Atualmente a tomaticultura brasileira ostenta o título de cultivo com o maior custo de produção médio no país. Dentre algumas alternativas a fim de minimizar os elevados gastos está o uso de híbridos de crescimento semideterminado em sistema de meia-estaca, devido a menor necessidade de podas, insumos e mão de obra. Realizaram-se dois experimentos, em condições de campo aberto, com os objetivos de (a) determinar o efeito de diferentes sistemas de poda (P1, P2 e P3) e distintos espaçamentos entre plantas (E1 e E2) sobre a produção e seus componentes para o híbrido Silvety, de hábito de crescimento semideterminado do segmento Salada Longa Vida (SLV) (Experimento 1); e (b) avaliar o efeito de distintos sistemas de poda (P1 e P2) sobre a produção e seus componentes para os híbridos Silvety, Sagatan e H003, de hábito de crescimento semideterminado do segmento SLV (Experimento 2). Em ambos os experimentos adotou-se o delineamento em blocos ao acaso com os tratamentos em esquema fatorial 3 x 2 e quatro repetições. O experimento 1 foi conduzido em Jacuí, MG, de maio de 2015 a novembro de 2015, já o experimento 2 foi conduzido em Holambra, SP, de fevereiro de 2016 a agosto de 2016. Avaliaram-se o número total de frutos (NTF), a produção total por parcela (PTP), o número de frutos comercializáveis (NFC), a produção comercial (PCO), a produção por planta (PPP), a massa média dos frutos (MMF), o número de frutos por classe AAA, AA e A (NAAA, NAA e NA), a massa de frutos por classe AAA, AA e A (MAAA, MAA e MA), o número de hastes (NH) o número de inflorescências (NI), o número de frutos por inflorescência (NFI), o número de frutos com rachaduras cuticulares (CUT), o número de frutos ocos (OCO) e o número de frutos manchados (MAN). Nas condições em que os experimentos foram conduzidos e de acordo com os resultados obtidos conclui-se que no experimento 1, a poda P2e o espaçamento E1 mostraram-se os mais indicados para o híbrido Silvety. No experimento 2, novamente, a poda P2 mostrou ser a mais recomendada para os híbridos Silvety, Sagatan e H003. / Currently, the Brazilian tomato crop holds the title with the highest average cost of production in the country. Among the alternatives to minimize the high expenses, is the use of hybrids with semi-determinate growth habit in a training system of \"half-stakes\", due to the less need of pruning, inputs and labor. Two experiments were carried out under open field conditions, aiming to (a) to determine the effect of different pruning system (P1, P2 and P3) and spacing between plants (E1 and E2) on the production and its components for the Silvety hybrid, of semi-determinate growth habit of varietal segment Salad Extended Shelf Life (SLV) (Experiment 1), and (b) to evaluate the effect of pruning system (P1 and P2) on the production and its components for the Silvety, Sagatan and H003 hybrids, of semi- determinate growth habit of segment SLV (Experiment 2). In both experiments it was adopted the randomized block design with the treatments in a 3 x 2 factorial scheme, and four replications. Experiment 1 was conducted in Jacuí, MG, from May 2015 to November 2015, and Experiment 2 was conducted in Holambra, SP, from February 2016 to August 2016. There were assessed the total number of fruits (NTF), the total production per plot (PTP), the number of commercial fruits (NFC), the commercial production (PCO), the production per plant (PPP), the average mass of fruits (MMF), the number of fruits per classes AAA,AA and A (NAAA, NAA and NA), the mass of fruits per classes AAA, AA and A (MAAA, MAA and MA), the number of stems (NH), the number of clusters (NI), number of fruits per cluster (NFI), number of fruits with micro-cracking (CUT), number of hollow fruits (OCO), number of fruits with blotchy (MAN). Under the conditions that the trials were conducted and according to the results obtained, it was concluded that in experiment 1, the pruning system P2 and E1 spacing were the most suitable for Silvety hybrid. In experiment 2, again, the pruning system P2 showed to be the most recommended for Silvety, Sagatan and H003 hybrids.

