Matrix mining
Publish: 2021-04-22 08:05:03
1. Dongmen station
2.
Shopping malls: Central City, Tianhong shopping plaza, cool apple (Central City store), Cocopark, Xinghe · Cocopark, huatiannu digital, JunShang department store, Yijing central city, east gate of Tianhong shopping plaza, west gate of Tianhong shopping plaza
banks: Shenzhen Branch of Ping'an bank, China Merchants Bank (Unicom branch), China Merchants Bank (central business branch) Funan branch of China CITIC Bank, Shenzhen Branch of Pangu bank, Times financial center branch of Bank of China, Huarong branch of Bank of communications, Central District branch of Everbright Bank, ty free building branch of ICBC, Wuzhou branch of Minsheng Bank, Shenzhen Branch of Instrial Bank Shunji restaurant, zuiwengting, Shanhai feast, real pot coffee, Shenzhen investment building, Beihai fishing village restaurant
Hotels: Marco Polo happy days hotel, investment building hotel, Beihai fishing village (Xinhua), Shunji restaurant, Futian Shangri La Hotel, parking lot of Marco Polo happy days hotel, Guohui Hotel, Shenzhen Investment Building Hotel Co., Ltd Daoxiang and Marco Polo Hotel Restaurants
supermarkets: Carrefour Central City store, Watsons (Central City store), Baijia super Plaza International store, Vango 24-hour convenience store (excellence Times Plaza store), 7-11 convenience store (news Road store), berison No.192, coco Park store, wanzhongle, Shenzhen Zhenlong Gift Co., Ltd Sui Bao Department Store (First World Plaza store)
3. Where is Meilin? If there is a subway nearby, take the bus to Futian station and walk about 300 meters to the Times financial center.
4. Shenzhen airport to the Times financial center
can take the airport bus, no traffic jam situation about upstairs two time
If time permits, stay far away, more than 300 yuan.. Taking a taxi to the Times financial center is more cost-effective than staying in a hotel near the Times financial center<
no more than 300 are:
sunflower hotel in Shenzhen 0755 -
address: No. 32, Futian South Road, Futian District, Shenzhen (near Huanggang port Convention and Exhibition Center)
Meiyi Hotel in Shenzhen 0755 -
address: Block C, Huang Plaza, 3008, intersection of Yitian road and Binhe Avenue, Futian District, Shenzhen (south of convention and Exhibition Center)
Shenzhen Yangfu Hotel
0755 - Address: 123 Fuhua Road, Futian District, Shenzhen (entrance of Fuhua road and Futian South Road)
Yinquan Hotel, Shenzhen
0755 - Address: Shawei store, qidouxing Business Hotel, building 11, Bagua 1st Road, Futian District, Shenzhen
address: Century handicraft trade market, Xinzhou South Road, Futian District, Shenzhen Building 315
Shenzhen Difu Hotel
address: 111 Zhenhua Road, Futian District, Shenzhen
Shenzhen 7-day Economy Chain Hotel Shenzhen Huaqiang hotel reservation center Tel: 0755 -
address: No.1 building, Huaqiang South saigeyuan, Futian District, Shenzhen
Shenzhen Guanyin hotel reservation center Tel: 0755 -
address: Futian District, Shenzhen Next to Binhe Avenue, Guangyin building, 38 South Road (Huanggang port)
enough
can take the airport bus, no traffic jam situation about upstairs two time
If time permits, stay far away, more than 300 yuan.. Taking a taxi to the Times financial center is more cost-effective than staying in a hotel near the Times financial center<
no more than 300 are:
sunflower hotel in Shenzhen 0755 -
address: No. 32, Futian South Road, Futian District, Shenzhen (near Huanggang port Convention and Exhibition Center)
Meiyi Hotel in Shenzhen 0755 -
address: Block C, Huang Plaza, 3008, intersection of Yitian road and Binhe Avenue, Futian District, Shenzhen (south of convention and Exhibition Center)
Shenzhen Yangfu Hotel
0755 - Address: 123 Fuhua Road, Futian District, Shenzhen (entrance of Fuhua road and Futian South Road)
Yinquan Hotel, Shenzhen
0755 - Address: Shawei store, qidouxing Business Hotel, building 11, Bagua 1st Road, Futian District, Shenzhen
address: Century handicraft trade market, Xinzhou South Road, Futian District, Shenzhen Building 315
Shenzhen Difu Hotel
address: 111 Zhenhua Road, Futian District, Shenzhen
Shenzhen 7-day Economy Chain Hotel Shenzhen Huaqiang hotel reservation center Tel: 0755 -
address: No.1 building, Huaqiang South saigeyuan, Futian District, Shenzhen
