Search Results

Search found 1251 results on 51 pages for 'negative'.

Page 35/51 | < Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42  | Next Page >

  • Accounting for waves when doing planar reflections

    - by CloseReflector
    I've been studying Nvidia's examples from the SDK, in particular the Island11 project and I've found something curious about a piece of HLSL code which corrects the reflections up and down depending on the state of the wave's height. Naturally, after examining the brief paragraph of code: // calculating correction that shifts reflection up/down according to water wave Y position float4 projected_waveheight = mul(float4(input.positionWS.x,input.positionWS.y,input.positionWS.z,1),g_ModelViewProjectionMatrix); float waveheight_correction=-0.5*projected_waveheight.y/projected_waveheight.w; projected_waveheight = mul(float4(input.positionWS.x,-0.8,input.positionWS.z,1),g_ModelViewProjectionMatrix); waveheight_correction+=0.5*projected_waveheight.y/projected_waveheight.w; reflection_disturbance.y=max(-0.15,waveheight_correction+reflection_disturbance.y); My first guess was that it compensates for the planar reflection when it is subjected to vertical perturbation (the waves), shifting the reflected geometry to a point where is nothing and the water is just rendered as if there is nothing there or just the sky: Now, that's the sky reflecting where we should see the terrain's green/grey/yellowish reflection lerped with the water's baseline. My problem is now that I cannot really pinpoint what is the logic behind it. Projecting the actual world space position of a point of the wave/water geometry and then multiplying by -.5f, only to take another projection of the same point, this time with its y coordinate changed to -0.8 (why -0.8?). Clues in the code seem to indicate it was derived with trial and error because there is redundancy. For example, the author takes the negative half of the projected y coordinate (after the w divide): float waveheight_correction=-0.5*projected_waveheight.y/projected_waveheight.w; And then does the same for the second point (only positive, to get a difference of some sort, I presume) and combines them: waveheight_correction+=0.5*projected_waveheight.y/projected_waveheight.w; By removing the divide by 2, I see no difference in quality improvement (if someone cares to correct me, please do). The crux of it seems to be the difference in the projected y, why is that? This redundancy and the seemingly arbitrary selection of -.8f and -0.15f lead me to conclude that this might be a combination of heuristics/guess work. Is there a logical underpinning to this or is it just a desperate hack? Here is an exaggeration of the initial problem which the code fragment fixes, observe on the lowest tessellation level. Hopefully, it might spark an idea I'm missing. The -.8f might be a reference height from which to deduce how much to disturb the texture coordinate sampling the planarly reflected geometry render and -.15f might be the lower bound, a security measure.

    Read the article

  • Are You Meeting Social Customer Service Expectations?

    - by Mike Stiles
    Whether it’s B2B or B2C, one sure path to repeat business is making sure your buyer has a memorably pleasant and successful customer service experience with you. If they get that kind of treatment consistently, that’s called a relationship. And those aren’t broken easily. Social customer service, driven by integrated SRM (social relationship management) technology, is the venue that can effectively connect customers not only to the brand, but to other customers. Positive experiences, once administered, don’t just rest with the recipient. They’re published in the form of public raves and peer-to-peer recommendation, a force far more actionable than push advertising. What’s more, your customers have come to expect access to you and satisfaction from you using social. An NM Incite study shows 83% of Twitter users and 71% of Facebook users expect to get an answer from brands the same day they post to them on their social assets. To make sure you’re responding, you’ve got to have a tech platform that’s set up to moderate and alert so you’ll know ASAP a customer needs help. The more integrated your social enterprise is, the faster you can not only respond, but respond with the answer they’re looking for, because your system is connected to the internal resources that can surface the answer or put wheels in motion to rectify the situation in the shortest amount of time possible. But if you go to the necessary lengths to make sure your customers feel valued and important, will they really reward you? The study says 71% of consumers who got quick and effective responses from companies they contacted via social were more likely to recommend the brand to their friends and followers. So yes, sweeping people off their feet pays big dividends in terms of word-of-mouth marketing. But you should be keenly aware of the reverse side of that coin. Give people a negative experience, either in real world or virtual customer service, and that message is highly likely to get amplified through social channels faster and louder. Only 36% of the NM Incite study’s respondents reported that their problems were solved quickly and effectively. 36%? That’s hardly an impressive number. It gets worse. 10% never got so much as a response - at all. Going back to the relationship analogy, companies that are this deep in the ditch where customer service is concerned are making their girl or boyfriends really easy for a competitor to steal. Given the technology tools and data available right now for having an intimate knowledge of the customer, what products they’ve purchased, likely problems with those products, effective resolutions to those problems, and follow-up communication to gauge satisfaction, there are fewer excuses than ever for making the lifeblood of your business feel like you couldn’t care less. @mikestiles

    Read the article

  • WmiPrvSE memory leak on Windows 2008 *R2*

    - by MichaelGG
    I've seen references on Windows 2008 to WmiPrvSE leaks, but nothing about Windows 2008 R2. We're running R2 on top of Hyper-V (2008). We are also running NSClient++ for monitoring from opsview. Over time, WmiPrvSE.exe starts to use a lot of memory, causing memory alert issues (less than 10% free). VM has 2GB, WmiPrvSE consumes up to 500-600MB before I kill it. Killing the process doesn't seem to have any negative effect; it starts up again and I haven't noticed any problems. But after a day or two, it's back in the same situation. Any ideas on what to do? Resource Monitor doesn't show any Disk or Network IO by WmiPrvSE.exe. Just slowly climbing private memory... Edited to add: We aren't running clustering, or Windows System Resource Manager. The only regular WMI user I can guess is NSClient++, but we don't seem to have this problem on other servers.

    Read the article

  • Virtual SMTP not sending mails

    - by DoStuffZ
    Hi I have been googling for the better part of the last two hours without finding any conclusion. My mails are not being sent from the production webserver. If I stop/start the Virtual SMTP I get this in the event log: No usable TLS server certificate for SMTP virtual server instance '1' could be found. TLS will be disabled for this virtual-server. We recently updated the webapplication running and I assume something went amiss during that. Googling the message straight up gave me a list that just as well could have been in greek. I found a security certificate on the server, installing that gave no change. I basicly played russian roulette with the certificate file (.cer), though I was somewhat certain it would not have a negative effect. (Russian roulette with a 6 chamber gun and 2 bullets.) I found a .pfx in our local documentation folder, though I'm far from certain that to have a positive effect. (6 chambers and 5 bullets). I found a site describing how the Virtual SMTP - Properties - Access - should have a button saying Certificates. I have a text saying "Did not find any TLS certificates" and a grayed out tick box saying "Require TLS certificate". I found the TLS being SSL ver3.1+ (3.1-3.3). So question goes - How do I enable the SMTP to once again send emails, like before.

