recommender system
     
     j.TensorFlow Recommenders https://www.tensorflow.org/recommenders
     
     k.suprise baseline model
     
     https://analyticsindiamag.com/top-open-source-recommender-systems-in-python-for-your-ml-project/

 1.Stacking models

 2.Bagging models

 3.Boosting models
 
 4.Blending
 
 5.Voting (Hard Voting,Soft Voting)

  2 types a)model free   b)model based

  agent,environment,policy(On-Policy vs Off-Policy),reward function,value function,state,action,episode,actor-critic

  agent apply action to environment get corresponding reward so that it learn environment
  
  1.Q-Learning
  
  2.Deep Q-Learning
  
  3.Deep Convolutional Q-Learning
  
  Deep Deterministic Policy Gradient
  
  4.Twin Delayed DDPG,DQN
  
  5.A3C  (Actor Critic)
  
  6.Advantage weighted actor critic (AWAC). 
  
  7.XCS
  
  8.genetic algorithm,sarsa
  
  https://simoninithomas.github.io/deep-rl-course/
  
   Environments-OpenAI Gym, DeepMind Lab, Unity ML-Agents
   
   https://analyticsindiamag.com/8-best-free-resources-to-learn-deep-reinforcement-learning-using-tensorflow/
   
   https://neptune.ai/blog/best-reinforcement-learning-tutorials-examples-projects-and-courses
   
   https://neptune.ai/blog/best-reinforcement-learning-tutorials-examples-projects-and-courses?utm_source=twitter&utm_medium=tweet&utm_campaign=blog-best-reinforcement-learning-tutorials-examples-projects-and-courses
   
   Open AI Gym - https://gym.openai.com/
   
   KerasRL https://github.com/keras-rl/keras-rl
   
   pyqlearning
   
   tensorforce https://tensorforce.readthedocs.io/en/latest/index.html  
   
   rl_coach https://github.com/IntelLabs/coach#installation        MushroomRL https://mushroomrl.readthedocs.io/en/latest/
   
   TFAgents  https://github.com/tensorflow/agents (https://www.tensorflow.org/agents)   https://deepmind.com/blog/article/trfl     
   
   Stable Baselines  https://github.com/openai/baselines
   
   https://neptune.ai/blog/the-best-tools-for-reinforcement-learning-in-python?utm_source=twitter&utm_medium=tweet&utm_campaign=blog-the-best-tools-for-reinforcement-learning-in-python

 1.Regression task

 2.Classification task

 1.Classification of image
 
   create own model,Lenet,Alexnet,Resenet,GoogleNet,Inception,Vgg,Efficient,Nasnet,STN,nasneta,senet,amoebanetc
 
 2.Localization of object in image
 
 3.Object detection and object segmentation 
 
   rcnn,fastrcnn,fastercnn,TensorFlow Object Detection,yolo v1,yolo v2,yolo v3,yolo v4,fast yolo,yolo tiny,yolo lite,yolo tiny++,yolo act++,
   
   maskrcnn,ssd,detectron,detectron2,mobilenet,retinanet,R-fcn,detr facebook,U-net,UNet++ 
   
   3 kind of object segmentation are available semantic segmentation,instance segmentation,panoptic segmentation
   
   PyTorch based low code object detection-https://github.com/alankbi/detecto
   
   https://awesomeopensource.com/project/hoya012/deep_learning_object_detection
 
 4.objecttracking  (mean shit and optical flow and kalman filter)
 
   Tracktor++,Trackrcnn,Jde,DeepSORT 
 
 5.Deepdream,Neural style transfer, Pose estimation 
 
 CNNs 'see' - FilterVisualizations, Heatmaps,Saliency Maps,Heat Map Visualizations
 
 imageai.Detection for Object detection
 
 DEEP LEARNING METHODS FOR 2D :OpenPose,DeepPose,MultiPoseNet,AlphaPose,VIBE,DeeperCut,Mask RCNN,DeepCut,Convolutional Pose Machines,PoseNet
 
 3D POSE ESTIMATION
 
 DEEP LEARNING METHODS FOR 3D:3D human pose estimation= 2D pose estimation + matching,Integral Human Pose Regression,Towards 3D Human Pose Estimation in the

