Recent Wilmington Real Estate Transactions

first_imgWILMINGTON, MA — Below are the real estate transactions in Wilmington that occurred from April 17, 2019 to April 23, 2019:Address: 2 King StreetPrice: $685,000Buyer: Dylan & Taylor SempleSeller: Samuel & Siddhi ChhoengDate: 4/19/19Use: 1-Family ResidenceLot Size: 16,000sfAddress: 6 Leonard LanePrice: $900,000Buyer: Shonna & Jonathan ScalfaniSeller: Kenneth & Jill ChisholmDate: 4/19/19Use: 1-Family ResidenceLot Size: 35,401sfAddress: 44 Nathan RoadPrice: $925,000Buyer: Peter & Lauren HaistSeller: Andrew & Nancy BarrDate: 4/19/19Use: 1-Family ResidenceLot Size: 60,871sfAddress: 4 Seneca LanePrice: $672,500Buyer: Srujith & Nikhlia KudikalaSeller: Kevin & Barbara MurrayDate: 4/17/19Use: 1-Family ResidenceLot Size: 25,000sfLike Wilmington Apple on Facebook. Follow Wilmington Apple on Twitter. Follow Wilmington Apple on Instagram. Subscribe to Wilmington Apple’s daily email newsletter HERE. Got a comment, question, photo, press release, or news tip? Email wilmingtonapple@gmail.com.Share this:TwitterFacebookLike this:Like Loading… RelatedRecent Wilmington Real Estate TransactionsIn “Business”Recent Wilmington Real Estate TransactionsIn “Business”Wilmington Real Estate Transactions (Week of August 13, 2019)In “Business”last_img read more

Australia vs Afghanistan ICC 2019 World Cup Live telecast preview players to

first_imgDefending champions Australia start off their campaign to regain the world title against a relatively weak opposition in Afghanistan at Bristol on June 1. However, the Afghans have done enough in their brief ODI history to make the Aussies and every other team wary. They have a trump card in the form of Rashid Khan whose success or failure may well decide the course of the match.When and where to watchBeing the second match of the day, this contest will begin at 1:30 PM local time and 6 PM IST. It will be telecasted on Star Sports 3 and streamed online on Hotstar. PreviewA lot of people are interested in this match, but not because of what is expected to happen on the field. The curiosity relates to the kind of reaction David Warner and Steve Smith are expected to get when they make their return to international cricket, that too, in front of an English crowd.When Steve Smith scored a hundred in a warm-up game, he got booed by the people in attendance. While some English players have requested the fans to not be so uncharitable, the English crowds are known to have a mind of their own.But looking at the match, Australia should record an easy win. Their batting order was looking quite solid even before the re-emergence of Warner and Smith. Aaron Finch scored heavily in the last ODI series of his team while Khawaja has been pumping out runs with great prolificacy since the 5-match rubber in India.Now the question is: Where will Warner bat? He is bound to be in top-3. Smith would come at no. 4 and by his century in the practice game has proven that he hasn’t lost his value in ODI cricket. Shaun Marsh and Glenn Maxwell might be next with Alex Carey finishing off the top-7. However, if Australia want a seamer as their fifth bowler, Marcus Stoinis may get in.The bowling would consist of Mitchell Starc, the hero of the last World Cup with his deadly speed and yorkers, and Pat Cummins, one of the very best in the World. Whether Aussies pick two spinners or one depends upon the nature of the wicket. If there is a bit of dryness, both Adam Zampa and Nathan Lyon would play. In case only one is needed, it will be tough to choose between Zampa and Lyon as both have done well. The former gives away runs but picks up wickets while Lyon is great for controlling the game through accurate bowling. The comeback of David Warner will be something keenly watchedDaniel Kalisz/Getty ImagesFor the third seamer’s spot, there are three options: Nathan Coulter Nile, Jason Behrendorff and Kane Richardson. None of the three have set the stage on fire but could be good as a partner with Cummins and Starc.On the Afghanistan side, the hopes would rest on Rashid Khan. But he won’t be alone among spinners. Mohammad Nabi and Mujeeb Ur Rahman have been good in recent times. A collective effort would be required from all of them. The pace bowling department of Afghanistan would be followed closely to see how Hamid Hassan, making a comeback into the team does. Among their batsmen, hopes are high from Hazratullah Zazai, a dashing, aggressive batsmen who has played some great knocks in T20 cricket. The likes of Hashmatullah Shahidi and Asghar Afghan are also expected to rear up the Afghan innings. But they will have their task cut out against the Cummins and co.Essentially, we are looking at an easy win for the 5-time world champions. But who knows, there may be a surprise in store. The hope for Afghans lies in Rashid’s bowling. If he can bamboozle the batsmen in this match, they will have an opprtunity, if not, it’s a foregone conclusion. Players to watchDavid Warner: Away from international cricket for more than a year, Warner doesn’t seem to have lost his great ability. After performing well in the IPL, he might want to make a big statement in the World Cup.Hamid Hassan: When Afghanistan first appeared on the international scene in the 2010 World T20, Hassan was the most impressive performer. He could bowl genuinely fast, reaching speeds up to 140/kph+. It would be interesting to see if he has maintained that pace.Predicted XIsAustralia: Aaron Finch (C), David Warner, Usman Khawaja, Steve Smith, Glenn Maxwell, Marcus Stoinis, Alex Carey (WK), Nathan Lyon, Pat Cummins, Nathan Coulter-Nile, Mitchell StarcAfghanistan: Mohammad Shahzad (WK), Hazratullah Zazai, Hashmatullah Shahidi, Asghar Afghan, Najibullah Zadran, Rahmat Shah, Mohammad Nabi, Gulbadin Naib (C), Rashid Khan, Mujeeb Ur Rehman, Hamid Hassanlast_img read more

IRMA UPDATE Nearly 125 Million Without Power In Florida

first_imgThe Latest on Irma (all times local):6:30 p.m.Florida emergency management officials estimate nearly 12.5 million residents remain without power.The updated number came during a briefing on Monday evening at the state’s emergency management center in Tallahassee.Hurricane and Tropical Storm Irma has left 62 percent of customers without power.Of the 6.5 million reported outages, nearly a third are in South Florida. Nineteen of Florida’s 67 counties report more than 100,000 outages.The biggest outages are in Miami-Dade County (801,648) followed by Broward (629,134) and Palm Beach (505,520). Hendry County in the southwest part of the state is the only county listing 100 percent outages.___6 p.m.Authorities are reporting the first death in South Carolina related to Tropical Storm Irma.Abbeville County Coroner Ronnie Ashley said 57-year-old Charles Saxon was cleaning limbs and debris outside his home in Calhoun Falls around 3 p.m. Monday when a limb fell on him.Ashely said in a news release that Saxon died at the scene. An autopsy has been ordered.The National Weather Service says winds in the area were gusting to around 40 mph (65 kph) at the time Saxon was killed. Calhoun Falls is located 60 miles (95 kilometers) south of Greenville, South Carolina.___5:30 p.m.Irma’s eye has finally left Florida and exited the state as a weak tropical storm with 50-mph (85-kph) winds.The National Hurricane Center says the storm’s center is over southwestern Georgia, about 10 miles (15 kilometers) east of Albany. It is forecast to take a northwest turn Tuesday morning, moving into Alabama.It is zipping north-northwest at 17 mph (28 kph) It is still a 415-mile (665-kilometers) wide storm.Some, but not all, storm warnings in Florida have been discontinued, but storm surge is still expected along western Florida and from around Daytona Beach to South Carolina. South Carolina, Alabama and north central Georgia are expected to get 3 to 6 inches (7 to 15 centimeters) of rain with spots hitting 10 inches (25 centimeters). Northern Mississippi and southern Tennessee and parts of North Carolina are forecast to get 2 to 4 inches (5 to 10 centimeters) of rain.