    Read the article

  • Macbook Pro 2010 Ethernet Jumbo Frame(9k MTU) Support?

    - by Troggy
    Has anyone been able to use jumbo frame support on their 2010 Macbook Pros? This is kind of negative news here, but I am seeing many reports that this is not available anymore due to Apple's choice of network card in the new machines. I cannot set my MTU speed over 1500 on my new 2010 MBP i7, but my old early 2008 MBP (Core2) has the 9000 MTU setting for use. Everything I have is setup to use jumbo frames and I thought apple kept that feature in their "pro" lineup. It sounds like the Mac Pro still has it. Did they decide to use a chipset that doesn't support it? I am trying to pinpoint some solid chipset numbers and the feature support. Maybe they just need to update the drivers? Is there some more official information about this feature? This might seem minor, but this is really frustrating if apple removed this feature from their pro laptop line. From what I have read so far, it sounds like I am not alone in my frustrations with this. http://discussions.info.apple.com/message.jspa?messageID=12258067 http://discussions.apple.com/thread.jspa?messageID=12130158 Anyone have any experience or further knowledge about this issue ... beyond typing my question into google and giving the top 5 results as answers? :)

    Read the article

  • Crash and Centos

    - by Jackob
    Hello guys and girls, I've a big problem it's the second time that my server crashed. I tried to check every thing and seems correct except this /var/log/messages: May 25 20:16:11 srv1 kernel: swap_free: Bad swap offset entry 00300000 May 25 20:16:33 srv1 kernel: swap_free: Unused swap offset entry 00080000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00340000 May 25 20:16:33 srv1 kernel: swap_free: Unused swap offset entry 000c0000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00a00000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00200000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00890000 May 25 20:16:33 srv1 kernel: swap_free: Unused swap offset entry 00080000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00f00000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00f90000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00980000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00980000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00820000 May 25 20:16:33 srv1 kernel: swap_free: Unused swap offset entry 000d0000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00a60000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00a20000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 009a0000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00170000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00f20000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00b60000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00a30000 May 25 20:16:33 srv1 kernel: swap_free: Bad swap offset entry 00320000 May 25 20:16:47 srv1 kernel: swap_free: Bad swap offset entry 002c0000 May 25 20:16:47 srv1 kernel: Eeek! page_mapcount(page) went negative! (-1) What can be my problem? Thanks so much!