 Data Augmentation apply to increase size of dataset and performance of model
 
 low code object detection -  detecto  https://github.com/alankbi/detecto 
 
 Object Detection with 10 lines of code-https://www.datasciencecentral.com/profiles/blogs/object-detection-with-10-lines-of-code

 1.RNN
 
 2.GRU
 
 3.LSTM (have memory cell,forget gate  etc..)
 
 all above 3 models have bidirectional also based on problem statement use bidirectional models

 Cycle gan,Dcgan,SRGAN,InfoGAN,stargan,attan gan,stylegan,,PixelRNN,DiscoGAN,lsGAN,Conditional GAN(Pix2Pix),Progressive GANs( produces higher resolution images,Image-to-Image Translation),Face Inpainting,Super-resolution
 
 https://github.com/hindupuravinash/the-gan-zoo

  1.sparse Autoencoder
  
  2.denoising Autoencoder
  
  3.Contractive Autoencoder
  
  4.stacked Autoencoder
  
  5.deep Autoencoder
  
  6.variational autoencoder

 Clean data(removing stopwords depond on problem ,lowering data,tokenization,postagging,stemmimg or lemmatization depond on problem,skipgram,n-gram,chunking)
 
 Nltk,spacy,genism,textblob,inltk,Pattern,stanza,OpenNLP,polygot,corenlp,polyglot,PyDictionary,Huggiing face,spark nlp,allen nlp,rasa nlu,Megatron,texthero,Flair,textacy,finetune,gluon-nlp,VnCoreNLP  libraries
 
 NLU,NLG,NER,text summarization,Sentiment Analysis,Text Classifications,machine translation,chat bot,Text Generation,Speech Recognition
  
 1.bag of words
 
 2.Tfidf
 
 3.wordembedding
    
    a.using pretrained model 
      
      i)word2vec( cbow,skipgram)
      
      ii)glove
      
      iiI)fasttext
    
    b.creating own embedding  (use when have huge data)
    
      i)word2vec library
      
      ii)keras embedding 
    
 4.Document embedding-Doc2vec
  
 5.sentence embedding

   sense2vec,SENT2VEC,Universal sentence encoder
 
 6.using rnn,lstm,gru
 
   for above 3 models have bidirectional also
 
 7.Encoder and Decoder(sequence to sequence), ProphetNet(new pretrained seq2seq model)
  
 8.attention 
 
   self attention,Global Attention,Multi-Head Attention,Local Attention (monotonic,predictive)    https://github.com/uzaymacar/attention-mechanisms
 
 9.Transformer (big breakthrough in NLP) - http://jalammar.github.io/illustrated-transformer/  
 
    Shrinking Transformers (reduce size)  1.quantization,distillation,pruning,
  
 10.BERT,Quantized MobileBERT,ALBERT,DistilBERT,ELMo,ROBERTA,XLNet,XLM-RoBERTa,T5,DISTILBERT,GPT,GPT2,GPT3,PRADO,PET,BORT
 
    http://jalammar.github.io/    http://jalammar.github.io/illustrated-bert/   http://jalammar.github.io/a-visual-guide-to-using-bert-for-the-first-time/
    
 11.Speech
   
    speech to text   
    
    text to speech
    
    Acoustic model,Speaker diarisation,apis
    
 googletrans (google Translator)
    
 https://medium.com/towards-artificial-intelligence/natural-language-processing-nlp-with-python-tutorial-for-beginners-1f54e610a1a0
 
 MT5-https://venturebeat.com/2020/10/26/google-open-sources-mt5-a-multilingual-model-trained-on-over-101-languages/?utm_content=144321587&utm_medium=social&utm_source=linkedin&hss_channel=lcp-3740012
 
 VADER does not require any training data https://pypi.org/project/vaderSentiment/  https://analyticsindiamag.com/sentiment-analysis-made-easy-using-vader/
 
 APPLICATIONS OF MACHINE TRANSLATIO-Text-to-text,Text-to-speech,Speech-to-text,Speech-to-speech,Image (of words)-to-text
 
