Hello, Habr! Ina gabatar muku da fassarar labarin ta Kristencarter7519.
Ko da yake da alama sauran kwanaki kadan ne ya rage a shekarar 2020, amma wadannan ranaku ma suna da muhimmanci a fagen bunkasa manhaja. Anan a cikin wannan labarin, za mu ga yadda shekara mai zuwa 2020 za ta canza rayuwar masu haɓaka software.
Makomar ci gaban software yana nan!
Haɓaka software na gargajiya shine haɓaka software ta hanyar rubuta lambar bin wasu ƙayyadaddun ƙa'idodi. Amma ci gaban software na zamani ya ga canji mai ma'ana tare da ci gaba a cikin basirar wucin gadi, koyon injina da zurfafa ilmantarwa. Ta hanyar haɗa waɗannan fasahohin guda uku, masu haɓakawa za su iya ƙirƙirar mafita na software waɗanda ke koya daga umarnin kuma ƙara ƙarin fasali da alamu zuwa bayanan da ake buƙata don samar da sakamakon da ake so.
Bari mu gwada da wani lamba
A tsawon lokaci, tsarin ci gaban software na cibiyar sadarwa na jijiyoyi sun zama mafi rikitarwa dangane da haɗin kai da kuma matakan aiki da musaya. Masu haɓakawa, alal misali, na iya gina hanyar sadarwa mai sauƙi mai sauƙi tare da Python 3.6. Anan akwai shirin misali wanda ke yin rabe-raben binary tare da 1s ko 0s.
Tabbas, zamu iya farawa ta hanyar ƙirƙirar ajin cibiyar sadarwa na jijiyoyi:
shigo da NumPy azaman NP
X=np.array([[0,1,1,0],[0,1,1,1],[1,0,0,1]])
y=np.array([[0],[1],[1]])Aikace-aikacen aikin sigmoid:
def sigmoid ():
return 1/(1 + np.exp(-x))
def derivatives_sigmoid ():
return x * (1-x)Horar da samfuri tare da ma'aunin farko da son zuciya:
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))Don masu farawa, idan kuna buƙatar taimako game da hanyoyin sadarwar jijiya, zaku iya bincika intanit don gidajen yanar gizon manyan kamfanonin haɓaka software ko kuna iya hayar masu haɓaka AI/ML don yin aiki akan aikin ku.
Gyaran lamba ta amfani da abin fitarwa na 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)
Kuskuren ƙididdigewa don ɓoyayyun lambar lambar
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) *lrFita
print (output)
[[0.03391414]
[0.97065091]
[0.9895072 ]]Yana da kyau koyaushe a ci gaba da sabunta sabbin yarukan shirye-shirye da dabarun yin coding, kuma ya kamata masu shirye-shirye su san yawancin sabbin kayan aikin da ke taimakawa apps ɗin su dace da sabbin masu amfani.
A cikin 2020, masu haɓaka software yakamata suyi la'akari da haɗa waɗannan kayan aikin haɓaka software guda 5 cikin samfuran su, komai yaren shirye-shirye da suke amfani da shi:
1. Tsarin Harshen Halitta (NLP)
Tare da chatbot wanda ke daidaita sabis na abokin ciniki, NLP yana samun hankalin masu shirye-shirye da ke aiki akan haɓaka software na zamani. Suna amfani da kayan aikin NLTK irin su Python NLTK don haɗa NLP cikin sauri a cikin chatbots, mataimakan dijital, da samfuran dijital. A tsakiyar 2020 ko nan gaba kadan, zaku ga NLP ta zama mafi mahimmanci a cikin komai daga kasuwancin dillalai zuwa motoci da na'urori masu zaman kansu don gida da ofis.
Ci gaba tare da ingantattun kayan aikin haɓaka software da fasaha, kuna iya tsammanin masu haɓaka software za su yi amfani da NLP ta hanyoyi daban-daban, daga mu'amalar mai amfani da murya zuwa mafi sauƙin kewaya menu, nazarin ji, gano mahallin, motsin rai, da samun damar bayanai. Duk waɗannan za su kasance ga mafi yawan masu amfani, kuma kamfanoni za su iya samun ci gaban yawan aiki har zuwa dala biliyan 430 nan da 2020 (bisa ga IDC, wanda Deloitte ya ambata).
2. GraphQL yana maye gurbin REST Apis
A cewar masu haɓakawa a kamfani na, wanda kamfani ne na haɓaka software na ketare, REST API yana rasa rinjayensa akan sararin samaniya saboda jinkirin ɗaukar bayanan da ake buƙatar yin daga URLs da yawa daban-daban.
GraphQL sabon salo ne kuma mafi kyawun madadin gine-gine na tushen REST wanda ke dawo da duk bayanan da suka dace daga shafuka da yawa ta amfani da tambaya guda ɗaya. Wannan yana inganta hulɗar abokin ciniki da uwar garken kuma yana rage jinkiri, yana sa aikace-aikacen ya fi dacewa da mai amfani sosai.
Kuna iya haɓaka ƙwarewar haɓaka software lokacin da kuke amfani da GraphQL don haɓaka software. Bugu da ƙari, yana buƙatar ƙasa da lamba fiye da REST Api kuma yana ba ku damar yin tambayoyi masu rikitarwa a cikin ƴan layukan masu sauƙi. Hakanan ana iya sanye shi da wasu fasalulluka na Backand a matsayin Sabis (BaaS) waɗanda ke sauƙaƙa amfani da masu haɓaka software a cikin harsunan shirye-shirye daban-daban, gami da Python, Node.js, C++ da Java.