    Read the article

  • I need advices: small memory footprint linux mail server with spam filtering

    - by petermolnar
    I have a VPS which is originally destined to be a webserver but some minimal mail capabilities are needed to be deployed as well, including sending and receiving as standalone server. The current setup is the following: Postfix reveices the mail, the users are in virtual tables, stored in MySQL on connection all servers are tested with policyd-weight service against some DNSBLs all mail is runs through SpamAssassin spamd with the help of spamc client the mail is then delivered with Dovecot 2' LDA (local delivery agent), virtual users as well As you saw... there's no virus scanner running, and that's for a reason: clamav eats all the memory possible and also, virus mails are all filtered out with this setup (I've tested the same with ClamAV enabled for 1,5 years, no virus mail ever got even to ClamAV) I don't use amavisd and I really don't want to. You only need that monster if you have plenty of memory and lots of simultaneous scanners. It's also a nightmare to fine tune by hand. I run policyd-weight instead of policyd and native DNSBLs in postfix. I don't like to send someone away because a single service listed them. Important statement: everything works fine. I receive very small amount of spam, nearly never get a false positive and most of the bad mail is stopped by policyd-weight. The only "problem" that I feel the services at total uses a bit much memory alltogether. I've already cut the modules of spamassassin (see below), but I'd really like to hear some advices how to cut the memory footprint as low as possible, mostly: what plugins SpamAssassin really needs and what are more or less useless, regarding to my current postfix & policyd-weight setup? SpamAssassin rules are also compiled with sa-compile (sa-update runs once a week from cron, compile runs right after that) These are some of the current configurations that may matter, please tell me if you need anything more. postfix/master.cf (parts only) dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/bin/spamc -e /usr/lib/dovecot/deliver -d ${recipient} -f {sender} postfix/main.cf (parts only) smtpd_helo_required = yes smtpd_helo_restrictions = permit_mynetworks, reject_invalid_hostname, permit smtpd_recipient_restrictions = permit_mynetworks, permit_sasl_authenticated, reject_invalid_hostname, reject_non_fqdn_hostname, reject_non_fqdn_recipient, reject_unknown_recipient_domain, reject_unauth_pipelining, reject_unauth_destination, check_policy_service inet:127.0.0.1:12525, permit policyd-weight.conf (parts only) $REJECTMSG = "550 Mail appeared to be SPAM or forged. Ask your Mail/DNS-Administrator to correct HELO and DNS MX settings or to get removed from DNSBLs"; $REJECTLEVEL = 4; $DEFER_STRING = 'IN_SPAMCOP= BOGUS_MX='; $DEFER_ACTION = '450'; $DEFER_LEVEL = 5; $DNSERRMSG = '450 No DNS entries for your MTA, HELO and Domain. Contact YOUR administrator'; # 1: ON, 0: OFF (default) # If ON request that ALL clients are only checked against RBLs $dnsbl_checks_only = 0; # 1: ON (default), 0: OFF # When set to ON it logs only RBLs which affect scoring (positive or negative) $LOG_BAD_RBL_ONLY = 1; ## DNSBL settings @dnsbl_score = ( # host, hit, miss, log name 'dnsbl.ahbl.org', 3, -1, 'dnsbl.ahbl.org', 'dnsbl.njabl.org', 3, -1, 'dnsbl.njabl.org', 'dnsbl.sorbs.net', 3, -1, 'dnsbl.sorbs.net', 'bl.spamcop.net', 3, -1, 'bl.spamcop.net', 'zen.spamhaus.org', 3, -1, 'zen.spamhaus.org', 'pbl.spamhaus.org', 3, -1, 'pbl.spamhaus.org', 'cbl.abuseat.org', 3, -1, 'cbl.abuseat.org', 'list.dsbl.org', 3, -1, 'list.dsbl.org', ); # If Client IP is listed in MORE DNSBLS than this var, it gets REJECTed immediately $MAXDNSBLHITS = 3; # alternatively, if the score of DNSBLs is ABOVE this level, reject immediately $MAXDNSBLSCORE = 9; $MAXDNSBLMSG = '550 Az levelezoszerveruk IP cime tul sok spamlistan talahato, kerjuk ellenorizze! / Your MTA is listed in too many DNSBLs; please check.'; ## RHSBL settings @rhsbl_score = ( 'multi.surbl.org', 4, 0, 'multi.surbl.org', 'rhsbl.ahbl.org', 4, 0, 'rhsbl.ahbl.org', 'dsn.rfc-ignorant.org', 4, 0, 'dsn.rfc-ignorant.org', # 'postmaster.rfc-ignorant.org', 0.1, 0, 'postmaster.rfc-ignorant.org', # 'abuse.rfc-ignorant.org', 0.1, 0, 'abuse.rfc-ignorant.org' ); # skip a RBL if this RBL had this many continuous errors $BL_ERROR_SKIP = 2; # skip a RBL for that many times $BL_SKIP_RELEASE = 10; ## cache stuff # must be a directory (add trailing slash) $LOCKPATH = '/var/run/policyd-weight/'; # socket path for the cache daemon. $SPATH = $LOCKPATH.'/polw.sock'; # how many seconds the cache may be idle before starting maintenance routines #NOTE: standard maintenance jobs happen regardless of this setting. $MAXIDLECACHE = 60; # after this number of requests do following maintenance jobs: checking for config changes $MAINTENANCE_LEVEL = 5; # negative (i.e. SPAM) result cache settings ################################## # set to 0 to disable caching for spam results. To this level the cache will be cleaned. $CACHESIZE = 2000; # at this number of entries cleanup takes place $CACHEMAXSIZE = 4000; $CACHEREJECTMSG = '550 temporarily blocked because of previous errors'; # after NTTL retries the cache entry is deleted $NTTL = 1; # client MUST NOT retry within this seconds in order to decrease TTL counter $NTIME = 30; # positve (i.,e. HAM) result cache settings ################################### # set to 0 to disable caching of HAM. To this number of entries the cache will be cleaned $POSCACHESIZE = 1000; # at this number of entries cleanup takes place $POSCACHEMAXSIZE = 2000; $POSCACHEMSG = 'using cached result'; #after PTTL requests the HAM entry must succeed one time the RBL checks again $PTTL = 60; # after $PTIME in HAM Cache the client must pass one time the RBL checks again. #Values must be nonfractal. Accepted time-units: s, m, h, d $PTIME = '3h'; # The client must pass this time the RBL checks in order to be listed as hard-HAM # After this time the client will pass immediately for PTTL within PTIME $TEMP_PTIME = '1d'; ## DNS settings # Retries for ONE DNS-Lookup $DNS_RETRIES = 1; # Retry-interval for ONE DNS-Lookup $DNS_RETRY_IVAL = 5; # max error count for unresponded queries in a complete policy query $MAXDNSERR = 3; $MAXDNSERRMSG = 'passed - too many local DNS-errors'; # persistent udp connection for DNS queries. #broken in Net::DNS version 0.51. Works with Net::DNS 0.53; DEFAULT: off $PUDP= 0; # Force the usage of Net::DNS for RBL lookups. # Normally policyd-weight tries to use a faster RBL lookup routine instead of Net::DNS $USE_NET_DNS = 0; # A list of space separated NS IPs # This overrides resolv.conf settings # Example: $NS = '1.2.3.4 1.2.3.5'; # DEFAULT: empty $NS = ''; # timeout for receiving from cache instance $IPC_TIMEOUT = 2; # If set to 1 policyd-weight closes connections to smtpd clients in order to avoid too many #established connections to one policyd-weight child $TRY_BALANCE = 0; # scores for checks, WARNING: they may manipulate eachother # or be factors for other scores. # HIT score, MISS Score @client_ip_eq_helo_score = (1.5, -1.25 ); @helo_score = (1.5, -2 ); @helo_score = (0, -2 ); @helo_from_mx_eq_ip_score= (1.5, -3.1 ); @helo_numeric_score= (2.5, 0 ); @from_match_regex_verified_helo= (1,-2 ); @from_match_regex_unverified_helo = (1.6, -1.5 ); @from_match_regex_failed_helo = (2.5, 0 ); @helo_seems_dialup = (1.5, 0 ); @failed_helo_seems_dialup= (2, 0 ); @helo_ip_in_client_subnet= (0,-1.2 ); @helo_ip_in_cl16_subnet = (0,-0.41 ); #@client_seems_dialup_score = (3.75, 0 ); @client_seems_dialup_score = (0, 0 ); @from_multiparted = (1.09, 0 ); @from_anon= (1.17, 0 ); @bogus_mx_score = (2.1, 0 ); @random_sender_score = (0.25, 0 ); @rhsbl_penalty_score = (3.1, 0 ); @enforce_dyndns_score = (3, 0 ); spamassassin/init.pre (I've put the .pre files together) loadplugin Mail::SpamAssassin::Plugin::Hashcash loadplugin Mail::SpamAssassin::Plugin::SPF loadplugin Mail::SpamAssassin::Plugin::Pyzor loadplugin Mail::SpamAssassin::Plugin::Razor2 loadplugin Mail::SpamAssassin::Plugin::AutoLearnThreshold loadplugin Mail::SpamAssassin::Plugin::MIMEHeader loadplugin Mail::SpamAssassin::Plugin::ReplaceTags loadplugin Mail::SpamAssassin::Plugin::Check loadplugin Mail::SpamAssassin::Plugin::HTTPSMismatch loadplugin Mail::SpamAssassin::Plugin::URIDetail loadplugin Mail::SpamAssassin::Plugin::Bayes loadplugin Mail::SpamAssassin::Plugin::BodyEval loadplugin Mail::SpamAssassin::Plugin::DNSEval loadplugin Mail::SpamAssassin::Plugin::HTMLEval loadplugin Mail::SpamAssassin::Plugin::HeaderEval loadplugin Mail::SpamAssassin::Plugin::MIMEEval loadplugin Mail::SpamAssassin::Plugin::RelayEval loadplugin Mail::SpamAssassin::Plugin::URIEval loadplugin Mail::SpamAssassin::Plugin::WLBLEval loadplugin Mail::SpamAssassin::Plugin::VBounce loadplugin Mail::SpamAssassin::Plugin::Rule2XSBody spamassassin/local.cf (parts) use_bayes 1 bayes_auto_learn 1 bayes_store_module Mail::SpamAssassin::BayesStore::MySQL bayes_sql_dsn DBI:mysql:db:127.0.0.1:3306 bayes_sql_username user bayes_sql_password pass bayes_ignore_header X-Bogosity bayes_ignore_header X-Spam-Flag bayes_ignore_header X-Spam-Status ### User settings user_scores_dsn DBI:mysql:db:127.0.0.1:3306 user_scores_sql_password user user_scores_sql_username pass user_scores_sql_custom_query SELECT preference, value FROM _TABLE_ WHERE username = _USERNAME_ OR username = '$GLOBAL' OR username = CONCAT('%',_DOMAIN_) ORDER BY username ASC # for better speed score DNS_FROM_AHBL_RHSBL 0 score __RFC_IGNORANT_ENVFROM 0 score DNS_FROM_RFC_DSN 0 score DNS_FROM_RFC_BOGUSMX 0 score __DNS_FROM_RFC_POST 0 score __DNS_FROM_RFC_ABUSE 0 score __DNS_FROM_RFC_WHOIS 0 UPDATE 01 As adaptr advised I remove policyd-weight and configured postfix postscreen, this resulted approximately -15-20 MB from RAM usage and a lot faster work. I'm not sure it's working at full capacity but it seems promising.