___5 p.m.Officials say the 42-bridge roadway that connects the Florida Keys to each other and the mainland must be checked for safety before motorists can be allowed back onto the islands.Florida Gov. Rick Scott said Monday that once officials are able to inspect, and to clear debris and sand from the Overseas Highway, it should be usable again.___5 p.m.Officials are reporting a second death in Georgia related to Tropical Storm Irma.Georgia Emergency Management Agency spokeswoman Catherine Howden said the death was confirmed Monday in Sandy Springs, north of Atlanta. She said she had no further details.The storm is also being blamed for the death of a 62-year-old man in rural southwest Georgia. Worth County sheriff’s spokeswoman Kannetha Clem said the man use a ladder to climb onto a shed Monday morning as sustained winds in the county exceeded 40 mph (65 kph).Clem says the man’s wife called 911 saying he suffered a heart attack, and first responders found his body lodged between two beams on the shed’s roof with debris on top of him.The dead man’s name was not immediately released.___4 p.m.At least one of the Orlando, Florida, theme parks popular with tourists around the world has plans to reopen now that Hurricane Irma has moved out of the state.Universal Orlando said Monday that all three of its parks will reopen at 9 a.m. Tuesday. Universal was closed down ahead of Irma.Universal said its facility suffered relatively minor damage to fences, trees and building facades.Disney World, Sea World and Busch Gardens Tampa Bay said Monday that they assessing damage and would announce their reopening plans later. All three theme parks said they never lost power. Sea World also said all of its animals are safe.___3:45 p.m.As South Carolina’s governor was issuing warnings about Tropical Storm Irma for the state’s residents, the storm toppled a massive oak tree on an apartment building he owns.Gov. Henry McMaster says the tree fell on a building he owns in Columbia around noon Monday.McMaster says the college students living at the apartments are safe. The governor says “no one suspected it might fall,” but the tree destroyed two apartments in the two-story building.The Columbia Fire Department says the fallen tree has left up to eight people without a home, but no one was injured.One displaced resident told WIS-TV the tree crashed through her apartment to the one below, taking furniture with it.___3:30 p.m.Florida Gov. Rick Scott says the Navy has deployed the USS Iwo Jima, USS New York and the aircraft carrier Abraham Lincoln to help with search and rescue and “a lot of other things” in the state.Scott says he flew over the Keys and saw a lot of flood damage and boats that had washed ashore.He says there is “devastation” and he hopes everyone who stayed behind survived Irma. He said almost every mobile home park in the Keys had overturned homes.Scott also flew over the west coast of Florida on Monday and said the damage was not as bad as he thought it would be.__3:30 p.m.President Donald Trump’s homeland security adviser says Irma is still a dangerous storm despite being downgraded to a tropical depression.Tom Bossert says while Irma’s category of strength may have been reduced, its combined effects might replicate that of a more powerful storm. Irma was once rated at Category 5 storm, the most powerful on record.Bossert notes that Jacksonville, Florida, is experiencing some of the worst flooding it has seen in 100 years.He says Tennessee and Kentucky, both targets as Irma moves to the U.S. interior, could experience inland flooding.Bossert says his message to the millions of Floridians who evacuated before the storm hit is not to rush back home because conditions are still dangerous.3:20 p.m.Georgia officials say at least one person has been killed by Tropical Storm Irma.Georgia Emergency Management Agency spokeswoman Catherine Howden said Monday that one storm-related death has been confirmed in Worth County, about 170 miles (270 kilometers) south of Atlanta. She had no further details.The county is located in southwest Georgia, where Irma’s center was churning northwestward toward Alabama on Monday afternoon. With tropical storm winds extending more than 400 miles (640 kilometers) from its center, Irma has caused damage across the state from trees falling on inland homes to flooding in neighborhoods on the Georgia coast.The storm has also been blamed for one death in Florida. At least 36 people died in the storm’s wake across the Caribbean.___3 p.m.Communities along the Georgia coast are seeing extensive flooding from Tropical Storm Irma.Irma’s storm surge pushed water ashore at the high tide Monday afternoon, and heavy rainfall made the flooding even worse. On Tybee Island east of Savannah, Hollard Zellers saw waist-deep water in the street as he went to fetch a kayak.About 3,000 people live on Tybee Island, which is Georgia’s largest public beach. City manager Shawn Gillen said the waters seemed to be receding quickly, but most of the island appeared to have some level of flooding and water was in many homes.Storm surge also sent floodwaters into downtown St. Marys just north of the Georgia-Florida line. St. Marys police Lt. Shannon Brock said piers and boat docks were heavily damaged and many boats sunk.___3 p.m.A massive sinkhole opened up at the edge of an apartment building in Orange County, Florida, swallowing air-conditioning units and bushes and a concrete slab. The sinkhole destabilized the building so seriously that firefighters evacuated dozens of residents amid the hurricane’s winds and pouring rain.Ronnie Ufie heard a loud bang and her 6-year-old grandson saw sparks shoot up behind the building, then their power flickered out.The fire alarm started screaming.Ernest Almonor, who lives next door to Ufie, ran outside but saw no fire and went back inside.But firefighters arrived and told them they had to leave the building. Ufie, who cares for her two young grandsons, grabbed some coloring books and crayons and headed through the rain for a neighbor’s house.But most residents, around 25 people, ended up scrambling through the storm to hunker for the night in the complex’s clubhouse.___3 p.m.Dutch King Willem-Alexander has flown to St. Maarten to see firsthand the devastation wreaked by Hurricane Irma on the tiny Caribbean territory and express gratitude to relief workers struggling to deliver aid and start the process of rebuilding shattered communities.Images broadcast by Dutch news outlets showed the king, wearing sunglasses and a khaki shirt with the sleeves rolled up, touring the badly damaged Princess Juliana International Airport . The airport, named for his grandmother, has become a vital hub for flights bringing in relief supplies as well as a gathering point for tourists and residents waiting to leave the island in the aftermath of last week’s devastating direct hit by Irma.Later Monday, the king was scheduled to visit the hospital in the capital, Philipsburg, and a school that is being used as a coordination center for distributing aid. Willem-Alexander also was expected to meet police and troops who have been struggling to maintain order on St. Maarten, where widespread looting broke out after Irma had passed.St. Maarten is part of the Kingdom of the Netherlands, but has had broad autonomy since 2010.After spending the night in St. Maarten, the king is flying Tuesday to two nearby Dutch islands, Saba and St. Eustatius, which also were hit by Irma, but suffered less damage.___2:30 p.m.State and federal environmental regulators have issued a blanket waiver for Florida electricity companies to violate clean air and water standards for the next two weeks.The Environmental Protection Agency announced the decision in a letter issued Monday as Hurricane Irma blew through the state. The agency said the so-called No Action Assurance granted through Sept. 26 will provide Florida utility generators needed flexibility to maintain and restore electricity supplies.The assurance letter will allow utilities to operate outside restrictions mandated by their permits, including potentially using dirtier fuels, running for longer hours or electively bypassing pollution-control equipment.The Associated Press reported last week that air pollution levels spiked in the Houston area after a similar enforcement waiver was granted to petrochemical facilities ahead of Hurricane Harvey.