The effect of source-sink relationships on the performance of tomato and hot pepper was investigated in glasshouse experiments by pruning tomato trusses and hot pepper fruit from plants trained to a single stem. The objectives were to characterize the effect of time, method and intensity of pruning on the yield and quality of tomato and hot pepper. Pruning at anthesis as compared to pruning at fruit-set had little effect on yield and fruit quality of both crops. Yield per truss increased steadily with intensity of pruning in tomato, due to increase in fruit size and fruit number per truss. Thus, total yield was not affected by pruning. In hot pepper fruit size increased with intensity of pruning but total yield was significantly reduced and total plant dry mass depressed at higher source : sink ratios (two and three fruit pruned out of a total of six). Occurrence of fruit disorders such as blossom-end rot and fruit cracking increased with increasing source: sink ratio. Pruning of one truss in tomato and one fruit in hot pepper gave the best fruit quality in terms of fruit size, pericarp thickness and freedom from defects, without decreasing total and marketable yield. Removing a middle truss of tomato (third truss) gave the highest yield as compared to removing the youngest truss (sixth truss) or the oldest truss (first truss). The yield increase (relative to the control) in the remaining individual trusses tended to decrease with increasing distance of the trusses from the pruned truss. In hot pepper removal of the youngest fruit (sixth fruit) resulted in the highest yield in comparison to removal of the middle fruit (third fruit) and the oldest fruit (first fruit). The quantity of yield increase in the remaining individual fruits had no consistent trend regarding the relative distance of the fruits from the pruned fruit. Two pruning methods were tried on hot pepper and tomato to compare yield and fruit quality. The first method involved pruning of the first three consecutive trusses of tomato and the first three consecutive fruit of hot pepper out of a total of six. The second method involved pruning of three alternating tomato trusses or hot pepper fruit. Yield and yield components did not differ significantly for the two methods, but pruning alternate trusses of tomato and fruit of hot pepper reduced occurrence of fruit disorders. / Dissertation (MSc (Agric) Agronomy)--University of Pretoria, 2007. / Plant Production and Soil Science / unrestricted

No description available.

Orientador: Sarita Leonel / Banca: Regina Marta Evangelista / Banca: Luís Eduardo Correa Antunes / Resumo: O pessegueiro (Prunus persica L. Bastch) é uma fruteira de clima temperado, pertencente à família das rosáceas. É uma planta de origem chinesa, sendo hoje cultivada em diversos países, inclusive em áreas subtropicais. Destacam-se como grandes produtores mundiais desta frutífera a China, EUA, Itália e Espanha. O pessegueiro, assim como outras frutíferas, necessita de tratos culturais importantes como a poda e o uso de técnicas para a quebra da dormência em cultivares utilizados em regiões subtropicais, já que nessas regiões, a planta não tem suas exigências em frio totalmente satisfeitas. Neste contexto, o presente trabalho teve como objetivo avaliar o desempenho agronômico e a qualidade dos frutos de 3 cultivares de pessegueiro, submetidas à diferentes épocas de poda de frutificação. O experimento foi realizado na Fazenda Experimental Lageado, da Faculdade de Ciências Agronômicas da UNESP de Botucatu/SP, localizada nas seguintes coordenadas geográficas: 22° 51‟ 55" S e 48° 26‟ 22" O e a 810 m de altitude, onde foram avaliados pessegueiros com 2 anos de idade, cultivados no espaçamento de 6,0 x 4,0 m. O delineamento foi em parcelas subdivididas com 4 blocos, onde as parcelas corresponderam aos tratamentos cultivares (Douradão, BRS Kampai e BRS Rubimel) e as subparcelas às épocas de poda (maio, junho, julho e agosto). Foram empregadas 10 plantas por parcela experimental, onde as 4 centrais foram consideradas úteis e as demais serviram como bordadura. As épocas de poda em junho e julho apresentaram os melhores resultados para porcentagem de fixação de frutos, com melhor resultado para „BRS Rubimel‟ podado em junho, com 44,96 %. A cultivar BRS Rubimel podada em julho, obteve a maior produção, 18,70 kg planta-1. A poda realizada em maio antecipou a colheita da cultivar BRS Rubimel em 13 dias, e, a poda... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The peach tree (Prunus persica L. Bastch) is a temperate zone fruit, belonging to the family Rosaceae. Is a plant of Chinese origin, being now cultivated in many countries, including in subtropical zone. Stand out as major world producers of peach China, USA, Italy and Spain. The peach tree, as others fruits, require cultural practices such as pruning and using techniques to break dormancy in cultivars in subtropical regions, since these regions, the peach tree not has its demands in cold fully satisfied. In this context, the present study aimed to evaluate the agronomic performance and fruit quality of three cultivars, submitted to different pruning times production. The experiment was conducted at the Experimental Farm Lageado of Faculty of Agricultural Sciences of UNESP at Botucatu/SP, located at the following coordinates: 22° 51‟ 55" S e 48° 26‟ 22" O e a 810 m height, where peach trees were evaluated with 2 years old, grown at a spacing of 6.0 x 4.0 meters. The experimental design was a split plot design with four blocks, where the plots corresponded to treatments cultivars (Douradão, BRS Kampai and BRS Rubimel) and subplots to pruning times (May, June, July and August). Were used 10 plants per plot, where the 4 central were considered useful and others served as margin. The pruning times in June and July showed the best results for percentage fruit set, with best results for „BRS Rubimel‟ pruned in June, with 44,96 %. The cultivar BRS Rubimel pruned in July, had the highest production, 18,7 kg plant-1. The pruning in may anticipated harvest cultivar BRS Rubimel in 13 days, and pruning carried out in july and August, provided a late harvest for cultivars Douradão and BRS Kampai for the state of São Paulo. The cultivar BRS Rubimel presented the highest values of average mass, diameter and lengh of th... (Complete abstract click electronic access below) / Mestre

