Webinar on tools for assessing national #climate pledges (NDCs) by @WRIClimate. @davidwaskow reminds: #ParisAgreement is meant to work thru an iterative process of increasing ambition (faster emission reductions), leading to #netzero emissions by 2050. (Thread 1/n)

Great seeing one of the fastest natural drivers of the last decade Robert Kubica today with @WilliamsRacing at @COTA #USGP #f1 #robertkubica #Poland #richenergy

My System for momentum trading🧵:

1. Stock Identification:

Daily and Weekly Delivery Analysis using @mystockedge
Theme Confirmation using Internet and Network

2. Stock Confirmation:

Relative Strength of a stock visavis market using RS55 model listed in @mystockedge

1/n

3. Stock Participation:

Only when chart showing short term breakout along with RS55>0 and RSI>50. I am ready to leave some initial movements.

4. Position buildup:

Every time the stock gives a breakout from the previous swing high add more position and go all-in.

2/n

5. Stock Exit:

I run two portfolios: Short term Swing and Long term Swing.

For Exit in Short term swings, I use a 2hourly Chart with super trend and RS. I am using @tradingview for the same.

For long term swing, I exit when the RS of the stock goes below 0. Period.

n/n

#NationalAuthorsDay Hi! Here's a thread about how I love you for buying my books anywhere, but I especially love when you support independent bookstores. You can get delivery through @indiebound via a store near you. Read on for places I love around the US and beyond...

In Asheville, North Carolina, go buy or order some great books by authors you love (hint, hint) via @Malaprops! #NationalAuthorsDay

In Chapel Hill, NC visit @FlyleafBooks. If they don't have my books in stock, I bet they'll order them for you! This is also true for other authors! #NationalAuthorsDay

In New York and NJ visit @wordbookstores! Ask for my books if they don't have them, and also buy lots of stuff by other authors! #NationalAuthorsDay

In Washington, D.C. visit @PoliticsProse for all your #NationalAuthorsDay needs. Ask for my books and those of hometown hero @baratunde and also buy cute objects!

राम-रावण युद्ध समाप्त हो चुका था। जगत को त्रास देने वाला रावण अपने कुटुम्ब सहित नष्ट हो चुका था।श्रीराम का राज्याभिषेक हुआ और अयोध्या नरेश श्री राम के नेतृत्व में चारों दिशाओं में शन्ति थी।
अंगद को विदा करते समय राम रो पड़े थे ।हनुमान को विदा करने की शक्ति तो राम में थी ही नहीं ।

माता सीता भी हनुमान को पुत्रवत मानती थी। अत: हनुमान अयोध्या में ही रह गए ।राम दिनभर दरबार में, शासन व्यवस्था में व्यस्त रहते थे। संध्या को जब शासकीय कार्यों में छूट मिलती तो गुरु और माताओं का कुशल-मंगल पूछ अपने कक्ष में जाते थे। परंतु हनुमान जी हमेशा उनके पीछे-पीछे ही रहते थे ।

उनकी उपस्थिति में ही सारा परिवार बहुत देर तक जी भर बातें करता ।फिर भरत को ध्यान आया कि भैया-भाभी को भी एकांत मिलना चाहिए ।उर्मिला को देख भी उनके मन में हूक उठती थी कि इस पतिव्रता को भी अपने पति का सानिध्य चाहिए ।

एक दिन भरत ने हनुमान जी से कहा,"हे पवनपुत्र! सीता भाभी को राम भैया के साथ एकांत में रहने का भी अधिकार प्राप्त है ।क्या आपको उनके माथे पर सिन्दूर नहीं दिखता?इसलिए संध्या पश्चात आप राम भैया को कृप्या अकेला छोड़ दिया करें "।
ये सुनकर हनुमान आश्चर्यचकित रह गए और सीता माता के पास गए ।

माता से हनुमान ने पूछा,"माता आप अपने माथे पर सिन्दूर क्यों लगाती हैं।" यह सुनकर सीता माता बोलीं,"स्त्री अपने माथे पर सिन्दूर लगाती है तो उसके पति की आयु में वृद्धि होती है और वह स्वस्थ रहते हैं "। फिर हनुमान जी प्रभु राम के पास गए ।

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises. [thread]

https://t.co/7QIx5CFdfJ

The paper is a good example of lots of elements of good experimental design. They validate their metric by showing lots of variants give consistent results. They tune hyperparamters separately for each condition, check that optimum isn't at the endpoints, and measure sensitivity.

They have separate experiments where the hold fixed # iterations and # epochs, which (as they explain) measure very different things. They avoid confounds, such as batch norm's artificial dependence between batch size and regularization strength.

When the experiments are done carefully enough, the results are remarkably consistent between different datasets and architectures. Qualitatively, MNIST behaves just like ImageNet.

Importantly, they don't find any evidence for a "sharp/flat optima" effect whereby better optimization leads to worse final results. They have a good discussion of experimental artifacts/confounds in past papers where such effects were reported.