___2 p.m. In Florida, Hurricane Irma has brought dangerous floodwaters, knocked out power to millions and turned human possessions into debris in the past 24 hours. After making landfall on Sunday, the huge storm remained a Category 1 hurricane as it moved over the state’s northwest early Monday, before finally being downgraded to a tropical storm.Irma has maximum sustained winds of 60 mph, with stronger gusts. At 2 p.m. ET, Irma’s center was about 50 miles south-southeast of Albany, Ga., and 55 miles east of Tallahassee.National Hurricane CenterIrma is now a tropical storm — but it still poses severe flooding threats across Florida and into neighboring states. The storm’s predicted path is seen here in forecasters’ 11 a.m. ET release.Flash flood warnings have been issued in Jacksonville — a victim, like other eastern Florida areas, of the heaviest rains that are commonly found in hurricanes’ northeast quarter. Flood and tornado warnings have been issued for parts of coastal Georgia and South Carolina.More than 6 million electricity accounts in Florida are currently without power, state officials say. While the full extent of Irma’s damage isn’t yet known, the storm has weakened at a faster rate than expected.As the sun rose Monday, many Floridians anxiously began to assess the damage wrought by the hurricane, either inspecting their houses and neighborhoods for themselves or contacting those who stayed behind.Full radar loop showing the trek of Hurricane #IRMA through the #FL peninsula since very early Sunday morning. #flwx #NWS #NHC pic.twitter.com/IgEJZ2gexU— NWS Tampa Bay (@NWSTampaBay) September 11, 2017For those caught in Irma’s path — and wondering what to do after it passes — member station WLRN in Miami has assembled a guide to help.Irma was a hurricane for part of 12 days, having been dubbed a hurricane on Aug. 31, when it was far from land in the Atlantic Ocean. For days, it was a Category 5 storm, wreaking havoc in the Caribbean, where recovery efforts are still underway.On Monday morning, the storm brought storm surge risks as high as 4 to 6 feet above normal water levels in parts of Florida. Irma was also extending tropical-storm-force winds outward up 415 miles, the National Hurricane Center says.The perilous storm and the massive evacuation it sparked were reflected in an unusual scene in Florida early Monday, when the state’s skies were empty of airliners. A screenshot taken by journalist Sam Sweeney shows, as he wrote, “not a single airplane over the state of Florida.”7:24am EST Monday– There is not a single airplane over the state of Florida. pic.twitter.com/fWGwU5VkFP— Sam Sweeney (@SweeneyABC) September 11, 2017Irma is moving north-northwest at nearly 17 mph — a motion that’s expected to continue through Tuesday. The storm’s center is forecast to move near the Florida Peninsula’s northwestern coast and cross the eastern Florida Panhandle into southern Georgia Monday afternoon. It will then move through southwestern Georgia and eastern Alabama through the night and into Tuesday. National Hurricane CenterHurricane Jose, which had shadowed part of Irma’s route toward the Leeward Islands, has veered off north, as expected — but forecasters say the storm will follow an odd curlicue path over the next several days, putting it south of its current position and possibly threatening parts of the Bahamas this weekend.When it arrived in Florida, Irma was more than 400 miles wide; the storm remains huge. Consider that when it made landfall at the bottom of Florida’s peninsula on Sunday, Irma’s thick bands of rain were already drenching parts of northern Florida and southern Georgia.Hurricane Irma made landfall twice in Florida on Sunday, smacking into the Florida Keys as a Category 4 storm before moving over water and hitting Marco Island as a Category 3 on Florida’s southwest coast.Here are some of the stories we’re seeing in Florida:JacksonvilleThe city has already seen bad flooding, as the St. Johns River overflowed its banks — and it’s going to get worse around 2 p.m., when high tide will come, the Jacksonville Sheriff’s Office says.Historic flood levels already. Levels will continue to rise. Expect 4-6 feet above normal high tide levels. You need to be concerned. https://t.co/doN3zYLezh— Jax Sheriff’s Office (@JSOPIO) September 11, 2017The sheriff’s office writes Monday morning: “Historic flood levels already. Levels will continue to rise. Expect 4-6 feet above normal high tide levels. You need to be concerned.”Officials had been calling for people to leave areas along the St. Johns River, and to contact emergency numbers for help leaving. To those who didn’t leave, the sheriff’s office says, “If you can’t get out, you need to get to a house with a second story.”Key WestThe far-flung island lost power and phone connections, but some news began to emerge from Key West late Sunday night. Photographer Rob O’Neal relayed an account of damage, highlighted by flooding, and downed trees and utility poles. Some buildings had also lost their rooftops.“Counted 10-20 utility lines & 20 trees down in Oldtown. Downed trees & tree limbs scattered throughout KW. Counted at least 100 ppl walking around outside after eye wall passed,” O’Neal said.“Heaviest damage from Marathon to Lower Keys,” O’Neal said.Thousands of people are believed to have resisted mandatory evacuation orders to stay in the Keys. Sunday night, Monroe County Sheriff Rick Ramsay ordered a dusk-to-dawn curfew for the Keys.OnlinePeople in the hurricane’s path are using a Facebook group to check in and mark themselves as safe, with nearly 400,000 people in the group as of Monday morning, offering peace of mind to loved ones and giving a hint at this storm’s reach.Members of the group also used it to commiserate, share tips, and vent their feelings about Irma.As the sun rose Monday, they also shared some of what they’re seeing. A sample, from Davenport, in central Florida, where Benjamin McKinney wrote:“Walked around to check out my yard. The neighbor’s screen patio was torn off and is laying in the street. The stop sign on the corner is missing. The neighbor behind us fence is down. We’re missing some pieces from our roof overhang. Our boughanvilla bush was uprooted and pulled our fence out. Nothing too crazy.”TampaAs the storm approached, it sucked water out of Tamp Bay and other areas, prompting curious residents to head out onto the muddy ground — and, in at least are, to rescue stranded manatees.The hurricane center warned on Sunday, “Don’t be fooled by low water conditions on the FL W coast ahead of Irma, water will rapidly surge back when the winds change direction.”Around midday on Monday, the National Weather Service office in Tampa extended flood warnings for at least 10 rivers, saying they’re close to flood levels — and will rise further.“Significant river flooding will continue over the next several days as heavy rainfall from Hurricane Irma drains into West Central Florida rivers,” the agency says.OrlandoFrom member station WMFE:“Florida Highway Patrol said they’re seeing too many people out ‘sightseeing’ and not obeying the curfew. FHP reports trees are down and there’s standing water from the exit ramps at 528 to 520. Those ramps have been closed in both directions… “FHP says traffic’s starting to build up on 95 and the Turnpike southbound. They expect traffic to pick up as the day goes on and people want to go home and check out their property. Officials say people can’t enter the Keys, so evacuees need to stay put.”MiamiThe city is focusing on clearing roads of debris trees, and sand — in some areas, parts of the beach have been pushed over roadways. Mayor Tomás Regalado said 72 percent of the city of Miami is currently without power, NPR’s Hansi Lo Wang reports. Police report at least 26 looting incidents, with 13 arrests made.“At one time, Miami was dead in the center of the track, and could have been hit very hard,” NPR’s Jon Hamiton says on Morning Edition. But today, Jon says, much of the city looks “surprisingly normal” — at least out toward the airport.