Shenzhen Guanyin hotel reservation center Tel: 0755 -
address: Futian District, Shenzhen Next to Binhe Avenue, Guangyin building, 38 South Road (Huanggang port)
enough
5. 1. It is recommended to take Longhua line from Shenzhen North Station to children's palace station, transfer to Longgang line to Futian station, take exit 1 and walk to Times financial center
2. Take m344 from Shenzhen North Station Bus hub station to great wall development station, transfer to 329 to radio and television building station, get off the bus and walk to Times financial center
2. Take m344 from Shenzhen North Station Bus hub station to great wall development station, transfer to 329 to radio and television building station, get off the bus and walk to Times financial center
6. Take bus 380b (10 stops) from Henggang market station to Tongxin Hotel station, then take bus 373 (18 stops) to Zhuoyue Times Square Station. Take bus 520 meters to Times financial center [get on and off map] [send SMS] [feedback information] 2. Take taxi: 27.8 km, red Taxi: 77.00 yuan. After 23:00 p.m., 99.00 yuan; Green Taxi: 84.00 yuan. After 23:00 PM, 88.00 yuan; Note: the information is for reference only and can not be used as the basis for payment. This fee does not include toll and Bridge toll. Traffic jam and other factors may significantly increase the cost [getting on and off the map] [sending SMS] [feedback information] 3. Take bus b679 (original 831b) from Henggang market station (10 stops) to Henggang mansion station (Song Road), Go to Henggang building terminal located on another road (Shenhui highway), transfer to k528 (Henggang railway station) (take 1 station) to railway station terminal (Luohu), transfer to 101 (take 8 stations) to radio and television building station 2. Walk about 510 meters to Times financial center [get on and off map] [send SMS] [feedback information] 4. Walk about 510 meters from Henggang market station to Henggang building terminal, take k528 (Henggang railway station) (take 1 station) to railway station Go to the entrance B of Luohu station and take subway line 1 (8 stops) to the entrance D of shopping park station. Walk about 480 meters to the Times financial center [get on and off map] [send SMS] [feedback] 5. From Henggang market station, take b679 Road (831b Road) (7 stops) to Henggang people's Hospital Station, take 357 Road (17 stops) to Buji joint inspection station, and take 310-315 ring line (12 stops) to radio and Television University Take about 430 meters from Henggang market station to Tongxin Hotel station, take no.380b (11 stops) to Dafen Wal Mart station, then take No.398 (19 stops) to excellence Times Square Station. Take about 520 meters to Times financial center [map of getting on and off] [send SMS] [feedback] 7. Take no.b679 (83 stops) from Henggang market station Take bus no.1b (7 stops) to Henggang people's Hospital Station, take bus No.357 (28 stops) to Lianhua Ercun east station, take bus no.329 (2 stops) to radio and television building (1 stop). Walk about 430 meters to Times financial center [map of getting on and off] [send SMS] [feedback] 8. Walk about 190 meters from Henggang market station to shangjiekou station, take bus No.357 (17 stops) to Honghu Park (1 stop) and transfer to 310-315 loop line (11 stops) to radio and television building Take bus b679 (original 831b) from Henggang market station to Henggang people's Hospital Station, then take bus 357 (take bus 20) to children's Park, then take bus 320 (take bus 11) to Guangdian building, and then walk 430 meters to Times financial center
7.
bus line: Luobao line, the whole journey is about 20.4 km
1. Walk about 530 meters from Bao'an District to Bao'an central station
2. Take Luobao line, pass 15 stops to Shopping Park Station
3. Walk about 480 meters to Times financial center
8. Low rank matrix is a low rank matrix
for the rank of a matrix, we need to introce the SVD decomposition of the matrix: x = USV & # 39;, U. V orthogonal matrix and s diagonal matrix. If it is a complete SVD decomposition, then the number of non-zero elements on the s diagonal is the rank of the matrix (these diagonal elements are called singular values), and there are some zero elements, which have no contribution to the rank
1.
