Leonardo Ferreira

The final project of the Computational Methods in Social Sciences course with the following objective: You will web scrape two different YouTube channels using: http://webscraper.io with the objectives: Take at least 5 screenshots of different stages of your web scraping progress. all 5+ images. Screenshots (taken on Mac or Windows ) will be in image format (extensions like .JPG, .JPEG, .PNG, .TIFF, .GIF, .BMP, .PDF, etc), the spreadsheet file for your first YouTube channel… Mostrar más

The final project of the Computational Methods in Social Sciences course with the following objective: You will web scrape two different YouTube channels using: http://webscraper.io with the objectives: Take at least 5 screenshots of different stages of your web scraping progress. all 5+ images. Screenshots (taken on Mac or Windows ) will be in image format (extensions like .JPG, .JPEG, .PNG, .TIFF, .GIF, .BMP, .PDF, etc), the spreadsheet file for your first YouTube channel. Spreadsheets will have extensions like .csv (most generic), .xls, .xlsx, .numbers, etc e the spreadsheet file for your second YouTube channel. Mostrar menos

Conclusion project for week 1 of the Seguence Models course of the specialization in Deep Learning. Use of jupyter notebook with numpy, rnn, among others with the objective: implement key components of a Recurrent Neural Network in numpy. Recurrent Neural Networks (RNN) are very effective for Natural Language Processing and other sequence tasks because they have "memory".

Conclusion project for week 1 of the Seguence Models course of the specialization in Deep Learning. Use of jupyter notebook with numpy, rnn, utils among others with the objective: How to store text data for processing using an RNN, How to synthesize data, by sampling predictions at each time step and ing it to the next RNN-cell unit, How to build a character-level text generation recurrent neural network e Why clipping the gradients is important.

Conclusion project for week 2 of the Sequence Models course of the specialization in Deep Learning. Use of jupyter notebook with Numpy, TensorFlow, emoji packages, among others with the objective: In this project, you'll start with a baseline model (Emojifier-V1) using word embeddings e Then you will build a more sophisticated model (Emojifier-V2) that further incorporates an LSTM.

Conclusion project for week 1 of the Sequence Models course of the specialization in Deep Learning. Use of jupyter notebook with numpy, music21, grammar, rnn, music_utils, keras, among others with the objective: Apply an LSTM to music generation e Generate your jazz music with deep learning.

Conclusion project for week 3 of the Sequence Models course of the specialization in Deep Learning. Use of jupyter notebook with Numpy, Keras, faker , babel.dates packages, among others with the objective: You will build a Neural Machine Translation (NMT) model to translate human-readable dates ("25th of June, 2009") into machine-readable dates ("2009-06-25") e You will do this using an attention model, one of the most sophisticated sequence-to-sequence models.

Conclusion project for week 2 of the Sequence Models course of the specialization in Deep Learning. Use of jupyter notebook with Numpy, TensorFlow packages, among others with the objective: Load pre-trained word vectors, and measure similarity using cosine similarity, Use word embeddings to solve word analogy problems such as Man is to Woman as King is to __ e Modify word embeddings to reduce their gender bias.

Conclusion project for week 3 of the Sequence Models course of the specialization in Deep Learning. Use of jupyter notebook with Numpy, Pydub , io, glob packages, among others with the objective: Structure a speech recognition project, Synthesize and process audio recordings to create train/dev datasets e Train a trigger word detection model and make predictions.

Conclusion project for week 4 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, h5py packages, among others with the objective: Develop an intuition of the over all structure of a neural network, Write functions (e.g. forward propagation, backward propagation, logistic loss, etc...) that would help you decompose your code and ease the process of building a neural network e Initialize/update parameters… Mostrar más

Conclusion project for week 4 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, h5py packages, among others with the objective: Develop an intuition of the over all structure of a neural network, Write functions (e.g. forward propagation, backward propagation, logistic loss, etc...) that would help you decompose your code and ease the process of building a neural network e Initialize/update parameters according to your desired structure. Mostrar menos

Conclusion project for week 2 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Learn to use Keras, a high-level neural networks API (programming framework), written in Python and capable of running on… Mostrar más

Conclusion project for week 2 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Learn to use Keras, a high-level neural networks API (programming framework), written in Python and capable of running on top of several lower-level frameworks including TensorFlow and CNTK for built Emotion Detection in Images of Faces. Mostrar menos

Conclusion project for week 1 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: implement convolutional (CONV) and pooling (POOL) layers in numpy, including both forward propagation and backward… Mostrar más

Conclusion project for week 1 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: implement convolutional (CONV) and pooling (POOL) layers in numpy, including both forward propagation and backward propagation. Mostrar menos

Conclusion project for week 1 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement helper functions that you will use when implementing a TensorFlow model e Implement a fully functioning ConvNet… Mostrar más

Conclusion project for week 1 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement helper functions that you will use when implementing a TensorFlow model e Implement a fully functioning ConvNet using TensorFlow. Mostrar menos

Conclusion project for week 2 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with keras, numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement the basic building blocks of ResNets,Put together these building blocks to implement and train a… Mostrar más

Conclusion project for week 2 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with keras, numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement the basic building blocks of ResNets,Put together these building blocks to implement and train a state-of-the-art neural network for image classification. Mostrar menos

Conclusion project for week 3 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with keras, tensorflow, numpy, matplotlib, argparse, pandas, PIL, packages, among others with the objective: learn about object detection using the very powerful YOLO model, Use object detection on a car detection dataset e Deal with bounding boxes.