“I have seen a whole lot of power trucks going by on the freeway out here,” Jon adds. “There’s still water on the ground in places, they’re picking up lots of palm fronds, but the city looks like it’s getting back in business.”Miami International Airport saw wind gusts near 100 mph and “sustained significant water damage throughout,” the airport director, Emilio Gonzalez, said via Twitter. The facility is closed to passenger flights at least through Monday.RT StevePetyerak: #Irma evacuees emptying out GermainArena in Estero, FL pic.twitter.com/V3WZdugcMR— Zachary’s Weather (@ZacharysWeather) September 11, 20179:45 a.m.Actress Kristen Bell says she’s “singing in a hurricane” while riding out Irma in Florida.The “Frozen” star is in Orlando filming a movie and staying at a hotel at the Walt Disney World resort. She stopped by an Orlando middle school that was serving as a shelter and belted out songs from “Frozen.”Back at the hotel, Bell posted pictures on Instagram of her singing with one guest and dining with a group of seniors.Bell also helped out the parents of “Frozen” co-star Josh Gad by securing them a room at the hotel.Bell tells Sacramento, California, station KMAX-TV — where her father is news director — that the experience is her version of one of her favorite movies, “Singin’ in the Rain.”___9:30 a.m.People are being rescued from flooded homes Monday morning south of Jacksonville, Florida, as Tropical Storm Irma pounds the state with rain and wind.John Ward, the emergency operations manager of Clay County, says crews have pulled 46 people from flooded homes by early Monday and an undetermined number are still stranded as the area’s creeks and ponds are getting record flooding.Ward says between 400 and 500 homes received severe flood damage but there have been no serious injuries or deaths.Irma weakened to a tropical storm Monday morning, a day after hitting the state as a powerful Category 4 hurricane.WATCH: Police in Lakeland, Florida, show high winds and rain in the city right now as the eye of Hurricane #Irma moves through the area. pic.twitter.com/6PrNyDS3lN— NBC Nightly News (@NBCNightlyNews) September 11, 2017___9:05 a.m.An Atlanta airport official says there’ve been around 800 cancellations due to the threat of Irma, which weakened to a tropical storm Monday morning.Hartsfield-Jackson Atlanta International Airport spokesman Andrew Gobeil says the airport will still be operational Monday and will monitor storm conditions.Gobeil says the airport created an overflow parking plan to allow planes unable to land in areas such as Florida to park at the airport in Atlanta.Also in Atlanta, the city’s transit system has suspended all bus and rail service ahead of the weather conditions caused by Irma.Metropolitan Atlanta Rapid Transit Authority spokesman Erik Burton says both systems will be closed for Monday. He says officials will continue to coordinate with state and local officials along with emergency personnel to determine MARTA’s service schedule for Tuesday.___9:05 a.m.Much of central Florida, including Orlando, suffered significant damage as Irma blew through Sunday night and into Monday morning.Orange County Mayor Teresa Jacobs said Monday morning that there’s been widespread damage and significant power loss throughout the area.Jacobs said approximately 300,000 residents in Orlando are without power. She also said 60 percent of the fire stations are operating on backup generators and dispatchers received 1,381 calls between Sunday at midnight and 5:45 a.m. Monday morning.Residents are being asked to minimize usage such as flushing toilets, bathing, along with washing dishes and laundry.___8:30 a.m.Irma is causing record-setting flooding in Jacksonville, Florida, as it moves over the state Monday on its way to southern Georgia.The National Weather Service in Jacksonville says storm surge flooding exceeds the record set in 1965 during Hurricane Dora.The city on Florida’s northeastern Atlantic coast remains under a flash flood warning until 12:45 p.m. Monday.Jacksonville Sheriff’s officials warned residents not to drive through standing water, adding that in addition to flooding, there are trees and power lines down across the area.Irma weakened to a tropical storm Monday morning, a day after hitting the state as a Category 4 hurricane.___8:30 a.m.Florida Gov. Rick Scott says there’s damage across the state caused by Hurricane Irma and it’s still too dangerous for residents to go outside or return from evacuation.Scott said Monday on Fox News that he’s concerned about flooding now unfolding in Jacksonville and the amount of damage in the Florida Keys. The governor will be flying out of Mobile, Alabama, on a U.S. Coast Guard plane down to the Keys where he plans to inspect the extent of the damage there.Local officials in the Keys on Sunday warned of a humanitarian crisis due to Irma’s winds and storm surge.Scott asked Floridians to be patient and warned that roads are impassable and that there are downed power lines.___8:30 a.m.Two children evacuated from Florida as Hurricane Irma approached have been killed in separate car crashes in Georgia.News outlets report Woodstock police say 3-month-old Riley Hunt of Port St. Lucie, Florida, was struck by an SUV driven by a 17-year-old girl Saturday night and was later pronounced dead. DeKalb County police say an 11-year-old boy who was also traveling from Florida to Georgia was hit and killed by a car early Monday in Stone Mountain.The identity of the boy hasn’t been released, and the cause of the crash is currently unknown.The Cherokee County Sheriff’s Office says Hunt’s mother, 28-year-old Kaitlyn Hunt, and 61-year-old Kathy Deming were also hit and are listed in critical condition. The incident remains under investigation. No charges had been filed.___8 a.m.Irma has weakened to a tropical storm as it moves over Florida toward southern Georgia.The storm’s maximum sustained winds decreased Monday morning to near 70 mph (110 kph). The U.S. National Hurricane Center says it’s expected to weaken to a tropical depression by Tuesday afternoon.Irma is centered about 105 miles (170 kilometers) north-northwest of Tampa, Florida, and is moving north-northwest near 18 mph (30 kph).Irma hit southern Florida on Sunday as a Category 4 hurricane, bringing roof-ripping winds, gushing floodwaters and widespread power outages.___7:45 a.m.The National Guard and local fire rescue teams evacuated more than 120 flooding homes in Orange County, just outside Orlando, as Hurricane Irma moves over Florida.Trees and power lines were down across town and floods cut off roads to a neighborhood.As the sun rose in Orlando, many tried to go outside to survey the damage, but authorities warn that conditions remain dangerous and ask that people to abide by the curfew that lasts throughout most of the day.___7:45 a.m.Winds and rain from Hurricane Irma have moved into South Carolina and officials are warning residents to be very careful throughout the day.A hurricane watch is in effect Monday from Edisto Beach into Florida. A storm surge warning and a tropical storm warning are in effect from near Georgetown into Florida.A flash flood warning is in effect along the southern coast of South Carolina, where more than 40,000 were ordered to evacuate barrier islands.The storm surge could reach 6 feet (2 meters), especially from late morning to mid-afternoon. Up to 6 inches (15 centimeters) of rain is also possible.Wind gusts of up to 60 mph (97 kph) are expected along the South Carolina coast. Forecasters say tornadoes are also possible.South Carolina Electric & Gas reported more than 13,000 customers without service Monday morning.___7:45 a.m.The European Union says it has released 2 million euros ($2.4 million) to help the Caribbean islands hit hardest by Hurricane Irma.EU’s humanitarian aid commissioner, Christos Stylianides, announced the assistance Monday, saying the money will go to key sectors such as health, water and sanitation and waste management.The EU has already been involved in the emergency relief effort, and Stylianides said the bloc stands ready to provide longer-term assistance as well. He called it “our moral duty to help those in need whose lives and homes are being destroyed or severely threatened.”