KET Studio’ni noldan o‘rganing.

Bu dastur nimaga xizmat qiladi?

KET Studio Python tajribasini lokal Windows desktop’da yozish, ishga tushirish, terminal oqimini ko‘rish va natijani tushunarli vizual event’lar bilan tekshirish uchun mo‘ljallangan. Quyidagi oqim dasturning asosiy modelidir:

15 daqiqada birinchi natija.

Birinchi setup internetga va kompyuterda Python’ga muhtoj. KET Studio core quantum paketlarni o‘zi yaratgan izolyatsiyalangan muhitga o‘rnatishga urinadi; bu jarayon birinchi marta bir necha daqiqa olishi mumkin.

1. Downloads sahifasidan Windows x64 MSIX’ni oling. Self-signed paket bo‘lsa, certificate’ni Trusted People store’ga o‘rnating; trust muammosida EXE fallback’dan foydalaning.
2. KET Studio’ni oching va Settings → Environment’da Python interpreter’ni tekshiring. `python`, `python3`, `py` yoki aniq `python.exe` manzilidan foydalanish mumkin.
3. Rebuild environment tugmasini bosing. Dastur `ket_venv` yaratadi, `qiskit[visualization]`, `qiskit-aer` va `numpy` core paketlarini o‘rnatadi hamda verification circuit ishga tushiradi.
4. Yangi project papkasini oching, `.py` fayl yarating va Editor’da kod yozing.
5. Run yoki F5 ni bosing. stdout/stderr Terminal’da, `KET_VIZ` event’lari esa Visualization, Metrics, Inspector, Estimator yoki History panellarida ko‘rinadi.

Papka va interpreter qanday ishlaydi?

Script joylashgan papka project root bo‘ladi. Internal Python environment esa loyiha ichida emas, application support ichidagi `ket_venv`da saqlanadi. Bu loyiha papkasini toza qoldiradi, lekin archive uchun environment ma’lumotini alohida yozib borish kerak.

my-project/
├── experiment.py       # your source
├── requirements-lock.txt
└── .ket/
    ├── out/             # PNG/SVG and other artifacts
    └── temp/            # launcher and temporary runtime files

Qiskit’dan tashqari paket kerak bo‘lsa Settings → Environment → package action orqali o‘rnating. Optional paketlar avtomatik o‘rnatilmaydi; package o‘rnatilishi internet, disk joyi va Python versiyasiga bog‘liq.

Ishlaydigan to‘liq Bell misoli.

Quyidagi template Qiskit circuit’ni yaratadi, AerSimulator’da 1024 shot bajaradi, histogram va metadata yuboradi, so‘ng circuit rasmini `.ket/out`ga saqlaydi. Bu docs oqimini tekshirish uchun eng yaxshi birinchi reproducer.

from qiskit import QuantumCircuit
from qiskit_aer import AerSimulator
import ket_viz

shots = 1024
seed = 7
qc = QuantumCircuit(2, 2)
qc.h(0)
qc.cx(0, 1)
qc.measure([0, 1], [0, 1])

backend = AerSimulator(seed_simulator=seed)
result = backend.run(qc, shots=shots).result()
counts = result.get_counts()

ket_viz.metrics({
    'backend': 'AerSimulator',
    'qubits': qc.num_qubits,
    'depth': qc.depth(),
    'shots': shots,
    'seed': seed,
    'status': 'completed',
})
ket_viz.histogram(dict(counts), title='Bell-state measurement')

qc.draw(output='mpl', filename='.ket/out/bell-circuit.png')
ket_viz.circuit('.ket/out/bell-circuit.png', title='Bell circuit')

Kutiladigan natija: Terminal’da process chiqishi; Visualization’da metrics va histogram; artifact ko‘rinishida esa Bell circuit PNG. `counts` qiymatlari seed va simulator versiyasiga qarab ozgina farq qilishi mumkin.

Natijani qaysi panelga qanday uzatish kerak?

Quyidagi API’lar Run/F5 jarayonida to‘g‘ridan-to‘g‘ri ishlaydi. Payload JSON-safe bo‘lsin: string, number, boolean, list yoki object. Scientific object’ni avval oddiy Python qiymatlariga aylantiring.

import ket_viz

counts = {'00': 510, '01': 2, '10': 4, '11': 508}
ket_viz.histogram(counts, title='Measurement distribution')

import ket_viz

matrix = [[1.0, 0.2, 0.0],
          [0.2, 0.8, 0.1],
          [0.0, 0.1, 0.4]]
ket_viz.heatmap(matrix, title='Qubit correlation')

import ket_viz

energies = [0.50, 0.31, 0.18, 0.09, 0.04]
ket_viz.chart(energies, title='VQE convergence')

import ket_viz

rows = [['Metric', 'Value'],
        ['Qubits', 4],
        ['Shots', 1024],
        ['Fidelity', 0.998]]
ket_viz.table('Experiment summary', rows)

import ket_viz

amplitudes = [
  {'label': '00', 'mag': 0.707, 'phase': 0.0},
  {'label': '11', 'mag': 0.707, 'phase': 3.14159},
]
ket_viz.statevector(amplitudes, title='Bell amplitudes')

import ket_viz

frames = [
  {'gate': 'H(0)', 'state_description': 'Superposition',
   'bloch': [{'theta': 1.57, 'phi': 0.0}]},
]
ket_viz.inspector('Algorithm steps', frames)
ket_viz.bloch({'theta': 1.57, 'phi': 0.0})

import ket_viz

ket_viz.metrics({
  'status': 'optimizing',
  'step': 12, 'energy': -1.1372,
  'seed': 20260914, 'backend': 'AerSimulator'
})

import ket_viz

ket_viz.text('Simulation started')
# Real exception bo‘lsa uni yashirmang:
# raise RuntimeError('backend failed')

import ket_viz

ket_viz.estimator({
  'qubits': 8, 'depth': 42,
  'total_gates': 180,
  'gate_counts': {'H': 8, 'CX': 64, 'RZ': 108}
})

Rasm, circuit va Matplotlib natijalarini ko‘rsatish.

