Coulibaly Zie Mamadou's Projects
Python scripts for atmospheric science written during grad school at the University of Utah (Python 3)
Jupyter Notebook from the "Scheduling Machine Learning Pipelines using Apache Airflow" workshop @ PyData Eindhoven 2019
Getting start with PySpark and MLlib
Implementation of Spark code in Jupyter notebook. Topics include: RDDs and DataFrame, exploratory data analysis (EDA), handling multiple DataFrames, visualization, Machine Learning
All Algorithms implemented in Python
Python Tutorials
Jupyter Notebooks for https://datascience.quantecon.org
Creation of quantum circuits using Qiskit a Python library designed by IBM
The recommender system takes two basic approaches, that is Collaborative Filtering or Content Based Filtering. The first approach is used in this project. Collaborative Filtering arrives at a recommendation that is based on a model of prior user behavior like movie ratings. I have used two datasets - MovieLens and Netflix Prize Data. This filtering technique has two popular approaches - Nearest Neighbor and Latent Factor based method. For the second technique, Sparks Alternating Least Squared implementation, an RDD - based API is used
This Project was made within the scope of the Big Data discipline. One of the objectives of this project is to learn to work with the pyspark library, learning the differences between pyspark (parallel distribution) and pandas. The dataset is a dataset referring to the red wine of Portugal, where it talks about the characteristics of the wine. An exploratory analysis is made to the dataset that includes the descriptions of variables and the creation of graphs, where we observe the correlation between variables. Then, models are created to predict the quality of red wine. Being the models, logistic regression and random forest.
Ridge and Lasso
Open sourced research notebooks by the QuantConnect team.
A curated collection of places where you can learn robotics, algorithms, and other useful tools for aspiring robotics software engineers.
Rocks are a fundamental component of Earth. The automatic identification of rock type in the field would aid geological surveying, education, and automatic mapping. It is a basic part of geological surveying and research, and mineral resources exploration. The automatic identification of rock type in the field would aid geological surveying, education, and automatic mapping. Working conditions in the field generally limit identification to visual methods, including using a magnifying glass for fine-grained rocks. Visual inspection assesses properties such as colour, composition, grain size, and structure. The attributes of rocks reflect their mineral and chemical composition, formation environment, and genesis. The colour of rock reflects its chemical composition. But these analysis is time taken process to identify the rocks.Its application here has effectively identified rock types from images captured in the field. This paper proposes an accurate approach for identifying rock types in the field based on image analysis using deep convolutional neural networks. Solution: Deep learning is receiving significant research attention for pattern recognition and machine learning. Its application here has effectively identified rock types from images captured in the field. This paper proposes an accurate approach for identifying rock types in the field based on image analysis using deep convolutional neural networks. The results show that the proposed approach based on deep learning represents an improvement in intelligent rock-type identification and solves several difficulties facing the automated identification of rock types in the field.Who are experienced in the field of geological they can identify the rocks easily. But who are new to the field, it can help to identify the type of rock.
Raspberry Pi code to collect data from temp sensors and send telegram alerts
This Python code will analyse the quality of field from 6 years data of MSAVI, NDVI, NDWI indices. We have designed Earth Engine Application to download spectrum data in the format of csv. Python code snippet will import that data and will calculate field parameters and it will predict the quality of field. For that we have implemented K-means clustering algorithm.
Resources for deep learning with satellite & aerial imagery
Work on predicting NVDI score with radar images (Sentinel-1 and Sentinel-2 images)
A garden for scikit-learn compatible trees
Mesure du ROI publicitaire Facebook Une introduction Ă la plateforme publicitaire Facebook Avec les rĂ©seaux de recherche et dâaffichage de Google, Facebook est lâun des grands acteurs quand il sâagit de la publicitĂ© en ligne. Comme les utilisateurs de Facebook interagissent avec la plate-forme, en ajoutant des informations dĂ©mographiques, en aimant des pages particuliĂšres et en commentant des publications spĂ©cifiques, Facebook construit un profil de cet utilisateur en fonction de qui ils sont et de ce quâils sont intĂ©ressĂ©s. Ce fait rend Facebook trĂšs attrayant pour les annonceurs. Les annonceurs peuvent crĂ©er des publicitĂ©s Facebook, puis crĂ©er un «auditoire» pour cette publicitĂ© ou groupe dâannonces. Les audiences peuvent ĂȘtre construites Ă partir dâun Ă©ventail dâattributs, y compris le sexe, lâĂąge, lâemplacement et les intĂ©rĂȘts. Ce ciblage spĂ©cifique signifie que les annonceurs peuvent adapter le contenu de maniĂšre appropriĂ©e pour un public spĂ©cifique, mĂȘme si le produit commercialisĂ© est le mĂȘme.Imaginons, par exemple, quâune entreprise veuille annoncer sa nouvelle voiture. Ils souhaitera peut-ĂȘtre promouvoir un ensemble de caractĂ©ristiques, de performances et de 2 kW stĂ©rĂ©o, pour les femmes dans leur dĂ©but de la vingtaine. Ils pourraient dĂ©cider quâils veulent parler de son efficacitĂ© Ă©nergĂ©tique et de rĂ©duire les Emmissions aux hommes dans la trentaine, et ils pourraient vouloir pousser lâintĂ©rieur spacieux et la note de sĂ©curitĂ© pour les hommes et les femmes dans les annĂ©es trente et les premiĂšres quarantaine qui sont intĂ©ressĂ©s par les familles magazine et qui aiment les pages de couches et les fabricants de vĂȘtements pour bĂ©bĂ©s.De quoi avons-nous besoin de notre analyse des publicitĂ©s Facebook? Lorsqu'il s'agit d'analyser l'ensemble de donnĂ©es sur les publicitĂ©s Facebook, nous pouvons poser beaucoup de questions et gĂ©nĂ©rer beaucoup d'informations. Toutefois, dâun point de vue commercial, nous souhaitons poser des questions qui nous apporteront des rĂ©ponses que nous pourrons utiliser pour amĂ©liorer les performances commerciales. Sans rien connaĂźtre de la stratĂ©gie marketing de l'entreprise ni des objectifs de sa campagne, nous ne savons pas quels indicateurs de performance clĂ©s (KPI) sont les plus importants. Par exemple, une nouvelle sociĂ©tĂ© peut se concentrer sur la notoriĂ©tĂ© de la marque et vouloir maximiser le nombre d'impressions, en se prĂ©occupant moins de la performance de ces annonces en termes de gĂ©nĂ©ration de clics et de revenus. Une autre entreprise peut simplement vouloir maximiser le montant de ses revenus tout en minimisant ses dĂ©penses en publicitĂ©. Ces deux objectifs Ă©tant trĂšs diffĂ©rents, il est important de travailler avec le client pour comprendre exactement ce quâils espĂšrent obtenir de leurs campagnes marketing avant de commencer toute analyse, afin de garantir la pertinence de nos conclusions et, en particulier, leur mise en oeuvre. Il ne sert Ă rien de produire un rapport riche en informations, si le client ne peut rien y faire. Commençons par importer les donnĂ©es, jetons un coup d'Ćil et analysons des analyses qui devraient ĂȘtre pertinentes pour une sĂ©rie d'objectifs.
Exploring sentinel hub python library to extract satellite images and NDVI values
Easy interactive web applications with R
A detailed research and study on sci-kit learn library in python
Capacitive soil moisture sensor calibration with Arduino and Python