___7:10 a.m.Nearly 4.5 million homes and businesses across Florida have lost power as Hurricane Irma moves over the state.And utility officials say it will take weeks to restore electricity to everyone. Farther north, more than 100,000 are in the dark in Georgia.Much of eastern Alabama and coastal South Carolina are under tropical storm warnings as Irma pummels Florida, weakening on its march northward.The National Hurricane Center predicts the storm will cross Monday into southwest Georgia, where a hurricane warning was in effect for a large rural area including the cities of Albany and Valdosta.Rain already is falling in parts of the state, including metro Atlanta, early Monday.___7 a.m.Tampa Mayor Bob Buckhorn says that while the city hasn’t escaped Hurricane Irma’s wrath, the situation isn’t as bad as they had feared.Speaking Monday morning on MSNBC, Buckhorn said “What we thought was going to be a punch in the face was a glancing blow.”Buckhorn did say there are a lot of downed power lines and debris.He said Tampa’s officials have vehicles positioned “to be sure that when that surge comes in we can keep people out of the streets.”He said he expected power to be out for some sections of Tampa for at least a couple more days.Hurricane Irma is getting weaker as it moves over the western Florida peninsula after hitting the state Sunday as a Category 4 storm.___6:45 a.m.Police in Miami are investigating reports of people looting stores as Hurricane Irma hit the state.On Sunday night, Miami police took two people into custody and detained two others.Deputy Police Chief Luis Cabrera told the Miami Herald the officers went to the Shops at Midtown on Sunday afternoon as the winds of Hurricane Irma were at their strongest in South Florida. Cabrera says a group in a white truck hit multiple locations. Police have also received additional reports of looting in the city.Police had issued a curfew Saturday night, partly to ward off looters by giving officers probable cause to stop anyone for being on the street during the storm.Cabrera didn’t have specific details about the looting incidents.___6:45 a.m.The British government is defending its response to Hurricane Irma amid claims it has been slow to help its overseas territories devastated by the storm.The British Virgin Islands, Anguilla and the Turks and Caicos islands were all pummeled by the hurricane last week, leaving thousands without electricity or shelter.Opposition politicians have compared Britain’s response unfavorably to that of France, which has sent more than 1,000 troops, police and emergency workers to St. Martin and St. Barts.Britain has dispatched a navy ship and nearly 500 troops, including medics and engineers.Foreign Secretary Boris Johnson said Monday that Britain had responded strongly to an “unprecedented catastrophe.” He says the government will soon increase the 32 million pounds ($42 million) it’s pledged to the relief effort.___6:30 a.m.Police in Lakeland, Florida, say a family with small children was rescued from a car that was submerged in water as Hurricane Irma crossed the area.Lakeland police said in a Facebook post that officers rescued the family of four early Monday as water reached the children’s car seats. No one was injured and police were able to get the family back to their home.“When you become a police officer you hope to make a difference in the lives of others,” the Facebook post said. “Tonight, there is no doubt these officers made a difference.”Lakeland is between Tampa and Orlando, off of Interstate 4.Hurricane Irma is getting weaker as it moves over the western Florida peninsula after hitting the state Sunday as a Category 4 storm.___6:30 a.m.A Florida sheriff’s sergeant and a paramedic were trapped in a sheriff’s vehicle when a live power pole fell on the cruiser as they were returning from dropping off an elderly patient as Hurricane Irma moved over the state.Polk County spokesman Kevin Watler said in a news release that Sgt. Chris Lynn and Polk County Fire Rescue paramedic James Tanner Schaill were trapped for about two hours late Sunday.Crews from Lakeland Electric crews disconnected the lines around 1:15 a.m. Monday. Both men have returned to their jobs to continue assisting hurricane recovery efforts.___6:10 a.m.More than 120 homes are being evacuated in Orange County, just outside Orlando, as floodwaters from Hurricane Irma started to pour in.The Orange County Emergency Operations Center said early Monday that the fire department and the National Guard are going door-to-door using boats to ferry families to safety. No injuries have been reported. The rescued families are being taken a shelter for safety.A few miles away, 30 others had to be evacuated when a 60-foot sinkhole opened up under an apartment building. No injuries were reported in that incident.Hurricane Irma is getting weaker as it moves over the western Florida peninsula after hitting the state Sunday as a Category 4 storm.___5 a.m.Hurricane Irma is getting weaker as it moves over the western Florida peninsula early Monday.Irma hit Florida on Sunday as a powerful Category 4 hurricane, hammering much of the state with roof-ripping winds, gushing floodwaters and widespread power outages.By Monday morning, Irma had weakened to a Category 1 hurricane with winds near 85 mph (135 kph). Additional weakening is forecast and Irma is expected to become a tropical storm over northern Florida or southern Georgia later in the day.___4:45 a.m.Dutch search and rescue experts are heading to the shattered former colony of St. Maarten to support the humanitarian relief effort in the aftermath of Hurricane Irma.A team of 59 urban search and rescue experts is flying Monday to the Dutch territory that’s home to some 40,000 people, where 70 percent of homes were badly damaged last week by a direct hit from the Category 5 storm. Four people were killed and dozens injured.The Dutch government also is sending extra troops to maintain order following widespread looting and robberies. The government says there are already nearly 400 extra troops in St. Maarten and that number will rise to some 550 over the next two days.Dutch King Willem-Alexander is expected to visit the island Monday to show his support for local residents and the emergency services working to restore infrastructure and begin the process of reconstruction.___2 a.m.Irma weakened to a Category 1 storm as the massive hurricane zeroed in on the Tampa Bay region early Monday after hammering much of Florida with roof-ripping winds, gushing floodwaters and widespread power outages.The hurricane’s maximum sustained winds weakened to 85 mph (135 kph) with additional weakening expected.As of 2 a.m. EDT, the storm was centered about 25 miles (40 kilometers) northeast of Tampa and moving north-northwest near 15 mph (24 kph).Irma continues its slog north along Florida’s western coast having blazed a path of unknown destruction. With communication cut to some of the Florida Keys, where Irma made landfall Sunday, and rough conditions persisting across the peninsula, many are holding their breath for what daylight might reveal.___An earlier version of this report, incorrectly said Tampa Mayor Bob Buckhorn’s comments were made on MSNBC’s “Morning Joe” show. Hurricane Jose is expected to make an unusual curlicue maneuver in the Atlantic as it passes north of Puerto Rico. By this weekend, it will be close to the northern Bahamas. Enlarge this image Sharelast_img read more

Bryson Starts Youth Sports Coalition to Improve PG County Athletics

first_imgBy Mark F. Gray, AFRO Staff Writer, mgray@afro.comDr. Rick Bryson has long been an advocate for youth sports and participatory activities in Prince George’s County. Bryson has put his money where his money mouth is by creating the County’s Scholastic Sports Awards Banquet and has now begun a new initiative designed to improve the quality of youth and participatory sports.Dr. Bryson’s latest cause is the creation of the Youth Scholastic Coalition, designed to support and advocate for the activities kids participate in while collaborating with the scholastic support network for young athletes throughout Prince George’s County. Building bridges for youth and scholastic sports athletes with an academic support network has been a passion for the longtime podiatrist since his days as a baseball dad when his son was growing up and competing in Accokeek.The PG Youth Sports Coalition hopes to improve the quality of baseball field throughout the county. (Courtesy Photo)During that time Dr. Bryson built a relationship with St. Paul’s United Methodist Church in Fort Washington. For three years youth baseball players from the District and Prince George’s County participated in an academic development program that served as the genesis for the establishment of the coalition. They competed on the field while “barnstorming” throughout the DMV for games during the summer.However, the players also participated in a Scholastic Aptitude Testing Program at the church for 12 weeks over three years. The synergy between the athletics and academics for that group of players helped prepare them for the road to college. Several players part of the initial group would head to schools such as the U.S. Air Force Academy and Morehouse College to pursue their college degrees not just as athletes.“Sports is one thing, but academics will carry you which is what we’re trying to stress,” Bryson told the {AFRO}. “It was a lot easier to get the kids to participate in the learning phase when their peers were a part of the program.”There are many organizations in the community that provide academic and athletic options for youth as individuals and the Coalition is working to connect students with groups that will able to provide additional information and services. For example, organizations with athletic–based programs can expose their participants to services offered by an academic-based program. Those programs can include everything from college preparatory test tutoring to advising families about the NCAA eligibility clearinghouse requirements for potential college student-athletes.“Most groups are focused on the specific activities of their respective organizations, but the Coalition’s focus is on addressing issues that concern everyone,” Bryson said. “The basic premise of establishing the Coalition is the belief that collectively, organizations can be more effective in addressing concerns and issues that have an impact on the community, if there is a positive atmosphere of collaboration.”The Coalition is also serving as the advocate for the various needs related to youth sports programs in P.G. County. The Coalition assists by lobbying to improve facilities and interacts with various state and county officials to engage them in addressing issues of concern for high school sports participants as well. In addition to providing academic support and access to potential scholarship opportunities, one of its goals is to make access to quality fields a priority for all sports participants.The greatest challenge that faces the Coalition is developing corporate partnerships to help sponsor the programs Dr. Bryson and his group are working to establish. Grantors typically look favorably on civic organizations that partner with other groups to address community needs and services. The Coalition helps facilitate these partnerships.“We are aware of the various needs related to youth programs,” Bryson said. “We are working to identify grant-funding sources from various government and non-profit organizations.”last_img read more

Hacking Doomsday Your Cyberattack Survival Checklist

first_img 6 min read Opinions expressed by Entrepreneur contributors are their own. On Oct. 21, 2016, the largest distributed denial of service attack (DDoS) in history was launched at around 7 a.m. EDT. It nailed servers belonging to a company called Dyn, an internet performance-management company that directs traffic for major sites.The Dyn attack caused a severe disruption of internet traffic to major sites, including many you may be familiar with, like Amazon, Reddit, Twitter, Tumblr, Verizon, Pinterest, Etsy, Spotify, PayPal, Comcast and even Playstation. The first attack was followed by at least two more.Back in 2012, there was a DDoS attack that took out the websites of JPMorgan Chase, Bank of America, Wells Fargo, Citigroup and other financial companies. In 2007, a DDoS attack literally knocked most of Estonia off the grid.How do these attacks work? The hackers look for a distribution model that can deliver sufficient traffic — in this latest case, a webcam that had sold well and had easily exploitable security. Then they focus on a target. The target here seems to have been companies that act as the internet’s phone book, making sure that traffic requests find the fastest route to any particular destination.If you’re a fan of The Walking Dead, imagine a huge herd of zombies lumbering toward a valley with three fenced communities. While all three are in harm’s way, the one with the weakest fence will be overrun. The other two are more likely to withstand the onslaught.Another way to imagine this scenario is a subway car or bus contaminated with a cold virus. While everyone on board is exposed, not everyone will get sick. The virus goes where it encounters the least amount of resistance, which in this most recent attack happened to be Dyn.We can make this a problem of the past.While our government has disappointed time and again, in particular on matters of cybersecurity, a DDoS solution may possibly be only a rule or two away from becoming reality. That is, of course, if the new administration truly focuses on the issue.In that case Congress will actually agree on the day of the week to discuss the issue, and we can all say adios to the debate about whether the “Great Wall of Mexico” will be the answer to most of our problems.Related: Expert Hacker Shares 3 Ways Small Businesses Can Minimize Cybersecurity Threats The solution may lie in the way many ATMs now handle the new chip technology in credit and debit cards. When you make a request for cash or any other transaction, many machines will not release your card until you grab your cash or completely finish your transaction. This ensures that consumers don’t leave their cards in machines, which of course can create a very real vulnerability.When it comes to the Internet of Things (IoT) devices, the six billion (and geometrically increasing) connected and interconnected devices that surround us in our personal and business lives, the same principle might help. However, it would have to become the law of the land to work — a law that would include authority over trade, i.e., products coming in from overseas.Meanwhile, the “insert and release only when the transaction is completed” ATM process is a security measure designed to better protect your card, your accounts and your financial institution. If all IoT devices required the user to set a long and strong password before the device would function, a major vulnerability available for use by hackers in DDoS attacks would go the way of three-card monte scams.In the meantime, there are a few things you can do to put your organization in the best position to survive a DDoS attack. While these measures are somewhat technical, they are within the skill sets of whomever is in charge of your digital security.1. Identify a DDoS attack early.You can do this with the right Intrusion Detection System (IDS) or a state-of-the art firewall that uses “Stateful Inspection.” These are techniques that can alert you to an attack and allow you to respond before systems fail. Another good move: having a technology professional that is capable of identifying traffic that looks like a DDoS attack and is ready to respond by immediately applying filters to the inbound DDoS traffic.Related: 7 Cybersecurity Layers Every Entrepreneur Needs to Understand 2. Have plenty of bandwidth available.Bandwidth is pretty cheap, and if you are able to over-provision what you need, you can ride out smaller attacks or at least buy some time to react properly to larger attacks.3. Have a rule book.Apply the right technical defenses at your perimeter. For example, you could “rate-limit” your web server so that only so many traffic requests are allowed. This will keep your server from getting swamped. Doubtless, it will still slow down during a DDoS attack, but it won’t fail.There are other things you can do that are beyond a generalist’s knowledge, but are straightforward for a professional who handles the engine room of your online presence. One strategy is to add network rules that dump requests from suspicious places or deny all traffic that doesn’t come from trusted sources.Your IT team may decide it makes sense to time-out suspicious connections, drop malformed packages or set lower flooding thresholds on certain types of traffic. It is crucial to have these rules — however you choose to throttle them — applied to all routers and firewalls and even internal internet-working components to filter out the most common kinds of DDoS-type traffic.4. Get your provider involved.Your service provider will have better tools than anyone else when it comes to locking down the DDoS traffic headed your way. Alert the provider as soon as you see something. The company can even “null route” your address (a network route that goes nowhere) so you will see no DDoS traffic at all while it figures out how to neutralize the attack.5. Outsource your traffic.There are technical firms that specialize in filtering traffic for you in an emergency. When a DDoS attack occurs, all your traffic moves through one of their routers first, and they are very good at blocking all the DDoS traffic that would be coming your way.Related Offer: Experience ESET’s award-winning antivirus software right now with a 30-day free trial.While you cannot prevent someone from launching a DDoS attack against your company, you can be prepared to mitigate its impact. The key thing is to assume that it’s going to happen and be as ready for it as possible. Free Webinar | Sept 5: Tips and Tools for Making Progress Toward Important Goals December 7, 2016 Attend this free webinar and learn how you can maximize efficiency while getting the most critical things done right. Register Now »last_img read more

Generative Adversarial Networks Generate images using Keras GAN Tutorial

first_imgYou might have worked with the popular MNIST dataset before – but in this article, we will be generating new MNIST-like images with a Keras GAN. It can take a very long time to train a GAN; however, this problem is small enough to run on most laptops in a few hours, which makes it a great example. The following excerpt is taken from the book Deep Learning Quick Reference, authored by Mike Bernico. The network architecture that we will be using here has been found by, and optimized by, many folks, including the authors of the DCGAN paper and people like Erik Linder-Norén, who’s excellent collection of GAN implementations called Keras GAN served as the basis of the code we used here. Loading the MNIST dataset The MNIST dataset consists of 60,000 hand-drawn numbers, 0 to 9. Keras provides us with a built-in loader that splits it into 50,000 training images and 10,000 test images. We will use the following code to load the dataset: from keras.datasets import mnistdef load_data(): (X_train, _), (_, _) = mnist.load_data() X_train = (X_train.astype(np.float32) – 127.5) / 127.5 X_train = np.expand_dims(X_train, axis=3) return X_train As you probably noticed, We’re not returning any of the labels or the testing dataset. We’re only going to use the training dataset. The labels aren’t needed because the only labels we will be using are 0 for fake and 1 for real. These are real images, so they will all be assigned a label of 1 at the discriminator. Building the generator The generator uses a few new layers that we will talk about in this section. First, take a chance to skim through the following code: def build_generator(noise_shape=(100,)): input = Input(noise_shape) x = Dense(128 * 7 * 7, activation=”relu”)(input) x = Reshape((7, 7, 128))(x) x = BatchNormalization(momentum=0.8)(x) x = UpSampling2D()(x) x = Conv2D(128, kernel_size=3, padding=”same”)(x) x = Activation(“relu”)(x) x = BatchNormalization(momentum=0.8)(x) x = UpSampling2D()(x) x = Conv2D(64, kernel_size=3, padding=”same”)(x) x = Activation(“relu”)(x) x = BatchNormalization(momentum=0.8)(x) x = Conv2D(1, kernel_size=3, padding=”same”)(x) out = Activation(“tanh”)(x) model = Model(input, out) print(“– Generator — “) model.summary() return model We have not previously used the UpSampling2D layer. This layer will take increases in the rows and columns of the input tensor, leaving the channels unchanged. It does this by repeating the values in the input tensor. By default, it will double the input. If we give an UpSampling2D layer a 7 x 7 x 128 input, it will give us a 14 x 14 x 128 output. Typically when we build a CNN, we start with an image that is very tall and wide and uses convolutional layers to get a tensor that’s very deep but less tall and wide. Here we will do the opposite. We’ll use a dense layer and a reshape to start with a 7 x 7 x 128 tensor and then, after doubling it twice, we’ll be left with a 28 x 28 tensor. Since we need a grayscale image, we can use a convolutional layer with a single unit to get a 28 x 28 x 1 output. This sort of generator arithmetic is a little off-putting and can seem awkward at first but after a few painful hours, you will get the hang of it! Building the discriminator The discriminator is really, for the most part, the same as any other CNN. Of course, there are a few new things that we should talk about. We will use the following code to build the discriminator: def build_discriminator(img_shape): input = Input(img_shape) x =Conv2D(32, kernel_size=3, strides=2, padding=”same”)(input) x = LeakyReLU(alpha=0.2)(x) x = Dropout(0.25)(x) x = Conv2D(64, kernel_size=3, strides=2, padding=”same”)(x) x = ZeroPadding2D(padding=((0, 1), (0, 1)))(x) x = (LeakyReLU(alpha=0.2))(x) x = Dropout(0.25)(x) x = BatchNormalization(momentum=0.8)(x) x = Conv2D(128, kernel_size=3, strides=2, padding=”same”)(x) x = LeakyReLU(alpha=0.2)(x) x = Dropout(0.25)(x) x = BatchNormalization(momentum=0.8)(x) x = Conv2D(256, kernel_size=3, strides=1, padding=”same”)(x) x = LeakyReLU(alpha=0.2)(x) x = Dropout(0.25)(x) x = Flatten()(x) out = Dense(1, activation=’sigmoid’)(x) model = Model(input, out)print(“– Discriminator — “)model.summary()return model First, you might notice the oddly shaped zeroPadding2D() layer. After the second convolution, our tensor has gone from 28 x 28 x 3 to 7 x 7 x 64. This layer just gets us back into an even number, adding zeros on one side of both the rows and columns so that our tensor is now 8 x 8 x 64. More unusual is the use of both batch normalization and dropout. Typically, these two layers are not used together; however, in the case of GANs, they do seem to benefit the network. Building the stacked model Now that we’ve assembled both the generator and the discriminator, we need to assemble a third model that is the stack of both models together that we can use for training the generator given the discriminator loss. To do that we can just create a new model, this time using the previous models as layers in the new model, as shown in the following code: discriminator = build_discriminator(img_shape=(28, 28, 1))generator = build_generator()z = Input(shape=(100,))img = generator(z)discriminator.trainable = Falsereal = discriminator(img)combined = Model(z, real) Notice that we’re setting the discriminator’s training attribute to False before building the model. This means that for this model we will not be updating the weights of the discriminator during backpropagation. We will freeze these weights and only move the generator weights with the stack. The discriminator will be trained separately. Now that all the models are built, they need to be compiled, as shown in the following code: gen_optimizer = Adam(lr=0.0002, beta_1=0.5)disc_optimizer = Adam(lr=0.0002, beta_1=0.5)discriminator.compile(loss=’binary_crossentropy’,optimizer=disc_optimizer,metrics=[‘accuracy’])generator.compile(loss=’binary_crossentropy’, optimizer=gen_optimizer)combined.compile(loss=’binary_crossentropy’, optimizer=gen_optimizer) If you’ll notice, we’re creating two custom Adam optimizers. This is because many times we will want to change the learning rate for only the discriminator or generator, slowing one or the other down so that we end up with a stable GAN where neither is overpowering the other. You’ll also notice that we’re using beta_1 = 0.5. This is a recommendation from the original DCGAN paper that we’ve carried forward and also had success with. A learning rate of 0.0002 is a good place to start as well, and was found in the original DCGAN paper. The training loop We have previously had the luxury of calling .fit() on our model and letting Keras handle the painful process of breaking the data apart into mini batches and training for us. Unfortunately, because we need to perform the separate updates for the discriminator and the stacked model together for a single batch we’re going to have to do things the old-fashioned way, with a few loops. This is how things used to be done all the time, so while it’s perhaps a little more work, it does admittedly leave me feeling nostalgic. The following code illustrates the training technique: num_examples = X_train.shape[0]num_batches = int(num_examples / float(batch_size))half_batch = int(batch_size / 2) for epoch in range(epochs + 1):for batch in range(num_batches):# noise images for the batchnoise = np.random.normal(0, 1, (half_batch, 100))fake_images = generator.predict(noise)fake_labels = np.zeros((half_batch, 1))# real images for batch idx = np.random.randint(0, X_train.shape[0], half_batch)real_images = X_train[idx]real_labels = np.ones((half_batch, 1))# Train the discriminator (real classified as ones and generated as zeros)d_loss_real = discriminator.train_on_batch(real_images, real_labels)d_loss_fake = discriminator.train_on_batch(fake_images, fake_labels) d_loss = 0.5 * np.add(d_loss_real, d_loss_fake)noise = np.random.normal(0, 1, (batch_size, 100))# Train the generatorg_loss = combined.train_on_batch(noise, np.ones((batch_size, 1)))# Plot the progressprint(“Epoch %d Batch %d/%d [D loss: %f, acc.: %.2f%%] [G loss: %f]” %(epoch,batch, num_batches, d_loss[0], 100 * d_loss[1], g_loss))if batch % 50 == 0:save_imgs(generator, epoch, batch) There is a lot going on here, to be sure. As before, let’s break it down block by block. First, let’s see the code to generate noise vectors: noise = np.random.normal(0, 1, (half_batch, 100)) fake_images = generator.predict(noise) fake_labels = np.zeros((half_batch, 1)) This code is generating a matrix of noise vectors called z) and sending it to the generator. It’s getting a set of generated images back, which we’re calling fake images. We will use these to train the discriminator, so the labels we want to use are 0s, indicating that these are in fact generated images. Note that the shape here is half_batch x 28 x 28 x 1. The half_batch is exactly what you think it is. We’re creating half a batch of generated images because the other half of the batch will be real data, which we will assemble next. To get our real images, we will generate a random set of indices across X_train and use that slice of X_train as our real images, as shown in the following code: idx = np.random.randint(0, X_train.shape[0], half_batch)real_images = X_train[idx]real_labels = np.ones((half_batch, 1)) Yes, we are sampling with replacement in this case. It does work out but it’s probably not the best way to implement minibatch training. It is, however, probably the easiest and most common. Since we are using these images to train the discriminator, and because they are real images, we will assign them 1s as labels, rather than 0s. Now that we have our discriminator training set assembled, we will update the discriminator. Also, note that we aren’t using the soft labels. That’s because we want to keep things as easy as they can be to understand. Luckily the network doesn’t require them in this case. We will use the following code to train the discriminator: # Train the discriminator (real classified as ones and generated as zeros)d_loss_real = discriminator.train_on_batch(real_images, real_labels)d_loss_fake = discriminator.train_on_batch(fake_images, fake_labels)d_loss = 0.5 * np.add(d_loss_real, d_loss_fake) Notice that here we’re using the discriminator’s train_on_batch() method. The train_on_batch() method does exactly one round of forward and backward propagation. Every time we call it, it updates the model once from the model’s previous state. Also, notice that we’re making the update for the real images and fake images separately. This is advice that is given on the GAN hack Git we had previously referenced in the Generator architecture section. Especially in the early stages of training, when real images and fake images are from radically different distributions, batch normalization will cause problems with training if we were to put both sets of data in the same update. Now that the discriminator has been updated, it’s time to update the generator. This is done indirectly by updating the combined stack, as shown in the following code: noise = np.random.normal(0, 1, (batch_size, 100))g_loss = combined.train_on_batch(noise, np.ones((batch_size, 1))) To update the combined model, we create a new noise matrix, and this time it will be as large as the entire batch. We will use that as an input to the stack, which will cause the generator to generate an image and the discriminator to evaluate that image. Finally, we will use the label of 1 because we want to backpropagate the error between a real image and the generated image. Lastly, the training loop reports the discriminator and generator loss at the epoch/batch and then, every 50 batches, of every epoch we will use save_imgs to generate example images and save them to disk, as shown in the following code: print(“Epoch %d Batch %d/%d [D loss: %f, acc.: %.2f%%] [G loss: %f]” % (epoch,batch, num_batches, d_loss[0], 100 * d_loss[1], g_loss))if batch % 50 == 0:save_imgs(generator, epoch, batch) The save_imgs function uses the generator to create images as we go, so we can see the fruits of our labor. We will use the following code to define save_imgs: def save_imgs(generator, epoch, batch): r, c = 5, 5 noise = np.random.normal(0, 1, (r * c, 100)) gen_imgs = generator.predict(noise) gen_imgs = 0.5 * gen_imgs + 0.5 fig, axs = plt.subplots(r, c) cnt = 0 for i in range(r):for j in range(c): axs[i, j].imshow(gen_imgs[cnt, :, :, 0], cmap=’gray’) axs[i, j].axis(‘off’) cnt += 1 fig.savefig(“images/mnist_%d_%d.png” % (epoch, batch)) plt.close() It uses only the generator by creating a noise matrix and retrieving an image matrix in return. Then, using matplotlib.pyplot, it saves those images to disk in a 5 x 5 grid. Performing model evaluation Good is somewhat subjective when you’re building a deep neural network to create images.  Let’s take a look at a few examples of the training process, so you can see for yourself how the GAN begins to learn to generate MNIST. Here’s the network at the very first batch of the very first epoch. Clearly, the generator doesn’t really know anything about generating MNIST at this point; it’s just noise, as shown in the following image: But just 50 batches in, something is happening, as you can see from the following image: And after 200 batches of epoch 0 we can almost see numbers, as you can see from the following image: And here’s our generator after one full epoch. These generated numbers look pretty good, and we can see how the discriminator might be fooled by them. At this point, we could probably continue to improve a little bit, but it looks like our GAN has worked as the computer is generating some pretty convincing MNIST digits, as shown in the following image: Thus, we see the power of GANs in action when it comes to image generation using the Keras library. If you found the above article to be useful, make sure you check out our book Deep Learning Quick Reference, for more such interesting coverage of popular deep learning concepts and their practical implementation. Read Next Keras 2.2.0 releases! 2 ways to customize your deep learning models with Keras How to build Deep convolutional GAN using TensorFlow and Keraslast_img read more