KET Studio circuit object’ni o‘zi parse qilmaydi. Qiskit yoki boshqa library avval fayl yaratadi, keyin `ket_viz.circuit` yoki `ket_viz.image` shu faylni panelga olib kiradi. Relative path project root’dan hisoblanadi.

from qiskit import QuantumCircuit
import matplotlib.pyplot as plt
import ket_viz

qc = QuantumCircuit(2)
qc.h(0)
qc.cx(0, 1)
qc.draw(output='mpl', filename='.ket/out/circuit.png')
ket_viz.circuit('.ket/out/circuit.png', title='Rendered circuit')

plt.plot([0, 1, 2, 3], [0.8, 0.4, 0.2, 0.1])
plt.title('Loss')
plt.savefig('.ket/out/loss.png', bbox_inches='tight')
ket_viz.image('.ket/out/loss.png', title='Loss curve')

• Event yuborishdan oldin fayl yozilib bo‘lgan bo‘lsin.
• Artifact publication yoki review uchun muhim bo‘lsa `.ket/out`ni project bilan birga commit yoki archive qiling.
• Katta raw figure’ni diskka saqlang, UI’ga esa kerakli preview’ni yuboring.

Har qanday Python yoki boshqa process bilan.

`ket_viz` ishlamaydigan adapter, subprocess yoki boshqa til uchun stdout’ga bitta UTF-8 qator qilib `KET_VIZ ` + JSON chiqaring. JSON boshqa log bilan bir qatorga aralashmasin.

import json
import time

event = {
  'kind': 'histogram',
  'payload': {
    'histogram': {'0': 490, '1': 534},
    'title': 'Raw protocol'
  },
  'ts': int(time.time() * 1000)
}
print('KET_VIZ ' + json.dumps(event, ensure_ascii=False), flush=True)

• `KET_VIZ ` prefiksi, JSON va newline bitta to‘liq qatorda bo‘lsin.
• `kind`ga mos `payload` yuboring; real-time oqim uchun `flush=True` ishlating.
• Rasm path event’idan oldin fayl haqiqatda yozilgan bo‘lsin.
• Noma’lum `kind` rendererga ega emas va UI’da ko‘rsatilmaydi; malformed yoki oversize event warning bilan tashlanadi.

Qancha data foydali bo‘lib qoladi?

Limitlar ilmiy hisobni to‘xtatish uchun emas, UI’ni katta oqimda muzlab qolishidan himoya qilish uchun. Katta result’ni JSON/CSV yoki image sifatida diskka saqlang, panelga sparse yoki agregatsiyalangan preview yuboring.

VS Code’ga yaqin local shell.

Windows desktop’dagi Terminal paneli xterm3 UI va ConPTY native host bilan ishlaydi. Shell komandalarini yozing, ANSI ranglarni ko‘ring, resize qiling, input bering yoki Ctrl+C bilan process’ni to‘xtating.

python --version
where python
python -c "print('stdout OK')"
python -m pip freeze

• Terminal’dagi `python` va Settings’da tanlangan interpreter har doim bir xil bo‘lmasligi mumkin; `where python` bilan tekshiring.
• Run action script process’ini alohida boshqaradi; Terminal qo‘lda berilgan komandalar uchun qoladi.
• Web preview shell ochmaydi — bu local process va fayl tizimini brauzerdan ajratadigan xavfsizlik chegarasi.
• Process tugagach event’lar bounded session/history’da qoladi; doimiy research archive’ni project fayllari bilan o‘zingiz saqlang.

Natijani qayta tiklanadigan qiling.

KET Studio visualization’ni ko‘rsatadi, lekin ilmiy xulosani o‘zi tasdiqlamaydi. Har bir muhim run bilan source, muhit, parametr, raw result va artifact’ni birga saqlang.

research-run/
├── experiment.py
├── requirements-lock.txt
├── metadata.json          # backend, shots, seed, versions
├── raw/                    # counts, arrays, CSV/JSON
├── logs/                   # stdout and stderr
└── .ket/out/               # images and rendered circuits

1. KET Studio version, Git commit va script SHA-256’ni yozing.
2. OS, Python version, backend/library version va dependency lockfile’ni saqlang.
3. Qubit, depth, gate count, shots, optimizer va seed’ni metadata’ga qo‘shing.
4. Raw counts, logs, event JSON va `.ket/out` artifact’larini arxivlang.
5. Rerun qilib, exit code, warning/error va natija farqini yozib qo‘ying.

Ishlamasa qayerdan boshlash kerak?

Muammo report qilganda KET Studio version’i, OS, Python path/version, minimal reproducer, terminal log’i va event JSON’ini birga yuboring.

Yangi foydalanuvchi bilishi kerak bo‘lgan javoblar.

Repository’dagi chuqur hujjatlar