    Read the article

  • How to configure amavisd-new for only scanning on particular senders/servers?

    - by mailq
    I'd like to know how to configure amavisd-new to only scan for Spam on particular clients (IPs, CIDRs or hostnames) or alternatively sender's email domain. I know that it is possible to do it on a recipient's mail address but not on how to do it for the sender's mail address. It is even possible to do it on a recipient's IP address with policy banks. But my approach should be to be independent of recipient and only relay on the sender. What I want to accomplish is to only scan mails originating from Yahoo, Google, Hotmail and the other big senders. So it is easier to configure which senders should be observed than the ones that shouldn't. I known that it is easier to achieve on the MTA side, but that is not part of the question because I already go a solution on the MTA side. I want to do it on amavisd-new. And it doesn't help to know how to put senders on a whitelist, as this still means that the mail goes through all the scanning but then gets a high negative score. The mail shouldn't be scanned at all unless sent by the big players. So which parameters in amavisd-new is the right one to enable scanning for particular senders and only for these?

    Read the article

  • Connecting to RDS database from EC2 instance using bind9 CNAME alias

    - by mptre
    I'm trying to get internal DNS up and running on a EC2 instance. The main goal is to be able to define CNAME aliases for other AWS services. For example: Instead of using the RDS endpoint, which might change over time, an alias mysql.company.int can be used instead. I'm using bind9 and here's my config files: /etc/bind/named.conf.local zone "company.int" { type master; file "/etc/bind/db.company.int"; }; /etc/bind/db.company.int ; $TTL 3600 @ IN SOA company.int. company.localhost. ( 20120617 ; Serial 604800 ; Refresh 86400 ; Retry 2419200 ; Expire 604800 ) ; Negative Cache TTL ; @ IN NS company.int. @ IN A 127.0.0.1 @ IN AAAA ::1 ; CNAME mysql IN CNAME xxxx.eu-west-1.rds.amazonaws.com. The dig command ensures me my alias is working as excepted: $ dig mysql.company.int ... ;; ANSWER SECTION: mysql.company.int. 3600 IN CNAME xxxx.eu-west-1.rds.amazonaws.com. xxxx.eu-west-1.rds.amazonaws.com. 60 IN CNAME ec2-yyy-yy-yy-yyy.eu-west-1.compute.amazonaws.com. ec2-yyy-yy-yy-yyy.eu-west-1.compute.amazonaws.com. 589575 IN A zzz.zz.zz.zzz ... As far as I can understand a reverse zone isn't needed for a simple CNAME alias. However when I try to connect to MySQL using my newly created alias the operation is giving me a timeout. $ mysql -uuser -ppassword -hmysql.company.int ERROR 2003 (HY000): Can't connect to MySQL server on 'mysql.company.int' (110) Any ideas? Thanks in advantage!

    Read the article

  • Are web service handler chains possible under IIS / ASP.NET

    - by Mike
    I'm working with a client who wants me to implement a particular design in an IIS/ASP.NET environment. This design was already implemented in Java, but I am not sure it is possible using Microsoft technologies. In a Tomcat/Java environment one can create so call Handler Chains. In essence a handler runs on the server on which the web service is running and it intercepts the SOAP message coming to the web service. The handler can perform a number of tasks before passing control to the web service. Some of these tasks may refer to authentication and authorization. Moreover, one can create handler chains, such that the handlers can run in a particular sequence before passing control to the web service. This is a very elegant solution, as certain aspects of authentication and authorization can be automatically performed, without the developer of the client application and of the web service having to invest anything in it. The code for the client application and the web service is not affected. You may find a number of articles on internet on this subject by searching on Google for "web service handler chain". I performed searches for web service handlers in IIS or ASP.NET. I get some hits, but apparently handlers in IIS have another meaning than that described above. My question therefor is: Can handler chains (as available in Java and Tomcat) be created in IIS? If so, how (any article, book, forum...)? Either a negative or a positive answer will be greatly appreciated. Mike

    Read the article

  • Weird fluctuating time on a XEN linux guest

    - by Vin-G
    I have a weird problem with some servers here at work. We have a few XEN guests who's current time fluctuates. # date;date;date;date;date;date;date Thu Feb 25 16:00:40 PHT 2010 Thu Feb 25 16:00:48 PHT 2010 Thu Feb 25 16:00:40 PHT 2010 Thu Feb 25 16:00:48 PHT 2010 Thu Feb 25 16:00:40 PHT 2010 Thu Feb 25 16:00:48 PHT 2010 Thu Feb 25 16:00:40 PHT 2010 As seen above, the time fluctuates between 16:00:48 and 16:00:40, which is problematic for us since computing for time differences in some of our scripts becomes inaccurate (ex. what should be a few ms differences becomes some few second differences, and even sometimes, negative differences). The problematic servers are linux guests on a XEN host. The time fluctuates on the guest systems, but it is okay in the host itself. I've ruled out ntpd since this happens irregardless of whether ntpd is running or not on the guest systems. Guest is on full virtualisation. The time on both the host and the guest does match except that the time in the guest fluctuates at about a few seconds from the host's time, and the host time does not fluctuate. /proc/sys/xen/independent_wallclock is 0 in the host and does not exist in the guest. Ntpd service was stopped and disabled. Setting independent_wallclock to 1 in the host has no effect (that is, time still fluctuates in the guest). Though I was not able to restart the guest as it is a production server. Might be able to do that over the weekend. Any ideas on what to check and how to resolve this problem?