Conclusion project for week 4 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with tensorflow, numpy, matplotlib, PIL, packages, among others with the objective: Implement the neural style transfer algorithm e generate novel artistic images using your algorithm for built an Art generation with Neural Style Transfer,

Conclusion project for week 4 of the Convolutional Neural Networks course of the specialization in Deep Learning. Use of jupyter notebook with keras, tensorflow, numpy, matplotlib packages, among others with the objective: Implement the triplet loss function, Use a pretrained model to map face images into 128-dimensional encodings. Use these encodings to perform face verification and face recognition for built an Face Recognition.

Conclusion project for week 4 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, h5py, time, PIL, scipy packages, among others with the objective: Learn how to use all the helper functions you built in the previous assignment to build a model of any structure you want, Experiment with different model architectures and see how each one behaves, Recognize that it is always easier to build your helper functions… Mostrar más

Conclusion project for week 4 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, h5py, time, PIL, scipy packages, among others with the objective: Learn how to use all the helper functions you built in the previous assignment to build a model of any structure you want, Experiment with different model architectures and see how each one behaves, Recognize that it is always easier to build your helper functions before attempting to build a neural network from scratch. Mostrar menos

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand that different initialization methods and their impact… Mostrar más

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand that different initialization methods and their impact on your model performance, Implement zero initialization and and see it fails to "break symmetry", Recognize that random initialization "breaks symmetry" and yields more efficient models, Understand that you could use both random initialization and scaling to get even better training performance on your model. Mostrar menos

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand that different regularization methods that could help… Mostrar más

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand that different regularization methods that could help your model,Implement dropout and see it work on data, Recognize that a model without regularization gives you a better accuracy on the training set but nor necessarily on the test set e Understand that you could use both dropout and regularization on your model. Mostrar menos

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement gradient checking from scratch, Understand how to use… Mostrar más

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement gradient checking from scratch, Understand how to use the difference formula to check your backpropagation implementation, Recognize that your backpropagation algorithm should give you similar results as the ones you got by computing the difference formula e Learn how to identify which parameter's gradient was computed incorrectly. Mostrar menos

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement gradient checking from scratch, Understand how to use… Mostrar más

Conclusion project for week 1 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils(sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Implement gradient checking from scratch, Understand how to use the difference formula to check your backpropagation implementation, Recognize that your backpropagation algorithm should give you similar results as the ones you got by computing the difference formula e Learn how to identify which parameter's gradient was computed incorrectly. Mostrar menos

Conclusion project for week 2 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils (sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand the intuition between Adam and RMS prop, Recognize the… Mostrar más

Conclusion project for week 2 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, sklearn, init_utils (sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: Understand the intuition between Adam and RMS prop, Recognize the importance of mini-batch gradient descent e Learn the effects of momentum on the overall performance of your model. Mostrar menos

Conclusion project for week 2 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with numpy, matplotlib, h5py packages, among others with the objective: build the general architecture of a learning algorithm, including: initializing parameters, calculating the cost function and its gradient. Using an optimization algorithm (gradient descent).

Conclusion project for week 3 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with tensorflow, numpy, matplotlib, sklearn, init_utils (sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: After completing this assignment you will also be… Mostrar más

Conclusion project for week 3 of the Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization course of the specialization in Deep Learning. Use of jupyter notebook with tensorflow, numpy, matplotlib, sklearn, init_utils (sigmoid, relu, compute_loss, forward_propagation, backward_propagation, update_parameters, predict, load_dataset, plot_decision_boundary, predict_dec) packages, among others with the objective: After completing this assignment you will also be able to implement your own deep learning models using Tensorflow. In fact, using our brand new SIGNS dataset, you will build a deep neural network model to recognize numbers from 0 to 5 in sign language with a pretty impressive accuracy. Mostrar menos

Conclusion project for week 3 of the Neural Networks and Deep Learning course of the specialization in Deep Learning. Use of jupyter notebook with sklearn, numpy, matplotlib, packages, among others with the objective: Develop an intuition of back-propagation and see it work on data, Recognize that the more hidden layers you have the more complex structure you could capture e Build all the helper functions to implement a full model with one hidden layer.