If a matrix is regarded as a sample set, and each row (or column, it doesn't matter) is a sample, then the rank of the matrix is the dimension of the linear subspace formed by these samples. If the rank of the matrix is much less than the dimension of the samples (i.e. the number of matrix columns), then these samples are equivalent to living in a low dimensional subspace in the outer space, so that the dimension rection operation can be implemented. For example, the same person's face image collected under different light conditions, assuming that each image is
192x168, and 50 images are collected, the data matrix is composed of 50 rows and 192x168 columns. However, if you do SVD decomposition, you will find that only the top 10 singular values are larger, and their
singular values are close to zero, Therefore, the dimensions corresponding to the singular values close to zero can be discarded, and only the subspaces corresponding to the first 10 singular values can be retained, thus recing the data dimension to the 10 dimensional subspace
2.
consider a matrix as a mapping. Since it is a mapping, we have to consider its effect ax on the vector X. Note that ax is equivalent to a linear combination of the columns of A. if the matrix is of low rank, it means that the space stretched by these columns is a low dimensional subspace of the outer space, which is expressed by ax (where x is arbitrary). In other words, this matrix maps R ^ n space to R ^ m space, but its mapping image only lives in a low dimensional subspace of R ^ m space
. From SVD, ax = USV & # 39; x. So there are three transformations: the first is v & # 39; x. It is equivalent to rotating the coordinate axis in the original R ^ n space, so that only the coordinate
changes, and does not change the vector itself (for example, the length does not change); The second is s (V) 39; x) This is equivalent to stretching along each coordinate axis, and if some elements on the diagonal of s are zero, then the coordinate axes corresponding to these elements
are directly lost; Finally, u (SV) & 39; x) It's also an axis rotation. In general, ax is equivalent to stretching a vector x along some specific directions (with some non essential rotations), or even discarding it. The number of directions that are not discarded is rank.
for the rank of a matrix, we need to introce the SVD decomposition of the matrix: x = USV & # 39;, U. V orthogonal matrix and s diagonal matrix. If it is a complete SVD decomposition, then the number of non-zero elements on the s diagonal is the rank of the matrix (these diagonal elements are called singular values), and there are some zero elements, which have no contribution to the rank
1.
If a matrix is regarded as a sample set, and each row (or column, it doesn't matter) is a sample, then the rank of the matrix is the dimension of the linear subspace formed by these samples. If the rank of the matrix is much less than the dimension of the samples (i.e. the number of matrix columns), then these samples are equivalent to living in a low dimensional subspace in the outer space, so that the dimension rection operation can be implemented. For example, the same person's face image collected under different light conditions, assuming that each image is
192x168, and 50 images are collected, the data matrix is composed of 50 rows and 192x168 columns. However, if you do SVD decomposition, you will find that only the top 10 singular values are larger, and their
singular values are close to zero, Therefore, the dimensions corresponding to the singular values close to zero can be discarded, and only the subspaces corresponding to the first 10 singular values can be retained, thus recing the data dimension to the 10 dimensional subspace
2.
consider a matrix as a mapping. Since it is a mapping, we have to consider its effect ax on the vector X. Note that ax is equivalent to a linear combination of the columns of A. if the matrix is of low rank, it means that the space stretched by these columns is a low dimensional subspace of the outer space, which is expressed by ax (where x is arbitrary). In other words, this matrix maps R ^ n space to R ^ m space, but its mapping image only lives in a low dimensional subspace of R ^ m space
. From SVD, ax = USV & # 39; x. So there are three transformations: the first is v & # 39; x. It is equivalent to rotating the coordinate axis in the original R ^ n space, so that only the coordinate
changes, and does not change the vector itself (for example, the length does not change); The second is s (V) 39; x) This is equivalent to stretching along each coordinate axis, and if some elements on the diagonal of s are zero, then the coordinate axes corresponding to these elements
are directly lost; Finally, u (SV) & 39; x) It's also an axis rotation. In general, ax is equivalent to stretching a vector x along some specific directions (with some non essential rotations), or even discarding it. The number of directions that are not discarded is rank.
9. 1
2
3
4
5
from numpy import random
randarray = random. Random (size = (2,4))
#output
#array ([[0.93848018,0.42005976,0.81470729,0.98797783], [0.12242703,0.42756378,0.59705163,0.36619101])
the random function receives the tuple of the shape that needs to generate random matrix as the only parameter. The above code will return a random matrix with two rows and four columns. The value of the random number is between 0 and 1. The matrix is numpy. Array type. In addition to random function, there is randInt function which can generate integer random matrix.
2
3
4
5
from numpy import random
randarray = random. Random (size = (2,4))
#output
#array ([[0.93848018,0.42005976,0.81470729,0.98797783], [0.12242703,0.42756378,0.59705163,0.36619101])
the random function receives the tuple of the shape that needs to generate random matrix as the only parameter. The above code will return a random matrix with two rows and four columns. The value of the random number is between 0 and 1. The matrix is numpy. Array type. In addition to random function, there is randInt function which can generate integer random matrix.
Hot content