I am a big fan of small data for big discovery in #DataScience:
https://mapr.com/blog/when-big-data-goes-local-small-data-gets-big-part-2/

———
But #DeepLearning needs large labeled training sets of #BigData:
https://hackernoon.com/%EF%B8%8F-big-challenge-in-deep-learning-training-data-31a88b97b282

———

Human-Machine Collaborative Annotation with #MachineLearning can help:
https://onlinelibrary.wiley.com/doi/full/10.1002/bult.2013.1720390414

✴️ @AI_Python_EN

Getting System Information in Linux using Python Script.
#BigData #Analytics #DataScience #IoT #PyTorch #Python #RStats #TensorFlow #DataScientist #Linux

http://bit.ly/2X56cZa

✴️ @AI_Python_EN

The latest Machine Learning Daily!
https://paper.li/MachineCoding/1370791453?edition_id=51a63f80-99b9-11e9-a7d8-0cc47a0d15fd

Skill from Video data + motion reconstruction
code: https://github.com/akanazawa/motion_reconstruction
CVPR'19 human motion paper training
code: https://github.com/akanazawa/human_dynamics

✴️ @AI_Python_EN

StarAi: FREE Deep Reinforcement Learning Course

https://www.starai.io/course/

✴️ @AI_Python_EN

By the creator of Keras.

#machinelearning #datascience #stats #math

✴️ @AI_Python_EN

All you need to know about Classification and Regression (Machine Learning in 270 pages)

In classification problems, we are trying to predict a discrete number of values. The labels(y) generally comes in the categorical form and represents a finite number of classes.

+Decision Trees
+Logistic Regression
+Naive Bayes
+K Nearest Neighbors
+Linear SVC (Support vector Classifier)s.

In regression problems, we try to predict continuous valued output, take this example. Given the size of the house predict the price(real value).

+Regression Algorithms
+Linear Regression
+Regression Trees(e.g. Random Forest)
+Support Vector Regression (SVR)

✴️ @AI_Python_EN

Does text recognition stuff excite you?
Or do you use google translate which makes you think 'Damn,this is magic!'

This is what I used to feel but after research and #positive approach in learning I landed up getting my basics clear for NLP. Sharing the same with you.

Natural Language Processing is no rocket science and a sub-field of computer science and AI that enables computers to understand and process human language.

What is a word?
-a sequence of meaningful characters, words can be silly as well,DUH ._.

And how it goes around??(basicsteps)
-some sequence of steps(just giving you a jiffy,rest well explained in the pdf)
-Tokenization
-Token Normalisation
-Stemming
-Lemmatization

Attaching a beautifuI research paper from arXiv,give it a read and start with your fun journey.
#datascience #nlp
🗣 @AI_Python_arXiv
✴️ @AI_Python_EN

Learn:

- Practical Deep Learning http://course.fast.ai/
- Deep Learning Foundations https://lnkd.in/dhJJYhw
- Computational Linear Algebra https://lnkd.in/e3zAvzF
- Intro Machine Learning http://course.fast.ai/ml

#artificialintelligence #deeplearning #machinelearning

✴️ @AI_Python_EN
🗣 @AI_Python_arXiv

Is this really a paradox as claimed by the authors? Because of small sample sizes, once 2-year data are in why don't we just ignore the individual yearly baseball performance figures?
http://qualitysafety.bmj.com/content/23/9/701

✴️ @AI_Python_EN
🗣 @AI_Python_arXiv

FREE COURSE Intro to TensorFlow for Deep Learning

This course is a practical approach to deep learning for software developers

https://www.udacity.com/course/intro-to-tensorflow-for-deep-learning--ud187

✴️ @AI_Python_EN
🗣 @AI_Python_arXiv

https://www.fast.ai/2019/06/28/course-p2v3

✴️ @AI_Python_EN
🗣 @AI_Python_arXiv

Interesting paper! Tensorflow 2.0 and PyTorch 1.1 already pushed the language to the limits of what it can do. As Julia and Swift mature their support for #deeplearning, we may need to switch
https://buff.ly/320IH76

✴️ @AI_Python_EN
🗣 @AI_Python_arXiv

A Review of “Compound Probabilistic Context-Free Grammars for Grammar Induction”

By Ryan Cotterell

https://lnkd.in/fVVvwud
paper https://lnkd.in/fr-U2vK

#MachineLearning
#NaturalLanguageProcessing #NLP

✴️ @AI_Python_EN

New Google Brain Optimizer Reduces BERT Pre-Training Time From Days to Minutes
http://bit.ly/30tZfDN
#AI #MachineLearning #DeepLearning #DataScience

✴️ @AI_Python_EN

A mathematical theory of semantic development in deep neural networks

https://lnkd.in/ejt9fe6

#MachineLearning #ArtificialIntelligence #Neurons #Cognition

✴️ @AI_Python_EN

This is the reference implementation of Diff2Vec - "Fast Sequence Based Embedding With Diffusion Graphs" (CompleNet 2018). Diff2Vec is a node embedding algorithm which scales up to networks with millions of nodes. It can be used for node classification, node level regression, latent space community detection and link prediction. Enjoy!

https://lnkd.in/dXiy5-U

#technology #machinelearning #datamining #datascience #deeplearning #neuralnetworks #pytorch #tensorflow #diffusion #Algorithms

✴️ @AI_Python_EN

Hey #DeepLearning #AI enthusiast, have you heard of this cool drag & drop AI from #DSAIL lab from MIT?

This is an amazing tool which #managers and #datascience professionals can use instantly!

The researchers evaluated the tool on 300 real-world datasets. Compared to other state-of-the-art #AutoML systems, VDS’ approximations were as accurate, but were generated within seconds, which is much faster than other tools, which operate in minutes to hours.

Next they want to add features like alerts users to potential data bias or errors. For example, to protect patient privacy, sometimes researchers will label medical datasets with patients aged 0 (if they do not know the age) and 200 (if a patient is over 95 years old). But beginners may not recognize such errors, which could completely throw off their analytics.

Here is link to their project Northstar
https://lnkd.in/dmHQugW

Take a look! This is pretty awesome.
#artificialintelligence #automation #autoML #visualization

✴️ @AI_Python_EN

💡 What is the bias-variance trade-off?

Bias refers to an error from an estimator that is too general and does not learn relationships from a data set that would allow it to make better predictions.

Variance refers to error from an estimator being too specific and learning relationships that are specific to the training set but will not generalize to new observations well.

👉 In short, the bias-variance trade-off is a the trade-off between underfitting and overfitting. As you decrease variance, you tend to increase bias. As you decrease bias, you tend to increase variance.

👉 Generally speaking, your goal is to create models that minimize the overall error by careful model selection and tuning to ensure sure there is a balance between bias and variance: general enough to make good predictions on new data but specific enough to pick up as much signal as possible.

#datascience

✴️ @AI_Python_EN

Generating adversarial patches against YOLOv2

Adversarial attacks on machine learning models have seen increasing interest in the past years. By making only subtle changes to the input of a convolutional neural network, the output of the network can be swayed to output a completely different result. The first attacks did this by changing the pixel values of an input image slightly to fool a classifier to output the wrong class.

paper: https://lnkd.in/d5SnGYv

#yolo #deeplearning

✴️ @AI_Python_EN

Swift Core ML 3 implementation of BERT for Question answering

Built Julien Chaumond, Lysandre Debut and Thomas Wolf at Hugging Face: https://lnkd.in/ejJabYh

#machinelearning #naturallanguageprocessing #nlp

✴️ @AI_Python_EN