756 matching collectives
Zest
About
Nous développons des sites & applications en utilisant des pratiques de conception et de développement durables.
4 missions successfully executed on Collective
Zebra3
About
Power Back to small Businesses
4 missions successfully executed on Collective
Kuartz
About
When polished design meet effective copywriting to boost conversion
4 missions successfully executed on Collective
Ace Studio Webflow & Bubble
About
La réalisation de votre site web et app de A à Z par des pointes du No Code ayant travaillé pour des Startups à succès.
4 missions successfully executed on Collective
Itxasoa
About
On a monté et boosté les équipes marketing & design des plus belles start-up ! Notre créativité pour atteindre l'objectif n'a pas de limites !
4 missions successfully executed on Collective
Leverage | Growth Collective
About
Collectif d'experts en Growth pour entreprises ambitieuses. Transparence, efficacité et performance.
4 missions successfully executed on Collective
Airmasters.io
About
Votre partenaire Airtable & automatisation.
4 missions successfully executed on Collective
Growth Advisors.
About
Générez de nouvelles opportunités grâce à l'outbound marketing !
3 missions successfully executed on Collective
Neptune
About
Un collectif de professionnels passionnés pour des vidéos créatives et impactantes !
3 missions successfully executed on Collective
We Are Motion
About
Le collectif spécialisé dans la production de motion design 🚀.
3 missions successfully executed on Collective
Collectif 201
About
Parce qu'on ne peut jamais être deux sans un.
3 missions successfully executed on Collective
Ahoy !
About
Un archipel de profils créatifs et techniques qui mutualisent leurs talents et leurs expertises pour vous offrir LE site web dont vous rêvez.
3 missions successfully executed on Collective
meeXers
About
Intelligence Applicative ✨ Transcription & Synthèse
3 missions successfully executed on Collective
Horizon
About
Éclairer, innover, transformer - Product, Design et Data
3 missions successfully executed on Collective
Magician
About
Mettez un coup de baguette 🪄 dans vos data ! 🧙
3 missions successfully executed on Collective
Moonimize
About
Moonimize vous propose d'élaborer des solutions personnalisées, évolutives et flexibles pour soutenir votre croissance, tout en considérant vos défis commerciaux.
3 missions successfully executed on Collective
Collectif Wild & Web
About
Création de sites internet, référencement, refontes
3 missions successfully executed on Collective
Koojira Studio
About
Passez de 0 à 1. Une team d'experts pour créer votre MVP et le faire évoluer.
3 missions successfully executed on Collective
Ile-de-France Mobilités
•Transport, Logistics & Supply Chain•Enterprise
Deep learning generative model to predict use of public transformation networks
Use of Ile-de-France Mobilités data to predict frequentation of a day with the first few fifteen minutes
With the rise of digitalized subway cards, an huge amount of data is created and recents attempts to eploit this data have proven quite successfull. The goal of this project was two fold : first obtain a deep representation of the different subway stations of the networks. Second to be able to detect very early in a day perturbations in the network.
The first goal aims at obtaining representation of station that can then be used in other deep learning model to achieve other goals. This is why the unbiased nature of the representation is of the upmost importance.
By nature, the timeseries of frequentation of the subway networks are extremely regular with a week day/week-end period. However, some exterior factor (such as weather or public events) can impact this regularity. The goal of the project was to use the first hour of frequentation in the subway to predict how the network would be impacted for the day, allowing agents to adapt early to the situation.
To achieve these goals we first used a disentanglement deep learning model for generation of time series. By verifying with historical data that the generated time series were correct, we were able to confirm the solidity of our representations. We then used some clustering algorithms to enhances these representation and remove some bias. These representation were then used in another model designed to predict the frequentation of an entire day for the first few 15 minutes. The second model, which uses recent advances in forecasting, disentanglement and reccurent networks, was able to detect as early as 45 minutes into the day wether the day would be regular or not.
Members
Expertises
From
Apr 2019 to Apr 2021