    Read the article

  • USB External HDD NOT spinning down on Windows Vista / Windows 7

    - by Deepak
    I have 3 external 2.5" USB HDDs - all from different manufacturers and with different capacities. I also have access to multiple Windows Vista / Windows 7 / Windows XP computers. My problem is that with the Windows Vista and Windows 7 computers, the external USB drives DO NOT spin down when I do "Safely remove hardware". Windows will tell me that I can safely remove the device, but I can see (and feel the rotations of the disk when I touch the casing) that the disks are still spinning and NEVER spin down. They also never go into their suspended state (which is generally signaled with a slow flashing of the activity LED). However, with Windows XP, when I do "Safely remove hardware", I can see that the drives do indeed spin down without any issues and go into their respective suspended states. I notice that this behaviour is consistent across all my 3 drives and on different hardware. Has anybody else noticed the same issues? Is there any way we can have the same behaviour as Windows XP on Windows Vista and 7, because I feel on the long run, disconnecting the drives while they are still spinning will have a negative effect on their life span. Thanks, Deepak.

    Read the article

  • How do I increase the buffer size for domain sockets in OS X 10.6

    - by Chas. Owens
    In Linux I have no problem dumping tons of data into a domain socket, but the same code on OS X 10.6.2 blows up after about 65 records. The socket reader code looks like #!/usr/bin/perl use strict; use warnings; use IO::Socket; unlink "foo"; my $sock = IO::Socket::UNIX->new ( Local => 'foo', Type => SOCK_DGRAM, Timeout => 600, ) or die "Could not create socket: $!\n"; while (<$sock>) { chomp; print "[$_]\n"; } And the client code looks like #!/usr/bin/perl use strict; use warnings; use IO::Socket; my $sock = IO::Socket::UNIX->new ( Peer => 'foo', Type => SOCK_DGRAM, Timeout => 600, ) or die "Could not create socket: $!\n"; for my $i (1 .. 1_000_000) { print $sock "$i\n" or die $!; } close $sock; The error message I get is No buffer space available at write.pl line 15.. It seems fairly obvious that there is a difference in the buffer size between Linux and OS X, but I don't know how to set it OS X (or what the possible negative side effects might be).

    Read the article

  • how do you add an A record for a root domain

    - by nbv4
    this seems really simple, but I can't figure it out. I'm using xname.org since it's free and I own a bunch of domains spread over a few different registrars. The setup I desire is very simple: one A record that points the bare domain name to my server, plus a wildcard CNAME record pointing all subdomains to the same server. So if the user goes to domain.com it will point them to 285.24.435.75, if they go to www.domain, blah.domain.com, or any other sub domain, they'll get sent to 285.24.435.75. All the examples I read on the internet about setting up A records all have the A record set to a subdomain such as www. WWW is deprecated so I want to have noting to do with it. Currently my xname.org zone looks like this: $TTL 86400 ; Default TTL domain.com. IN SOA ns0.xname.org. nbvfour.gmail.com. ( 2010052503 ; serial 10800 ; Refresh period 3600 ; Retry interval 604800 ; Expire time 10800 ; Negative caching TTL ) $ORIGIN domain.com. IN NS ns2.xname.org. IN NS ns0.xname.org. IN NS ns1.xname.org. @ IN A 65.49.73.148 * IN CNAME domain.com The '@' symbol is something that the godaddy domain interface uses to mean "this root domain', but that may have been specefic to that interface and has no meaning here. Before I had a 'www' entry in the A rcords and it worked in the sense that I could ping 'www.domain.com' and it returned a response, but pinging the root domain 'domain.com' returned "no host found".

    Read the article

  • Running Debian as guest operating system on a Hyper-V VM

    - by kce
    Hello. Layer-9 considerations are prompting a migration from Citrix XenServer to Hyper-V as our shop's virtualization platform of choice. This will require me to migrate our existing virtual machines from XenServer to Hyper-V. A hand full of these VMs are running Debian. Unfortunately, Debian does not seem to be on the list of approved/supported guest operating systems. In fact it seems that running Debian as a guest operating system of is rather difficult, although apparently not impossible. I have two interrelated questions: Does anyone have any experience running a Debian guest on Hyper-V? Is it one of those things where it just will not work at all or is more along the lines of "it will probably work fine, but we won't support it". Any experience here, positive or negative, would be helpful. How much of a bad idea is it to deviate from Hyper-V's list of supported guest operating systems? Again, is it either basically asking for Bad Things (TM) to happen or is just another instance of "it will probably work fine, but we won't support it"? Or is it somewhere in the middle? Thank you.

    Read the article

  • Server 2012, Jumbo Frames - should I expect problems?

    - by TomTom
    Ok, this sound might stupid - but is there any negative on just enabling jumbo frames in practice? From what I understand: Any switch or ethernet adapter that sees a jumbo frame it can not handle will just drop it. TCP is not a problem as max frame size is negotiated in the setinuo phase. UCP is a theoretical problem as a server may just send a LARGE UDP packet that gets dropped on the way. Practically though, as UDP is packet based, I do not really think any software WOULD send a UDP packet larger than 1500 bytes net without app level configuration changes - at least this is how I do my programming, as it is quite hard to get a decent MTU size for that without testing yourself, so you fall back in programming to max 1500 packets. The network in question is a standard small business network - we upgraded now from a non managed 24 port switch to a 52 port switch with 4 10g ports (netgear - quite cheap) and will mov a file server to 10g for also ISCSI serving. All my equipment on the Ethernet level can handle minimum 9000 bytes and due to local firewalls I really want to get packets larger (less firewall processing), but the network is also NAT'ed to the internet. On top, different machines move around (download) large files (multi gigabyte area) quite often for processing. The question is - can I expect problems when I just enable jumbo frames? Again, this is not totally ignorance - I just don't see programs sending more than 1500 byte UDP packets (if that is a practical problem please tell me) and for TCP the MTU is negotiated anyway. if there is a problem I can move to a dedicated VLAN, but this has it's own shares of problems as basically most workstations must then be on both VLAN's.