 Google-GNMT (Tensorflow),Facebook-fairseq (Torch),Amazon-Sockeye (MXNet),NEMATUS (Theano),THUMT (Theano),OpenNMT (PyTorch),StanfordNMT (Matlab),DyNet-lamtram(CMU),EUREKA(MangoNMT

  here data split is different (train,test,validate)
  
  here handling missing data different 
  
  generally used  to impute data in Time Series
  
  1.ffill
  
  2.bfill
  
  3.do mean of previous or future x samples and impute
  
  4.take previous season value and impute (data with trend)
  
  5.mean,mode,median,random sample imputation (data without trend and without seasonality)
  
  6.linear interpolation(data with trend and without seasonality)
  
  7.seasonal +interpolation(data with trend and with seasonality)
  
  here model selection deponds on different property of data like stationary,trend,seasonality,cyclic
  
  adfuller test  for  Stationarity
  
  models 
  
  1.Arima , auto arima ,seasonal arima
  
  2.Autoregressive 
  
  3.Moving average,Exponential Moving average,Exponential Smoothing
  
  4.Lstm(neural network)
  
  5.Autoregressive
  
  6.Navie forecasts
  
  7.Smoothing (moving average,exponential smoothing)
  
  8.Facebook prophet (note:expceted date column as ds and target column as y)
  
  9.Holts winter,Holts linear trend
  
  10.AutoTS-https://analyticsindiamag.com/hands-on-guide-to-autots-effective-model-selection-for-multiple-time-series/
  
  11.Temporal Convolutional Neural
  
  12.Atspy For Automating The Time-Series Forecasting-https://analyticsindiamag.com/hands-on-guide-to-atspy-for-automating-the-time-series-forecasting/
  
  13.Darts-https://analyticsindiamag.com/hands-on-guide-to-darts-a-python-tool-for-time-series-forecasting/
  
  14.Bayesian Neural Network 
  
  15.PyFlux (easy way to compare different models)-https://analyticsindiamag.com/pyflux-guide-python-library-for-time-series-analysis-and-prediction/
  
  16.Orbit , DeepAR ,NeuralProphet(https://github.com/ourownstory/neural_prophet    https://ourownstory.github.io/neural_prophet/model-overview/)
  
  best article-https://www.analyticsvidhya.com/blog/2018/02/time-series-forecasting-methods/,
  
  https://machinelearningmastery.com/time-series-forecasting-methods-in-python-cheat-sheet/
  
  https://www.machinelearningplus.com/time-series/time-series-analysis-python/
  
  https://github.com/Apress/hands-on-time-series-analylsis-python
  
  https://otexts.com/fpp2/simple-methods.html
      
  https://analyticsindiamag.com/top-time-series-deep-learning-methods/

a.GridSearchCV (check every given parameter so take long time)

b.RandomizedSearchCV (search randomly narrow down our time)

c.Bayesian Optimization , Hyperopt

d.Sequential Model Based Optimization(Tuning a scikit-learn estimator with skopt)

e.Optuna

f.Genetic Algorithms

g.Keras tuner

h.Scikit-Optimize

https://towardsdatascience.com/10-hyperparameter-optimization-frameworks-8bc87bc8b7e3

 1.Loocv
 
 2.Kfoldcv
 
 3.Stratfied cross validation
 
 4.Time Series cross-validation
 
 5.Holdout cross-validation
 
 6.Repeated cross-validation

 Always check bias variance tradeoff to know how model is performing
 
 Model can be overfitting(low bias,high variance),underfitting(high bias,high variance),good fit(low bias,low variance)
 
1.Regression task - mean-squared-error, Root-Mean-Squared-Error,mean-absolute error, R², Adjusted R²,Cross-entropy loss,Mean percentage error 

2.Classification task-Accuracy,confusion matrix,Precision,Recall,F1 Score,Binary Crossentropy,Categorical Crossentropy,AUC-ROC curve,log loss,Average precision,Mean average precision

3.Reinforcement learning - generally  use rewards

4.Incase of machine translation use bleu score

5.Clustering then use External: Adjusted Rand index, Jaccard Score, Purity Score    Internal:silhouette_score, Davies-Bouldin Index, Dunn Index

6.Object Detection loss-localization loss,classification loss,Focal Loss,IOU,L2 loss

7.Distance Metrics - Euclidean Distance,Manhattan Distance,Minkowski Distance,Hamming Distance

metric-Built-in metrics, Custom metric without external parameters,Custom metric with external parameters,Subclassing custom metric layer

https://medium.com/swlh/custom-loss-and-custom-metrics-using-keras-sequential-model-api-d5bcd3a4ff28

loss-Built-in loss, Custom loss without external parameters,Custom loss with external parameters,Subclassing loss layer

1.Azure

2.Heroku

3.Amazon Web Services

4.Google cloud platform

MODEL DEPLOYMENT USING TF SERVING
 
TensorFlow Extended (TFX) is an end-to-end platform for deploying production ML pipelines https://www.tensorflow.org/tfx

Models visualization using Tensorboard,netron, TensorBoard.dev

Python web Frameworks for App Development- Flask,Streamlit,fastapi,Django,Web2py,Pyramid,CherryPy,Voila,Kivy and Kivymd  https://analyticsindiamag.com/top-8-python-tools-for-app-development/

Web-Based GUI (Gradio)- https://analyticsindiamag.com/guide-to-gradio-create-web-based-gui-applications-for-machine-learning/

https://github.com/gradio-app/gradio

1.COURSERA

2.UDEMY

3.EDX

4.DATACAMP

5.Udacity

1.Krish Naik-https://www.youtube.com/user/krishnaik06

2.Codebasics-https://www.youtube.com/channel/UCh9nVJoWXmFb7sLApWGcLPQ  

3.Abhishek thakur-https://www.youtube.com/user/abhisheksvnit

4.AIEngineering-https://www.youtube.com/channel/UCwBs8TLOogwyGd0GxHCp-Dw

5.Ineuron-https://www.youtube.com/channel/UCb1GdqUqArXMQ3RS86lqqOw

6.Ken jee-https://www.youtube.com/c/KenJee1/featured       

7.3Blue1Brown-https://www.youtube.com/c/3blue1brown/featured

8.The AI Guy -https://www.youtube.com/channel/UCrydcKaojc44XnuXrfhlV8Q 

9.Unfold Data Science-https://www.youtube.com/channel/UCh8IuVJvRdporrHi-I9H7Vw

1.Towards data science-https://towardsdatascience.com/

2.Analyticsvidhya-https://www.analyticsvidhya.com/blog/?utm_source=feed&utm_medium=navbar

3.Medium-https://medium.com/

4.Machinelearningmastery-https://machinelearningmastery.com/blog/

5.ML+  -https://www.machinelearningplus.com/

1.Linkedin

2.Twitter

1.Google cloab (FREE)

2.Kaggle kernel(read terms and conditions before use) (FREE)

3.Paperspace Gradient(read terms and conditions before use)

4.knime - https://www.knime.com/(read terms and conditions before use)

5.RapidMiner (read terms and conditions before use)

https://github.com/zszazi/Deep-learning-in-cloud

1.E-commerce- Identifying consumers,Recommending Products,Analyzing Reviews

2.Manufacturing- Predicting potential problems,Monitoring systems,Automating manufacturing units, Maintenance Scheduling,Anomaly Detection

3.Banking- Fraud detection,Credit risk modeling,Customer lifetime value

4.Healthcare- Medical image analysis, Drug discovery,Bioinformatics,Virtual Assistants,image segmentation

5.Transport- Self-driving cars,Enhanced driving experience,Car monitoring system,Enhancing the safety of passengers

6.Finance- Customer segmentation,Strategic decision making,Algorithmic trading,Risk analytics

7.Marketing (Added from comments Credits: Jawad Ali)- LTV predictions,Predictive analytics for customer behavior,Ad targeting

and many more fields - https://www.topbots.com/enterprise-ai-companies-2020/ , https://venturebeat.com/2020/10/21/the-2020-data-and-ai-landscape/

                                                    HAPPY LEARNING