Blockchain's use in Vehicular Ad-Hoc Networks (VANETs) research demonstrates that the technology provides useful attributes to allow for the safe and secure operation of VANET applications. The growth of blockchain applications pose a threat to the efficient operation of the MANET-like environment found within VANETs. Floating Genesis Blocks (FGB) can be used to preserve the state of the blockchain up to a certain point, and allows for the safe pruning of the chain without information loss. The early work presented here demonstrates two pruning techniques and compares the effects of each blockchain through simulation measurement of the chain's space requirements. A discussion on the results and recommendations for future work conclude the author's work.

Recent advances in computer vision tasks are mainly due to the success of large deep neural networks. The current state-of-the-art models have high computational costs during inference and suffer from a high memory footprint. Therefore, deploying these large networks on edge devices remains a serious concern. Furthermore, training these over-parameterized networks is computationally expensive and requires a longer training time. Thus, there is a demand to develop techniques that can efficiently reduce training costs and also be able to deploy neural networks on mobile and embedded devices. This dissertation presents practices like designing a lightweight network architecture and increasing network resource utilization. These solutions improve the efficiency of large networks during training and inference. We first propose an efficient micro-architecture (slim modules) to construct a light-weight Slim-CNN to predicting face attributes. Slim modules uses depthwise separable convolutions with pointwise convolutions, making them computationally efficient for embedded applications. Next, we investigate the problem of obtaining a compact pruned model from an untrained original network in a single-stage process. We introduce our RAPID framework that distills knowledge to a pruned student model from a teacher model under online settings. Next, we analyze the phenomena of inactive channels in a trained neural network. We take a deep dive into the gradient updates of these channels and discover that these channels have no weight update after a few early epochs. Thus, we present our channel regeneration technique that reinitializes batch normalization gamma values of all inactive channels. The gradient updates of these channels improve after the regeneration step, resulting in an increase in the contribution of these channels to the network performance. Finally, we introduce a method to improve computational efficiency in pre-trained vision transformers by reducing redundancy in visual data. Our method selects image windows or regions with high objectness measures, as these regions may contain an object of any class. Across all works in this dissertation, we extensively evaluate our proposed methods and demonstrate that our techniques improve the computational efficiency of deep neural networks during training and inference.

Volunteer pruning programs are becoming an important tool in helping municipal arborists manage the urban forest. To find if volunteers can learn to prune trees well, the ability of volunteers to prune small trees after receiving training was assessed in three different ways, a written exam, a pruning prescription assessment, and a pruning cut assessment. Volunteers were assigned to either an indoor, lecture-based training or an outdoor, hands-on training session. After the training volunteers were asked to complete a written exam to gauge their understanding of the curriculum. Volunteers were then asked to perform a pruning prescription on small street trees indicating where they would prune to remove structural issues and branches growing into signs and walkways. Finally, volunteers were asked to make a removal cut and reduction cut using hand pruners. Volunteers who received indoor training performed significantly better on the written exam. Both training groups performed well on the pruning prescription assessment. At each location, volunteers who received the outdoor training scored significantly higher than the indoor group on their pruning prescription. The type of training received did not affect a volunteer’s ability to make good pruning cuts. These results indicate that, given proper training, volunteers can learn to prune small trees to the satisfaction of an arborist.