    Read the article

  • DNS propagation delay or bad configuration?

    - by Javier Martinez
    I have been waiting the DNS propagation for almost 24 hours. I'am no impatient, but I want to know if I configured my zone good or I have any error in it. I think that is good, because if I use my server dns like my DNS secondary I can resolve and lookup host well. ; ; BIND data file for mydomain.net ; $TTL 86400 @ IN SOA mydomain.net. mydomain.net. ( 20120629 ; Serial 10800 ; Refresh 3 hours 3600 ; Retry 1 hour 604800 ; Expire 1 week 86400 ) ; Negative Cache TTL ; @ IN NS ns1 @ IN NS ns2 IN MX 10 mail ns1 IN A 5.39.X.Y ns2 IN A 5.39.X.Z There is not any errors in /var/syslog about bind daemon. Is everything correct? Do I only need to wait up to 48 hours for the right DNS propagation? My nslookup from a remote machine with the nameserver of the bind host: $ nslookup mydomain.net Server: bind-host-ip Address: bind-host-ip#53 Name: mydomain.net Address: domain-ip

    Read the article

  • How do I increase the buffer size for domain sockets in OS X 10.6

    - by Chas. Owens
    In Linux I have no problem dumping tons of data into a domain socket, but the same code on OS X 10.6.2 blows up after about 65 records. The socket reader code looks like #!/usr/bin/perl use strict; use warnings; use IO::Socket; unlink "foo"; my $sock = IO::Socket::UNIX->new ( Local => 'foo', Type => SOCK_DGRAM, Timeout => 600, ) or die "Could not create socket: $!\n"; while (<$sock>) { chomp; print "[$_]\n"; } And the client code looks like #!/usr/bin/perl use strict; use warnings; use IO::Socket; my $sock = IO::Socket::UNIX->new ( Peer => 'foo', Type => SOCK_DGRAM, Timeout => 600, ) or die "Could not create socket: $!\n"; for my $i (1 .. 1_000_000) { print $sock "$i\n" or die $!; } close $sock; The error message I get is No buffer space available at write.pl line 15.. It seems fairly obvious that there is a difference in the buffer size between Linux and OS X, but I don't know how to set it OS X (or what the possible negative side effects might be).

    Read the article

  • Varnish with multiple sites/boxes

    - by jerhinesmith
    Is it possible for Varnish to redirect traffic to different IPs based on the url? For example, is the following setup feasible (and if so, what would the VCL look like): *.example.com points to Varnish IP address When a request is made to foo.example.com, varnish checks the cache and sends the request to Server1's IP address on a cache miss. When a request is made to bar.example.com, varnish checks the cache and sends the request to Server2's IP address on a cache miss. foo and bar are (for the most part) completely unrelated sites. They use the engine, but have different content and their own distinct database. Since there previously was no penalty for doing so (other than cost) we split them up into two separate boxes so that a ton of traffic to foo won't have a negative impact on visitors browsing around bar. I could set up two instances of varnish and have one serve up foo's static content and the other serve up bar's, but as there doesn't seem to be much overhead to running Varnish, I think (perhaps mistakenly) that it would make more sense to go with one Varnish server that redirects the traffic to the appropriate box on a cache miss.

    Read the article

  • Thomson router reboots unexpectedly with an apparent remote connection attempt

    - by ChrisF
    I've got a weird problem. Every so often my rooter (a Thomson TG585 v8 running version 8.2.7.8 of it's firmware) reboots itself. It seems to be associated with this message in the event log: FIREWALL replay check (1 of 2): Protocol: ICMP Src ip: 183.178.144.177 Dst ip: xxx.xxx.xxx.xxx Type: Destination Unreachable Code: Host Unreacheable xxx.xxx.xxx.xxx is my external IP address 183.178.144.177 resolves to 183178144177.ctinets.com We've got a student from Hong Kong staying with us at the moment and the reboots seem coincidental with him starting up his laptop. I say this because a check on ctinets.com shows it to be based in Hong Kong, though our guest's laptop doesn't appear to have any software related to this company installed. I say "apparently" as he is running the Chinese version of Windows and his English doesn't cover technical subjects like this. I know this is an incoming message but I was assuming that it was in response to something on the student's laptop which is why the first thought was malware, but we've got anti virus on all the other machines and have run malwarebytes on his with a negative result so I don't think the problem is due to a virus or (known) trojan. What else can I do to stop this and identify the cause?

    Read the article

  • ipv6 reverse DNS delegation

    - by user1709492
    I currently have 2001:1973:2303::/48 assigned to me and i'll be assigning /64's to customer's I'd like to have 1 zonefile for the /48 where i can essentially point / redirect query to different nameservers. Example ( Desired effect ) 2001:1973:2303:1234::/64 -> ns1.example.com, ns2.example.com 2001:1973:2303:2345::/64 -> ns99.example2.com, ns100.example2.com 2001:1973:2303:4321::/64 -> ns1.cust1.com, ns2.cust1.com Current /48 zonefile $TTL 3h $ORIGIN 3.0.3.2.3.7.9.1.1.0.0.2.ip6.arpa. @ IN SOA ns3.example.ca. ns4.example.ca. ( 2011071030 ; serial 3h ; refresh after 3 hours 1h ; retry after 1 hour 1w ; expire after 1 week 1h ) ; negative caching TTL of 1 hour IN NS ns3.example.ca. IN NS ns4.example.ca. 1234 IN NS ns1.example.com. NS ns2.example.com. 2345 IN NS ns99.example2.com. NS ns100.example2.com. 4321 IN NS ns1.cust1.com. NS ns2.cust1.com. Where am i going wrong ? My request seems simple to me atleast. To put it in terms of firewalling i want to redirect traffic client queries 2001:1973:2303:4321::1 - ns3.example.ca sees the request and redirects the query to ns1.cust1.com - ns1.cust1.com answers the query with omg.itworks.ca ( provided ns1.cust1.com is properly configured.