1. Machine Learning flight simulator.

2. Bird recognition in the city of Peacetopia (case study).

3. Autonomous driving (case study).

Foi criado, por meio da Portaria CNMP-PRESI nº 63, o Núcleo de Ciência de Dados, que faz parte da Secretaria de Gestão Estratégica. O núcleo foi criado no intuito de assessorar as unidades do Conselho nas atividades que envolvam coleta, organização e análise de dados; na elaboração de estudos para orientar discussões e subsidiar o processo de tomada de decisões estratégicas do CNMP; no desenvolvimento de pesquisas e diagnósticos sobre o Ministério Público, em suas diversas áreas temáticas, além… Mostrar más

Foi criado, por meio da Portaria CNMP-PRESI nº 63, o Núcleo de Ciência de Dados, que faz parte da Secretaria de Gestão Estratégica. O núcleo foi criado no intuito de assessorar as unidades do Conselho nas atividades que envolvam coleta, organização e análise de dados; na elaboração de estudos para orientar discussões e subsidiar o processo de tomada de decisões estratégicas do CNMP; no desenvolvimento de pesquisas e diagnósticos sobre o Ministério Público, em suas diversas áreas temáticas, além de disseminar a cultura de dados no âmbito do CNMP. Mostrar menos

Análise Exploratória de Dados conduzida sobre os dados da operação da AirBnb na cidade do Rio de Janeiro. Foi possível ter algumas insights interessantes no projeto.

O Laboratório de Inteligência do Gasto Público - LIGP tem como objetivos no âmbito do ciclo de aquisições governamentais: I - Contribuir para melhoria da qualidade do gasto público; II - Elaborar estudos ligados à temática da qualidade do gasto público; III - Propor medidas de racionalização do gasto público; e IV - Criar e monitorar indicadores de desempenho ligados às medidas propostas de racionalização do gasto público.O Laboratório de Inteligência de Gastos Públicos - LIGP será coordenado… Mostrar más

O Laboratório de Inteligência do Gasto Público - LIGP tem como objetivos no âmbito do ciclo de aquisições governamentais: I - Contribuir para melhoria da qualidade do gasto público; II - Elaborar estudos ligados à temática da qualidade do gasto público; III - Propor medidas de racionalização do gasto público; e IV - Criar e monitorar indicadores de desempenho ligados às medidas propostas de racionalização do gasto público.O Laboratório de Inteligência de Gastos Públicos - LIGP será coordenado pela Subsecretaria de Compras Governamentais - SCG/SAGA, em conjunto com as demais Subsecretarias da SEFP, a quem compete: I - Coletar, organizar e analisar dados ligados ao ciclo de aquisições governamentais; II - Elaborar relatórios e painéis gerenciais ligados ao ciclo de aquisições governamentais; III - Disponibilizar estudos ligados à temática da qualidade do gasto em áreas específicas integrantes do ciclo de aquisições governamentais; IV - Propor integrações e ou modificações de sistemas transacionais no âmbito do ciclo de aquisições governamentais; V - Apresentar portfólio de medidas de racionalização do gasto público no âmbito do ciclo de aquisições governamentais; VI - Criar e monitorar indicadores de desempenho ligados às medidas de racionalização do gasto propostas no âmbito do ciclo de aquisições governamentais; e VII - Dar publicidade aos trabalhos realizados no âmbito do LIGP no Portal de Compras do Distrito Federal..
Situação: Em andamento; Natureza: Desenvolvimento. Mostrar menos

A publicação do Decreto nº 37.729, de 26 de outubro de 2016, pela Subsecretaria de Compras Governamentais (SCG) da Secretaria de Fazenda, Planejamento, Orçamento e Gestão (SEFP), inaugurou uma nova era na gestão das compras públicas no Governo do Distrito Federal (GDF). Institui-se o Programa de Gestão de Compras Governamentais do Distrito Federal - COMPRASDF aplicável às aquisições e à contratação de serviços no âmbito distrital. O referido programa reúne ações que visam modernizar o ciclo de… Mostrar más

A publicação do Decreto nº 37.729, de 26 de outubro de 2016, pela Subsecretaria de Compras Governamentais (SCG) da Secretaria de Fazenda, Planejamento, Orçamento e Gestão (SEFP), inaugurou uma nova era na gestão das compras públicas no Governo do Distrito Federal (GDF). Institui-se o Programa de Gestão de Compras Governamentais do Distrito Federal - COMPRASDF aplicável às aquisições e à contratação de serviços no âmbito distrital. O referido programa reúne ações que visam modernizar o ciclo de compras públicas e aperfeiçoar o gerenciamento da cadeia integrada de suprimentos dos órgãos e entidades do Poder Executivo do Governo do Distrito Federal. Frente ao desafio de modernização, foi instituído o Portal de Compras Governamentais (PCG) do Distrito Federal, desenvolvido em plataforma web, que permitirá o gerenciamento online dos procedimentos licitatórios, da gestão contratual e da gestão de suprimentos dos órgãos e entidades do Distrito Federal. O referido portal é integrado pelos seguintes sistemas corporativos (SC): Sistema de Gestão de Compras Governamentais (e-ComprasDF), Sistema de Gestão de Contratos (e-ContratosDF) e Sistema de Gestão de Suprimentos (e-SupriDF). Focado no desenvolvimento futuro de ciência dos dados, com uso de ferramentas de análise de dados, mineração de dados, aprendizado de máquina, inteligência artificial, algoritmos evolutivos, internet das coisas (IoT), em e ao processo decisório e ao desenho de políticas públicas de aquisições governamentais do Distrito Federal..
Situação: Concluído; Natureza: Desenvolvimento. Mostrar menos