Federated Learning (FL) is a sub-domain of machine learning (ML) that enforces privacy by allowing the user's local data to reside on their device. Instead of having users send their personal data to a server where the model resides, FL flips the paradigm and brings the model to the user's device for training. Existing works share model parameters or use distillation principles to address the challenges of data heterogeneity. However, these methods ignore some of the other fundamental challenges in FL: device heterogeneity and communication efficiency. In practice, client devices in FL differ greatly in their computational power and communication resources. This is exacerbated by unbalanced data distribution, resulting in an overall increase in training times and the consumption of more bandwidth. In this work, we present a novel approach for resource-efficient FL called emph{REFT} with variable pruning and knowledge distillation techniques to address the computational and communication challenges faced by resource-constrained devices. Our variable pruning technique is designed to reduce computational overhead and increase resource utilization for clients by adapting the pruning process to their individual computational capabilities. Furthermore, to minimize bandwidth consumption and reduce the number of back-and-forth communications between the clients and the server, we leverage knowledge distillation to create an ensemble of client models and distill their collective knowledge to the server. Our experimental results on image classification tasks demonstrate the effectiveness of our approach in conducting FL in a resource-constrained environment. We achieve this by training Deep Neural Network (DNN) models while optimizing resource utilization at each client. Additionally, our method allows for minimal bandwidth consumption and a diverse range of client architectures while maintaining performance and data privacy. / Master of Science / In a world driven by data, preserving privacy while leveraging the power of machine learning (ML) is a critical challenge. Traditional approaches often require sharing personal data with central servers, raising concerns about data privacy. Federated Learning (FL), is a cutting-edge solution that turns this paradigm on its head. FL brings the machine learning model to your device, allowing it to learn from your data without ever leaving your device. While FL holds great promise, it faces its own set of challenges. Existing research has largely focused on making FL work with different types of data, but there are still other issues to be resolved. Our work introduces a novel approach called REFT that addresses two critical challenges in FL: making it work smoothly on devices with varying levels of computing power and reducing the amount of data that needs to be transferred during the learning process. Imagine your smartphone and your laptop. They all have different levels of computing power. REFT adapts the learning process to each device's capabilities using a proposed technique called Variable Pruning. Think of it as a personalized fitness trainer, tailoring the workout to your specific fitness level. Additionally, we've adopted a technique called knowledge distillation. It's like a student learning from a teacher, where the teacher shares only the most critical information. In our case, this reduces the amount of data that needs to be sent across the internet, saving bandwidth and making FL more efficient. Our experiments, which involved training machines to recognize images, demonstrate that REFT works well, even on devices with limited resources. It's a step forward in ensuring your data stays private while still making machine learning smarter and more accessible.

We investigate the coherent gradient hypothesis and show that the coherence measurements are different on real and random data regardless of the network's initialization. We introduce "diffs," an attempt at an element-wise approximation at coherence, and investigate their properties. We study how coherence is affected by increasing the width of simple fully-connected networks. We then prune those fully-connected networks and find that sparse networks outperform dense networks with the same number of nonzero parameters. In addition, we show that it is possible to increase the performance of a sparse network by scaling the size of the dense parent network it is derived from. Finally we apply our pruning methods to ResNet50 and ViT and find that diff-based pruning can be competitive with other methods.

Crotches were cut out of red maple (Acer rubrum), callery pear (Pyrus calleryana), and sawtooth oak (Quercus acutissima) trees (2.5â -7â d.b.h.) and then pulled apart in an engineering testing machine to identify physical parameters correlated with crotch strength. Parameters measured included the diameter of the branch and of the trunk above and below the crotch, angle of the branch and branch bark ridge, and the length of the crotch and the branch bark ridge. The force required to break each sample was used to calculate breaking strength based on the formula for bending stress. Each parameter was tested for correlation with crotch strength within the individual species and for the three species combined. The ratio of branch diameter over crotch width had the highest correlation coefficient for crotch strength. Branch angle was also correlated with crotch strength but not as highly as the ratio of the diameters. V-shaped crotches (those with included bark) were significantly weaker than U-shaped crotches for all species. The ratio of the two stem diameters greatly influenced the manner in which the crotches broke. In crotches where the branch diameter was 2/3 the size of the trunk or smaller, the crotch broke by being pulled directly out of the trunk. Crotches with branches more than 2/3 the diameter of the trunk broke when the trunk split longitudinally and had significantly lower strength values. These results indicate that increased crotch strength results from a small branch diameter relative to that of the trunk. / Master of Science