3. Low codeing/no code (low code)
Duk ƙananan kayan aikin haɓaka software suna ba da fa'idodi da yawa. Ya kamata ya zama mai inganci sosai yayin rubuta shirye-shirye da yawa daga karce. Ƙananan lambar tana ba da lambar da aka riga aka tsara wacce za a iya shigar da ita cikin manyan shirye-shirye. Wannan yana ba da damar ko da waɗanda ba su da shirye-shirye don ƙirƙirar samfuran hadaddun da sauri da sauƙi da haɓaka yanayin yanayin ci gaban zamani.
A cewar wani rahoton TechRepublic, an riga an yi amfani da kayan aikin no-code/low code a tashoshin yanar gizo, tsarin software, aikace-aikacen hannu da sauran wurare. Kasuwancin kayan aikin ƙananan code zai girma zuwa dala biliyan 15 nan da 2020. Waɗannan kayan aikin suna ɗaukar komai, gami da sarrafa dabaru na tafiyar aiki, tace bayanai, shigo da fitarwa. Anan ne mafi kyawun dandamali na ƙananan code a cikin 2020:
- Microsoft PowerApps
- Mendix
- Kayan aiki na waje
- Zoho Mahalicci
- Salesforce App Cloud
- Saurin tushe
- Takalmin bazara
4. 5G ruwa
Haɗin 5G zai yi tasiri sosai ga aikace-aikacen hannu da haɓaka software da haɓakar yanar gizo. Bayan haka, tare da fasaha irin su IoT, an haɗa komai. Don haka, software na na'urar za ta yi amfani da mafi kyawun damar hanyoyin sadarwar mara waya mai sauri tare da 5G.
A cikin wata hira da aka yi da Digital Trends, Dan Dery, mataimakin shugaban kamfanin Motorola, ya ce "a cikin shekaru masu zuwa, 5G zai isar da bayanai da sauri, mafi girman bandwidth, da kuma hanzarta software na waya sau 10 fiye da fasahar mara waya ta data kasance."
Ta wannan haske, kamfanonin software za su yi aiki don kawo 5G cikin aikace-aikacen zamani. A halin yanzu, fiye da masu aiki 20 sun sanar da haɓaka hanyoyin sadarwar su. Don haka, masu haɓakawa yanzu za su fara aiki akan amfani da APIs masu dacewa don cin gajiyar 5G. Fasahar za ta inganta abubuwa masu zuwa sosai:
- Tsaron shirin hanyar sadarwa, musamman don Yanke hanyar sadarwa.
- Samar da sabbin hanyoyin sarrafa ID na mai amfani.
- Yana ba ku damar ƙara sabbin ayyuka zuwa aikace-aikace tare da ƙarancin jinkiri.
- Zai tasiri ci gaban tsarin AR/VR.
5. Sauƙaƙe ingantaccen tabbaci
Tabbatarwa yana ƙara zama ingantaccen tsari don kare mahimman bayanai. Ƙwaƙwalwar fasahar ba wai kawai tana da rauni ga hacking ɗin software ba, har ma tana goyan bayan bayanan ɗan adam har ma da ƙididdigar ƙididdiga. Amma kasuwar haɓaka software ta riga tana ganin sabbin nau'ikan tantancewa da yawa, kamar nazarin murya, ƙirar halitta da sanin fuska.
A wannan mataki, masu kutse suna samun hanyoyi daban-daban don ƙirƙira ID na mai amfani da kan layi da kalmomin shiga. Tunda masu amfani da wayar hannu sun riga sun saba shiga wayoyinsu da hoton yatsa ko na fuska, don haka amfani da kayan aikin tantancewa, ba za su bukaci sabbin hanyoyin tantancewa ba saboda yiwuwar satar yanar gizo zai ragu. Anan akwai wasu kayan aikin tantance abubuwa da yawa tare da ɓoyewar SSL.
- Soft Tokens suna juyar da wayowin komai da ruwan ku zuwa ingantattun abubuwa masu yawa.
- Samfuran EGrid abu ne mai sauƙin amfani kuma sanannen nau'i na masu tabbatarwa a cikin masana'antar.
- Wasu mafi kyawun shirye-shiryen tantancewa don kasuwanci sune RSA SecurID Access, OAuth, Ping Identity, Authx, da Aerobase.
Akwai kamfanonin software a Indiya da Amurka suna gudanar da bincike mai zurfi a fannin tantancewa da na'urorin zamani. Hakanan suna haɓaka AI don ƙirƙirar software mafi inganci don murya, fuska-id, ɗabi'a da ingantaccen yanayin halitta. Yanzu zaku iya kare tashoshi na dijital da haɓaka damar dandamali.
ƙarshe
Yana kama da rayuwa ga masu shirye-shirye za su zama ƙasa da ƙalubale a cikin 2020 saboda saurin haɓaka software zai iya haɓaka. Kayan aikin da ke akwai za su zama sauƙin amfani. A ƙarshe, wannan ci gaban zai haifar da duniyar mai ƙarfi ta shiga sabon zamani na dijital.
source: www.habr.com