    Read the article

  • Problems with real-valued input deep belief networks (of RBMs)

    - by Junier
    I am trying to recreate the results reported in Reducing the dimensionality of data with neural networks of autoencoding the olivetti face dataset with an adapted version of the MNIST digits matlab code, but am having some difficulty. It seems that no matter how much tweaking I do on the number of epochs, rates, or momentum the stacked RBMs are entering the fine-tuning stage with a large amount of error and consequently fail to improve much at the fine-tuning stage. I am also experiencing a similar problem on another real-valued dataset. For the first layer I am using a RBM with a smaller learning rate (as described in the paper) and with negdata = poshidstates*vishid' + repmat(visbiases,numcases,1); I'm fairly confident I am following the instructions found in the supporting material but I cannot achieve the correct errors. Is there something I am missing? See the code I'm using for real-valued visible unit RBMs below, and for the whole deep training. The rest of the code can be found here. rbmvislinear.m: epsilonw = 0.001; % Learning rate for weights epsilonvb = 0.001; % Learning rate for biases of visible units epsilonhb = 0.001; % Learning rate for biases of hidden units weightcost = 0.0002; initialmomentum = 0.5; finalmomentum = 0.9; [numcases numdims numbatches]=size(batchdata); if restart ==1, restart=0; epoch=1; % Initializing symmetric weights and biases. vishid = 0.1*randn(numdims, numhid); hidbiases = zeros(1,numhid); visbiases = zeros(1,numdims); poshidprobs = zeros(numcases,numhid); neghidprobs = zeros(numcases,numhid); posprods = zeros(numdims,numhid); negprods = zeros(numdims,numhid); vishidinc = zeros(numdims,numhid); hidbiasinc = zeros(1,numhid); visbiasinc = zeros(1,numdims); sigmainc = zeros(1,numhid); batchposhidprobs=zeros(numcases,numhid,numbatches); end for epoch = epoch:maxepoch, fprintf(1,'epoch %d\r',epoch); errsum=0; for batch = 1:numbatches, if (mod(batch,100)==0) fprintf(1,' %d ',batch); end %%%%%%%%% START POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% data = batchdata(:,:,batch); poshidprobs = 1./(1 + exp(-data*vishid - repmat(hidbiases,numcases,1))); batchposhidprobs(:,:,batch)=poshidprobs; posprods = data' * poshidprobs; poshidact = sum(poshidprobs); posvisact = sum(data); %%%%%%%%% END OF POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% poshidstates = poshidprobs > rand(numcases,numhid); %%%%%%%%% START NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% negdata = poshidstates*vishid' + repmat(visbiases,numcases,1);% + randn(numcases,numdims) if not using mean neghidprobs = 1./(1 + exp(-negdata*vishid - repmat(hidbiases,numcases,1))); negprods = negdata'*neghidprobs; neghidact = sum(neghidprobs); negvisact = sum(negdata); %%%%%%%%% END OF NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% err= sum(sum( (data-negdata).^2 )); errsum = err + errsum; if epoch>5, momentum=finalmomentum; else momentum=initialmomentum; end; %%%%%%%%% UPDATE WEIGHTS AND BIASES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% vishidinc = momentum*vishidinc + ... epsilonw*( (posprods-negprods)/numcases - weightcost*vishid); visbiasinc = momentum*visbiasinc + (epsilonvb/numcases)*(posvisact-negvisact); hidbiasinc = momentum*hidbiasinc + (epsilonhb/numcases)*(poshidact-neghidact); vishid = vishid + vishidinc; visbiases = visbiases + visbiasinc; hidbiases = hidbiases + hidbiasinc; %%%%%%%%%%%%%%%% END OF UPDATES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% end fprintf(1, '\nepoch %4i error %f \n', epoch, errsum); end dofacedeepauto.m: clear all close all maxepoch=200; %In the Science paper we use maxepoch=50, but it works just fine. numhid=2000; numpen=1000; numpen2=500; numopen=30; fprintf(1,'Pretraining a deep autoencoder. \n'); fprintf(1,'The Science paper used 50 epochs. This uses %3i \n', maxepoch); load fdata %makeFaceData; [numcases numdims numbatches]=size(batchdata); fprintf(1,'Pretraining Layer 1 with RBM: %d-%d \n',numdims,numhid); restart=1; rbmvislinear; hidrecbiases=hidbiases; save mnistvh vishid hidrecbiases visbiases; maxepoch=50; fprintf(1,'\nPretraining Layer 2 with RBM: %d-%d \n',numhid,numpen); batchdata=batchposhidprobs; numhid=numpen; restart=1; rbm; hidpen=vishid; penrecbiases=hidbiases; hidgenbiases=visbiases; save mnisthp hidpen penrecbiases hidgenbiases; fprintf(1,'\nPretraining Layer 3 with RBM: %d-%d \n',numpen,numpen2); batchdata=batchposhidprobs; numhid=numpen2; restart=1; rbm; hidpen2=vishid; penrecbiases2=hidbiases; hidgenbiases2=visbiases; save mnisthp2 hidpen2 penrecbiases2 hidgenbiases2; fprintf(1,'\nPretraining Layer 4 with RBM: %d-%d \n',numpen2,numopen); batchdata=batchposhidprobs; numhid=numopen; restart=1; rbmhidlinear; hidtop=vishid; toprecbiases=hidbiases; topgenbiases=visbiases; save mnistpo hidtop toprecbiases topgenbiases; backpropface; Thanks for your time

    Read the article

  • Input field separator in awk

    - by Matthijs
    I have many large data files. The delimiter between the fields is a semicolon. However, I have found that there are semicolons in some of the fields, so I cannot simply use the semicolon as a field separator. The following example has 4 fields, but awk sees only 3, because the '1' in field 3 is stripped by the regex (which includes a '-' because some of the numerical data are negative): echo '"This";"is";1;"line of; data"' | awk -F'[0-9"-];[0-9"-]' '{print "No. of fields:\t"NF; print "Field 3:\t" $3}' No. of fields: 3 Field 3: ;"line of; data" Of course, echo '"This";"is";1;"line of; data"' | awk -F';' '{print "No. of fields:\t"NF}' No. of fields: 5 solves that problem, but counts the last field as two separate fields. Does anyone know a solution to this? Thanks! Matthijs

    Read the article

  • SQL Server 2000, large transaction log, almost empty, performance issue?

