Utkarsh Mital

Earth Scientist @ArvaIntel. Formerly @BerkeleyLab. Fan of @Arsenal. Cryptic crossword enthusiast. @Caltech and @VJTI alum.

[Arsenal @Arsenal]

We’re proud to become the only football club in the world to have a net-zero target approved by the globally respected Science Based Targets initiative (SBTi).

Read all the details here 👇

[Communications Earth & Environment @CommsEarth]

🚨Estimating the #snowpack in the Western USA under a warming #climate will require models that incorporate open-source satellite data not just measurements.

🔗https://t.co/ucDA3GZ40l

@mariannecowherd @NatureAtCal @eesalbnl @doescience @armnewsteam @UCB_CSSL @utmital

[Utkarsh Mital @utmital]

@ArsenalVPodcast Do you think Ben White takes too long to take throw-ins? After the Tomiyasu red, I feel paranoid whenever an Arsenal player hangs on to the ball too long.

[Utkarsh Mital @utmital]

@AFCAMDEN @Arsenal Lord Bendtner.

[Utkarsh Mital @utmital]

@celestelabedz If the null hypothesis was that both AM/PM are equally likely, you would fail to reject it!

[Utkarsh Mital @utmital]

@TeagWall That’s so stupid. Can you check with your pediatrician and/or try a different pharmacy?

[Dipankar @DipankarDw08]

https://t.co/s7uoaGoD3J

@ChaopengShen
@utmital
@ecohydraulics
@reza_soltanian
@kcc_carroll
@Tirthankar__Roy
@DebasMishra
@alisonappling
@jesusgomezvelez
@pkumar3691
@Leilaher_hydro
@YuanH_Wang
#PNNL

[Dipankar @DipankarDw08]

Now registration is open for the HydroML Symposium, which is hosted by Lawrence Berkeley National Laboratory.
The symposium will take place from May 22, 2023 – May 24, 2023
https://t.co/jVW2D6ADCs
@ChaopengShen @eesalbnl @BerkeleyLab

[Dipankar @DipankarDw08]

We are pleased to share that the abstract submission for HydroML 2023 is now open! Abstracts are due March 24, 2023. Please help spread the word!

Abstract submission link:
https://t.co/0r4ZSl7wJK

For more information:
https://t.co/8wSqCCC5f0

[Dipankar @DipankarDw08]

We are pleased to announce that Berkeley Lab will host the HydroML 2023 Symposium on May 22-24, 2023.
https://t.co/PDiIevhcpq
This will be the second Symposium in the series of HydroML symposia. The first HydroML Symposium was held on May 18-20, 2022, at PSU.

[Utkarsh Mital @utmital]

@arseblog @gunnerblog #arsecastextra Do you feel that Ramsdale struggles a bit while defending corners? I get the impression he tends to hang back a bit rather than trying to come out and punch the ball. He tried against united which led to their equalizer. Turner looked more assured against city.

[Santiago @svpino]

Apple just released their MacBook Pro laptops with their M2 chip.

Their GPU is extremely fast, yet many people have no idea how to use it to run their machine learning models.

Here is how you allow TensorFlow to use your GPU:

[Utkarsh Mital @utmital]

New paper! We used machine learning to develop a new dataset that estimates high-resolution daily precipitation and temperature in Colorado for 2008-2019. @eesalbnl #MachineLearning

[Earth Syst Sci Data @ESSD_journal]

New data description paper: Downscaled hyper-resolution (400 m) gridded datasets of daily precipitation and temperature (2008–2019) for the East–Taylor subbasin (western United States) https://t.co/wnW2t3FzTv

[Utkarsh Mital @utmital]

We developed a new framework to model snowpack in the mountains of Colorado. Paper out now👇🏼

[Utkarsh Mital @utmital]

We modeled snowpack using widely available data such as precipitation, temperature, surface reflectance, elevation, and canopy.

[Utkarsh Mital @utmital]

New paper alert!
We developed a new machine learning framework to model snowpack in the mountains of Colorado!
Early online release here: https://t.co/ayVz6H1hjE

@AMS_AIES @eesalbnl
#MachineLearning #Snow #Hydrology

[Utkarsh Mital @utmital]

Our framework leverages high-resolution snowpack estimates generated via lidar observations by @airbornesnow, and can estimate snowpack across a basin when lidar data is not available.

[Utkarsh Mital @utmital]

@ChaopengShen 🖐🏼

[Mark Tenenholtz @marktenenholtz]

Feature engineering is the most important part of building great models for tabular data.

However, it’s easy to run out of ideas.

Much like Writer’s Block, I call this Feature Engineering Block.

So here are a bunch of ideas to make sure you never run out: