Mhoro, Habr! Ndinokupa kutariswa kweshanduro yechinyorwa naKristencarter7519.
Kunyangwe zvichiita sekunge tangosara mazuva mashoma kubva muna 2020, mazuva ano akakoshawo mumunda wekuvandudza software. Pano mune ino chinyorwa, tichaona kuti iro rinouya gore ra2020 richachinja sei hupenyu hwevagadziri vesoftware.
Ramangwana rekuvandudza software rasvika!
Traditional software development ndiko kugadzirwa kwesoftware nekunyora kodhi uchitevera mimwe mitemo yakatarwa. Asi kuvandudzwa kwesoftware yemazuva ano kwakaona shanduko yeparadigm nekufambira mberi muhungwaru hwekugadzira, kudzidza kwemichina uye kudzidza kwakadzama. Nekubatanidza matekinoroji matatu aya, vanogadzira vanozokwanisa kugadzira zvigadziriso zvesoftware izvo zvinodzidza kubva mumirayiridzo uye nekuwedzera mamwe maficha uye mapatani kune data rinodiwa kuburitsa mhedzisiro inodiwa.
Ngatiedze neimwe kodhi
Nekufamba kwenguva, neural network software yekuvandudza masisitimu ave akanyanya kuomarara maererano nekubatanidzwa pamwe nemazinga ekushanda uye mainterface. Vagadziri, semuenzaniso, vanogona kuvaka yakapusa neural network nePython 3.6. Heino muenzaniso chirongwa chinoita mabhinari classification ne 1s kana 0s.
Ehe, isu tinogona kutanga nekugadzira neural network kirasi:
pinza NumPy seNP
X=np.array([[0,1,1,0],[0,1,1,1],[1,0,0,1]])
y=np.array([[0],[1],[1]])Kushandiswa kwebasa re sigmoid:
def sigmoid ():
return 1/(1 + np.exp(-x))
def derivatives_sigmoid ():
return x * (1-x)Kudzidzisa modhi ine huremu hwekutanga uye kusarerekera:
epoch=10000
lr=0.1
inputlayer_neurons = X.shape[1]
hiddenlayer_neurons = 3
output_neurons = 1
wh=np.random.uniform(size=(inputlayer_neurons,hiddenlayer_neurons))
bh=np.random.uniform(size=(1,hiddenlayer_neurons))
wout=np.random.uniform(size=(hiddenlayer_neurons,output_neurons))
bout=np.random.uniform(size=(1,output_neurons))Kune vanotanga, kana iwe uchida rubatsiro maererano neural network, unogona kutsvaga pawebhusaiti emakambani epamusoro ekugadzira software kana unogona kuhaya vanogadzira AI/ML kuti vashande pachirongwa chako.
Kugadziriswa kwekodhi uchishandisa yakabuda layer neuron
hidden_layer_input1=np.dot(X,wh)
hidden_layer_input=hidden_layer_input1 + bh
hiddenlayer_activations = sigmoid(hidden_layer_input)
output_layer_input1=np.dot(hiddenlayer_activations,wout)
output_layer_input= output_layer_input1+ bout
output = sigmoid(output_layer_input)
Kuverengera kukanganisa kwekodhi yakavanzika layer
E = y-output
slope_output_layer = derivatives_sigmoid(output)
slope_hidden_layer = derivatives_sigmoid(hiddenlayer_activations)
d_output = E * slope_output_layer
Error_at_hidden_layer = d_output.dot(wout.T)
d_hiddenlayer = Error_at_hidden_layer * slope_hidden_layer
wout += hiddenlayer_activations.T.dot(d_output) *lr
bout += np.sum(d_output, axis=0,keepdims=True) *lr
wh += X.T.dot(d_hiddenlayer) *lr
bh += np.sum(d_hiddenlayer, axis=0,keepdims=True) *lrBuda
print (output)
[[0.03391414]
[0.97065091]
[0.9895072 ]]Izvo zvinogara zvichikosha kuti urambe uchizivana neazvino mitauro yekuronga uye nzira dzekukodha, uye vanogadzira mapurogiramu vanofanirwawo kuziva maturusi mazhinji anobatsira kuti maapplication avo aenderane nevashandisi vatsva.
Muna 2020, vanogadzira software vanofanirwa kufunga nezvekubatanidza aya mashanu maturusi ekuvandudza software muzvigadzirwa zvavo, zvisinei kuti vanoshandisa mitauro yei:
1. Natural Language Processing (NLP)
Iine chatbot inokwenenzvera sevhisi yevatengi, NLP iri kuwana tarisiro yevagadziri vepurogiramu vari kushanda pakuvandudza software yemazuva ano. Ivo vanoshandisa NLTK maturusi emidziyo akadai sePython NLTK kukurumidza kubatanidza NLP mumachatbots, vabatsiri vedhijitari, uye zvigadzirwa zvedhijitari. Pakazosvika pakati pa2020 kana munguva pfupi iri kutevera, iwe uchaona NLP ichiwedzera kukosha mune zvese kubva kumabhizinesi ezvitoro kusvika kumotokari dzinozvimiririra uye zvishandiso zvemba nehofisi.
Kufambira mberi nezvishandiso zviri nani zvekuvandudza software uye matekinoroji, unogona kutarisira kuti vanogadzira software vashandise NLP munzira dzakasiyana siyana, kubva kunzwi-based user interfaces kusvika kune nyore kutenderera menyu, kuongorora manzwiro, kuzivikanwa kwemamiriro ezvinhu, manzwiro, uye kuwanikwa kwedata. Zvese izvi zvichave zviripo kune ruzhinji rwevashandisi, uye makambani achakwanisa kuwana kukura kwechigadzirwa kusvika kumadhora mazana mana nemakumi matatu emadhora ne430 (maererano neIDC, yakataurwa naDeloitte).
2. GraphQL ichitsiva REST Apis
Sekureva kwevagadziri vekambani yangu, inova kambani yekuvandudza software yekunze, iyo REST API iri kurasikirwa nekutonga kwayo pamusoro pesesese yekushandisa nekuda kwekunonoka kurodha data inoda kuitwa kubva kune akawanda maURL ega.
GraphQL inzira itsva uye imwe nzira iri nani kune REST-based architecture inotora data rese rakakodzera kubva kune akawanda saiti uchishandisa mubvunzo mumwe chete. Izvi zvinovandudza kupindirana kwevatengi-server uye zvinoderedza latency, zvichiita kuti chikumbiro chiite zvakanyanya kuterera kumushandisi.
Unogona kuvandudza hunyanzvi hwako hwekuvandudza software paunoshandisa GraphQL yekuvandudza software. Pamusoro pezvo, inoda kodhi shoma pane REST Api uye inobvumidza iwe kuita mibvunzo yakaoma mumitsetse mishoma iri nyore. Inogona zvakare kuve yakashongedzerwa nehuwandu hweBackand seSevhisi (BaaS) maficha anoita kuti zvive nyore kushandiswa nevagadziri vesoftware mumitauro yakasiyana-siyana yehurongwa, kusanganisira Python, Node.js, C++ uye Java.
3. Low coding level/hapana kodhi (yakaderera kodhi)
Yese yakaderera kodhi software yekuvandudza maturusi anopa akawanda mabhenefiti. Inofanira kunge ichishanda sezvinobvira pakunyora mapurogiramu akawanda kubva pakutanga. Yakaderera kodhi inopa preconfigured kodhi inogona kuisirwa muzvirongwa zvakakura. Izvi zvinobvumira kunyangwe vasiri-programmer kuti nekukurumidza uye nyore kugadzira zvigadzirwa zvakaomarara uye nekumhanyisa iyo yazvino budiriro ecosystem.
Sekureva kwemushumo weTechRepublic, hapana-kodhi / yakaderera kodhi maturusi ari kutoshandiswa mumawebhu portal, software masisitimu, nharembozha nedzimwe nzvimbo. Iyo yakaderera kodhi maturusi musika inokura kusvika pamadhora gumi nemashanu emadhora panosvika 15. Zvishandiso izvi zvinobata zvese, zvinosanganisira kutonga kufambiswa kwemabasa, kusefa data, kupinza uye kutumira kunze. Heano akanakisa epasi kodhi mapuratifomu muna 2020:
- Microsoft PowerApps
- Mendix
- Outsystems
- Zoho Musiki
- Salesforce App Cloud
- Quick Base
- Bhoti yechirimo
4. 5G wave
5G yekubatanidza ichakanganisa zvakanyanya nharembozha uye kusimudzira software pamwe nekuvandudzwa kwewebhu. Mushure mezvose, nematekinoroji akadai seIoT, zvese zvakabatana. Nekudaro, iyo software yemuchina ichaita zvakanyanya kugona kweakakwira-speed wireless network ne5G.
Mubvunzurudzo ichangoburwa neDigital Trends, Dan Dery, mutevedzeri wemutungamiri we Motorola, akati "mumakore anotevera, 5G ichaunza data nekukurumidza, yakakwirira bandwidth, uye nekumhanyisa software yefoni kanokwana kagumi kupfuura tekinoroji iripo isina waya."
Muchiedza ichi, makambani esoftware achashanda kuunza 5G mumashandisirwo emazuva ano. Parizvino, vanopfuura makumi maviri vanoshanda vakazivisa kukwidziridzwa kunetiweki yavo. Saka, vanogadzira zvino vachatanga kushanda pakushandisa maAPI akakodzera kutora mukana we20G. Iyo tekinoroji ichavandudza zvakanyanya zvinotevera:
- Network kuchengetedza chirongwa, kunyanya kune Network Slicing.
- Ipa nzira itsva dzekushandisa maID ID.
- Inokutendera kuti uwedzere mashandiro matsva kune maapplication ane low latency.
- Ichapesvedzera kuvandudzwa kweiyo AR/VR system.
5. Easy authentication
Kusimbisa kuri kuramba kuchive nzira inoshanda yekudzivirira data inonzwisisika. Iyo tekinoroji yakaomesesa haisi panjodzi chete kune software hacks, asi zvakare inotsigira yekunyepedzera kungwara uye kunyange quantum komputa. Asi iyo software yekuvandudza musika yave kutoona akawanda marudzi matsva ehuchokwadi, akadai sekuongorora inzwi, biometrics uye kuzivikanwa kwechiso.
Panguva ino, matsotsi anowana nzira dzakasiyana dzekugadzira online mushandisi ID nemapassword. Sezvo vashandisi venhare vagara vajaira kuwana mafoni avo nemunwe kana chiso chekutarisa, nekudaro vachishandisa maturusi echokwadi, ivo havazodi hutsva hwekusimbisa masimba sezvo mukana wekubirwa kwecyber uchange uri mushoma. Heano mamwe akawanda-factor echokwadi maturusi ane SSL encryption.
- Soft Tokens shandura mafoni ako kuti ave ari nyore ma-multi-factor authenticators.
- Egrid templates iri nyore kushandisa uye yakakurumbira fomu yevatenderi muindasitiri.
- Zvimwe zvezvakanakisa zvirongwa zvekusimbisa mabhizinesi ndezveRSA SecurID Access, OAuth, Ping Identity, Authx, uye Aerobase.
Kune makambani esoftware muIndia neUS ari kuita tsvakiridzo yakakura mumunda wekusimbisa uye biometric. Ivo zvakare vari kusimudzira AI kugadzira yepamusoro software yezwi, face-id, maitiro uye biometric authentication. Iye zvino unogona kuchengetedza dhijitari chiteshi uye kuvandudza hunyanzvi hwepuratifomu.
mhedziso
Zvinotaridza senge hupenyu hwevagadziri huchave husina kunetsa muna 2020 sezvo nhanho yekuvandudza software ingangowedzera. Zvishandiso zviripo zvichava nyore kushandisa. Pakupedzisira, kufambira mberi uku kuchagadzira nyika ine simba ichipinda muzera idzva redhijitari.
Source: www.habr.com