    - by Mafu Josh
    For a company that I have been helping troubleshoot their database. In SQL Server 2000, database is about 120 gig. Something caused the transaction log to grow MUCH larger than normal to over 100 gig, some hung transaction that didn't commit or roll back for a few days. That has been resolved and it now stays around 1% full or less, due to its hourly transaction log backups. It IS my understanding that a GROWING transaction log file size can cause performance issues. But what I am a little paranoid about is the size. Although mainly empty, MIGHT it be having a negative effect on performance? But I haven't found any documentation that suggests this is true. I did find this link: http://www.bigresource.com/MS_SQL-Large-Transaction-Log-dramatically-Slows-down-processing-any-idea-why--2ahzP5wK.html but in this post I can't tell if their log was full or empty, and there is not any replies to the post in this link. So I am guessing it is not a problem, anyone know for sure?

    Read the article

  • Problems with real-valued deep belief networks (of RBMs)

    - by Junier
    I am trying to recreate the results reported in Reducing the dimensionality of data with neural networks of autoencoding the olivetti face dataset with an adapted version of the MNIST digits matlab code, but am having some difficulty. It seems that no matter how much tweaking I do on the number of epochs, rates, or momentum the stacked RBMs are entering the fine-tuning stage with a large amount of error and consequently fail to improve much at the fine-tuning stage. I am also experiencing a similar problem on another real-valued dataset. For the first layer I am using a RBM with a smaller learning rate (as described in the paper) and with negdata = poshidstates*vishid' + repmat(visbiases,numcases,1); I'm fairly confident I am following the instructions found in the supporting material but I cannot achieve the correct errors. Is there something I am missing? See the code I'm using for real-valued visible unit RBMs below, and for the whole deep training. The rest of the code can be found here. rbmvislinear.m: epsilonw = 0.001; % Learning rate for weights epsilonvb = 0.001; % Learning rate for biases of visible units epsilonhb = 0.001; % Learning rate for biases of hidden units weightcost = 0.0002; initialmomentum = 0.5; finalmomentum = 0.9; [numcases numdims numbatches]=size(batchdata); if restart ==1, restart=0; epoch=1; % Initializing symmetric weights and biases. vishid = 0.1*randn(numdims, numhid); hidbiases = zeros(1,numhid); visbiases = zeros(1,numdims); poshidprobs = zeros(numcases,numhid); neghidprobs = zeros(numcases,numhid); posprods = zeros(numdims,numhid); negprods = zeros(numdims,numhid); vishidinc = zeros(numdims,numhid); hidbiasinc = zeros(1,numhid); visbiasinc = zeros(1,numdims); sigmainc = zeros(1,numhid); batchposhidprobs=zeros(numcases,numhid,numbatches); end for epoch = epoch:maxepoch, fprintf(1,'epoch %d\r',epoch); errsum=0; for batch = 1:numbatches, if (mod(batch,100)==0) fprintf(1,' %d ',batch); end %%%%%%%%% START POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% data = batchdata(:,:,batch); poshidprobs = 1./(1 + exp(-data*vishid - repmat(hidbiases,numcases,1))); batchposhidprobs(:,:,batch)=poshidprobs; posprods = data' * poshidprobs; poshidact = sum(poshidprobs); posvisact = sum(data); %%%%%%%%% END OF POSITIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% poshidstates = poshidprobs > rand(numcases,numhid); %%%%%%%%% START NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% negdata = poshidstates*vishid' + repmat(visbiases,numcases,1);% + randn(numcases,numdims) if not using mean neghidprobs = 1./(1 + exp(-negdata*vishid - repmat(hidbiases,numcases,1))); negprods = negdata'*neghidprobs; neghidact = sum(neghidprobs); negvisact = sum(negdata); %%%%%%%%% END OF NEGATIVE PHASE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% err= sum(sum( (data-negdata).^2 )); errsum = err + errsum; if epoch>5, momentum=finalmomentum; else momentum=initialmomentum; end; %%%%%%%%% UPDATE WEIGHTS AND BIASES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% vishidinc = momentum*vishidinc + ... epsilonw*( (posprods-negprods)/numcases - weightcost*vishid); visbiasinc = momentum*visbiasinc + (epsilonvb/numcases)*(posvisact-negvisact); hidbiasinc = momentum*hidbiasinc + (epsilonhb/numcases)*(poshidact-neghidact); vishid = vishid + vishidinc; visbiases = visbiases + visbiasinc; hidbiases = hidbiases + hidbiasinc; %%%%%%%%%%%%%%%% END OF UPDATES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% end fprintf(1, '\nepoch %4i error %f \n', epoch, errsum); end dofacedeepauto.m: clear all close all maxepoch=200; %In the Science paper we use maxepoch=50, but it works just fine. numhid=2000; numpen=1000; numpen2=500; numopen=30; fprintf(1,'Pretraining a deep autoencoder. \n'); fprintf(1,'The Science paper used 50 epochs. This uses %3i \n', maxepoch); load fdata %makeFaceData; [numcases numdims numbatches]=size(batchdata); fprintf(1,'Pretraining Layer 1 with RBM: %d-%d \n',numdims,numhid); restart=1; rbmvislinear; hidrecbiases=hidbiases; save mnistvh vishid hidrecbiases visbiases; maxepoch=50; fprintf(1,'\nPretraining Layer 2 with RBM: %d-%d \n',numhid,numpen); batchdata=batchposhidprobs; numhid=numpen; restart=1; rbm; hidpen=vishid; penrecbiases=hidbiases; hidgenbiases=visbiases; save mnisthp hidpen penrecbiases hidgenbiases; fprintf(1,'\nPretraining Layer 3 with RBM: %d-%d \n',numpen,numpen2); batchdata=batchposhidprobs; numhid=numpen2; restart=1; rbm; hidpen2=vishid; penrecbiases2=hidbiases; hidgenbiases2=visbiases; save mnisthp2 hidpen2 penrecbiases2 hidgenbiases2; fprintf(1,'\nPretraining Layer 4 with RBM: %d-%d \n',numpen2,numopen); batchdata=batchposhidprobs; numhid=numopen; restart=1; rbmhidlinear; hidtop=vishid; toprecbiases=hidbiases; topgenbiases=visbiases; save mnistpo hidtop toprecbiases topgenbiases; backpropface; Thanks for your time

    Read the article

< Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42  | Next Page >