Results 1 - 10 of 109,181

Table 4.a. Cache misses, non-image data (B=16)

in unknown title
by unknown authors
"... In PAGE 5: ... Table4 .... In PAGE 5: ... Table 4 compares the results achieved by simulating unified and split caches. The split cache has been simulated by equally dividing the overall cache space between image and non-image data; different sizes could be adopted, anyway, since the miss rate on non-image data is much lower than that on image references (see Table4 .... In PAGE 5: ...mage references (see Table 4.a). Table 4.b shows that in most cases the split cache allows a less number of misses, and this is especially true for the case of 2D pre-fetch. In addition, Table4 .b shows that 2D pre-fetching performs best for both kind of caches.... In PAGE 5: ...b shows that 2D pre-fetching performs best for both kind of caches. For the unified cache, two pre-fetching strategies have been used: 2D pre-fetch for all data, versus 2D pre-fetch for images and OBL for non-images; Table4 .b shows that the latter technique slightly reduces the number of misses.... ..."

Table 1: Summary of clinical, stress-test, non-imaging, and scintigraphic features and their rep- resentation used to train and test the neural network methods. A continuous representation is indicated by \..thru.. quot; in the value eld.

in The Functional Transfer of Knowledge for Coronary Artery Disease Diagnosis
by Daniel L. Silver, Robert E. Mercer, Gilbert A. Hurwitz 1997
Cited by 2

Table 2 Comparing compression ratios with other binary image coding methods and non-image based methods Test images Binary image coding methods Non- image based methods

in On a Fluency Image Coding System for Beef Marbling Evaluation
by Kazuo Toraichi , Paul Wing Hing Kwan, Kazuki Katagishi , Fumito Yoshikawa, et al.

Table 5: Average percentage of time spent on non image- related activities (+ range of responses) Non image-related activity % of time range

in
by unknown authors

Table 6.1 Technical characteristics of the HUTRAD non-imaging radiometers (Kurvonen and Hallikainen 1996, Hallikainen et al. 1998).

in Investigation of the Microwave Signatures of
by The Baltic Sea Ice, Marko Mäkynen, Teknillinen Korkeakoulu, Marko Mäkynen, Picaset Oy, Instructor Prof, Martti Hallikainen

Table 4.b. Split vs. unified caches, all data (B=16)

in unknown title
by unknown authors
"... In PAGE 5: ... The high probability of inter-block access, greater for the horizontal direction but substantial also for the vertical direction, is the reason for the usefulness of a bidimensional caching scheme, and explains the good performance of a 4 x 16 form factor . Table4 .a.... In PAGE 5: ... Table 4 compares the results achieved by simulating unified and split caches. The split cache has been simulated by equally dividing the overall cache space between image and non-image data; different sizes could be adopted, anyway, since the miss rate on non-image data is much lower than that on image references (see Table4 .... In PAGE 5: ...mage references (see Table 4.a). Table 4.b shows that in most cases the split cache allows a less number of misses, and this is especially true for the case of 2D pre-fetch. In addition, Table4 .b shows that 2D pre-fetching performs best for both kind of caches.... In PAGE 5: ...b shows that 2D pre-fetching performs best for both kind of caches. For the unified cache, two pre-fetching strategies have been used: 2D pre-fetch for all data, versus 2D pre-fetch for images and OBL for non-images; Table4 .b shows that the latter technique slightly reduces the number of misses.... ..."

Table 2 Country representation in the Archive by percentage Country In Archive Not in archive

in A fair history of the Web? Examining country balance
by Mike Thelwall, Liwen Vaughan 2004
Cited by 4

Table 1. Parameters of archival data.

in Infrared and visible coronal lines in NGC1068
by Marconi Van, P. P. Van Der Werf, E. Oliva
"... In PAGE 2: ... 1994) and con- sisted of object, lamp and standard star frames taken with the Intermediate Dispersion Spectrograph (IDS) on the 2.5m Isaac Newton Telescope (INT; for more information see Table1 ). When recorded, slit orientations were in the range PA 37-40 i.... ..."

Table 1. Image data collections.

in Image Classification for Digital Archive Management
by Cheng-hung Li, Chih-yi Chiu, Hsiang-an Wang
"... In PAGE 5: ... Part of each class of images from the digital archive management system. Table1 lists the image collections. There are 2716 collected images (# CI) 385 training images (# TRI), and 2331 testing images (# TI).... ..."

Table 1. Country representation in the Archive. Note: The main number is the number of sites found and the number in brackets is the expected count. Country In Archive Not in archive Total

in A fair history of the Web? Examining country balance
by Mike Thelwall, Liwen Vaughan 2004
Cited by 4
Results 1 - 10 of 109,181