Big Data To Good Data: Andrew Ng Urges ML Community To Be More Data-Centric And Less Model-Centric
Big Data To Good Data: Andrew Ng Urges ML Community To Be More Data-Centric And Less Model-Centric “If 80 percent of our work is data preparation, then ensuring data quality is the important work of a machine learning team.” “If 80 percent of our work is data preparation, then ensuring data quality is the important work of a machine learning team.” Andrew Ng The progress in machine learning progre
Optuna - A hyperparameter optimization framework
A simple optimization problem: Define objective function to be optimized. Let's minimize (x - 2)^2 Suggest hyperparameter values using trial object. Here, a float value of x is suggested from -10 to 10 Create a study object and invoke the optimize method over 100 trials import optuna def objective(trial): x = trial.suggest_float('x', -10, 10) return (x - 2) ** 2 study = optuna.create_study() study
Elasticsearch Optimizations at Lyft
Photo by LyftBy Stefan Zier and Vinay Kakade IntroductionAt Lyft, we use an in-house Feature Service to store batch and streaming features used by ML models, making them accessible in both offline mode (for training) and online mode (for inference). The service replicates these features in Elasticsearch to enable advanced queries. For Growth products, we use Elasticsearch to power internal tools l
Learn Kaggle techniques from Kaggle #1, Owen Zhang
If you know kaggle.com(the biggest data scientists competition platform), you must have known Owen Zhang. He has competed in and won several high profile challenges, and is currently ranked No. 1 out of a community of over 240,000 data scientists on Kaggle. Owen Zhang is the Chief Product Officer at DataRobot. Owen spent most of his career in the Property & Casualty insurance industry. Most recent
Featured Talk: #1 Kaggle Data Scientist Owen Zhang
Many thanks to Jared Hatch from ThoughtWorks for sponsoring us the wonderful space. Many thanks to Owen Zhang for giving us A GREAT TALK! It was certainly full of joy and satisfaction! (Photo: As a tradition of the meetup, we voted for the outcome of the meetup. Thumbs are up in the air means " we are super happy and satisfied"!) =========================================== Original event announcem
An Analysis of Deep Neural Network Models for Practical Applications
Since the emergence of Deep Neural Networks (DNNs) as a prominent technique in the field of computer vision, the ImageNet classification challenge has played a major role in advancing the state-of-the-art. While accuracy figures have steadily increased, the resource utilisation of winning models has not been properly taken into account. In this work, we present a comprehensive analysis of importan
機械学習アルゴリズムのボトルネックをCythonで改善する話 - Qiita
class MFbpr(Recommender): ''' コンストラクタとか他の処理 ''' def buildModel(self): loss_pre = sys.float_info.max nonzeros = self.trainMatrix.nnz hr_prev = 0.0 sys.stderr.write("Run for BPR. \n") for itr in xrange(self.maxIter): start = time.time() # Each training epoch for s in xrange(nonzeros): # sample a user u = np.random.randint(self.userCount) itemList = self.trainMatrix.getrowview(u).rows[0] if len(itemL
直積量子化(Product Quantization)を用いた近似最近傍探索についての簡単な解説
"aka motsu-nabe" by chatani 概要 冬の寒さも一段と厳しくなってまいりました。おでんや鍋が恋しくなる季節です。 さて、最近ようやっと一仕事が終わりまして、長ったらしい記事が書けるようになりました。ですので、今回は2011年にTPAMIで発表された、近似最近傍探索についての論文『Product quantization for nearest neighbor search』について簡単に紹介したいと思います。 この論文は2011年に発表された、最近傍探索アルゴリズムの決定打です。シンプルな理論でありながら既存手法を打ち破るほどの強力な性能を有し、速度も非常に高速、かつ省メモリなのでスマートフォンに載せ、リアルタイムで動作させることも可能です。 以前この手法はCV勉強会@関東で紹介されたらしいのですが、具体的に紹介しているページは(最近すぎるので当たり前ですが)現在
1
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
