This study presents a setup of the wind wave forecasting model WAM working over the Baltic Sea in operational mode. The work is a part of the PROZA project financed from the Polish Innovative Economy Programme and the European Regional Development Fund. The modelling results are verified against observations recorded with a directional waverider buoy. The results of the validation studies are presented in this paper. The agreement between the modelled and observed wave data obtained was quite good. The model setup is part of the system allowing for operational decision-making based on atmospheric and sea conditions.In the next stage of the project we are seeking for improvement of the modelling accuracy by taking into account wave-current interactions. To this end a new version of POM with embedded wave sub-model has been implemented. Based on cross-comparisons against observed waverider buoy data and the results obtained with WAM, some adjustments of the parameters controlling the coupled circulation-wave POM were applied.

This paper presents an application of a Convolutional Neural Network as a solution for a task associated with ESENSEI Challenge: Marking Hair Follicles on Microscopic Images. As we show in this paper quality of classification results could be improved not only by changing architecture but also by ensemble networks. In this paper, we present two solutions for the task, the first one based on benchmark convolutional neural network, and the second one, an ensemble of VGG-16 networks. Presented models took first and third places in the final competition leaderboard.

The exponential growth of the internet community has resulted in the production of a vast amount of unstructured data, including web pages, blogs and social media. Such a volume consisting of hundreds of billions of words is unlikely to be analyzed by humans. In this work we introduce the tool LanguageCrawl, which allows Natural Language Processing (NLP) researchers to easily build web-scale corpora using the Common Crawl Archive—an open repository of web crawl information, which contains petabytes of data. We present three use cases in the course of this work: filtering of Polish websites, the construction of n-gram corpora and the training of a continuous skipgram language model with hierarchical softmax. Each of them has been implemented within the LanguageCrawl toolkit, with the possibility to adjust specified language and n-gram ranks. This paper focuses particularly on high computing efficiency by applying highly concurrent multitasking. Our tool utilizes effective libraries and design. LanguageCrawl has been made publicly available to enrich the current set of NLP resources. We strongly believe that our work will facilitate further NLP research, especially in under-resourced languages, in which the lack of appropriately-sized corpora is a serious hindrance to applying data-intensive methods, such as deep neural networks.

No abstract