Using the Coulomb Blockade Models | Conductor Quantum

Installation

We recommend using pip, poetry, or uv to install the package.

$ pip install conductorquantum

Authentication

The SDK requires an API key for authentication. Sign in and create a new API key. Remember, your API key is your access secret—keep it safe with environment variables.

Using environment variables:

Python

1 from dotenv import load_dotenv
2 import os
3 from conductorquantum import ConductorQuantum
4 
5 # Load API key from .env file
6 load_dotenv()
7 TOKEN = os.getenv("CONDUCTOR_API_KEY")
8 
9 # Initialize client
10 client = ConductorQuantum(token=TOKEN)

Or provide the API key directly:

Python

1 from conductorquantum import ConductorQuantum
2 
3 # Initialize client with API key
4 client = ConductorQuantum(token="YOUR_API_KEY")

Usage Examples

Using the Coulomb Blockade Classifier

The Coulomb Blockade Classifier is a model that can classify a given current measurement as either in the Coulomb Blockade regime or not.

You can download an example file to follow along with the example:

Python

1 from dotenv import load_dotenv
2 import os
3 import numpy as np
4 from conductorquantum import ConductorQuantum
5 
6 
7 # Load API key from .env file
8 load_dotenv()
9 TOKEN = os.getenv("CONDUCTOR_API_KEY")
10 
11 # Initialize client
12 client = ConductorQuantum(token=TOKEN)
13 
14 # Load Coulomb blockade data (current measurement)
15 data = np.load("coulomb-blockade-classifier-v0.npy")  # shape (n, )
16 
17 # Detect Coulomb blockade peaks
18 result = client.models.execute(
19     model="coulomb-blockade-classifier-v0",
20     data=data
21 )
22 
23 # Access the classification result
24 is_coulomb_blockade = result.output["classification"]
25 print(f"Is Coulomb blockade: {is_coulomb_blockade}")

Output

1 Is Coulomb blockade: True

Using the Coulomb Blockade Peak Detector

The Coulomb Blockade Peak Detector is a model that can detect the peaks in a given current measurement.

You can download an example file to follow along with the example:

Python

1 from dotenv import load_dotenv
2 import os
3 import numpy as np
4 from conductorquantum import ConductorQuantum
5 
6 
7 # Load API key from .env file
8 load_dotenv()
9 TOKEN = os.getenv("CONDUCTOR_API_KEY")
10 
11 # Initialize client
12 client = ConductorQuantum(token=TOKEN)
13 
14 # Load Coulomb blockade data (current measurement)
15 data = np.load("coulomb-blockade-peak-detector-v0.npy")  # shape (n, )
16 
17 # Detect Coulomb blockade peaks
18 result = client.models.execute(
19     model="coulomb-blockade-peak-detector-v0",
20     data=data
21 )
22 
23 # Access the peak locations (indices)
24 peak_indices = result.output["peak_indices"]
25 print(f"Detected peaks at: {peak_indices}")

Output

1 Detected peaks at: [30, 38, 47, 55, 64, 72, 81, 89, 98]

Plotting the Output

Python

1 import matplotlib.pyplot as plt
2 
3 plt.figure(figsize=(10, 5), dpi=300)
4 plt.plot(example_input, 'k-')
5 count = 0
6 for peak_index in result.output["peak_indices"]:
7     if count == 0:
8         plt.axvline(x=peak_index, color='r', linestyle='--', label='Peaks')
9     elif count > 0:
10         plt.axvline(x=peak_index, color='r', linestyle='--')
11 
12 
13     count += 1
14 plt.xlabel('Voltages (Indices)')
15 plt.ylabel('Current (a.u.)')
16 plt.legend()
17 plt.show()

Similarly, for v1 of the Coulomb Blockade Peak Detector, you can download an example file and replace the model name in the example above with coulomb-blockade-peak-detector-v1, and you should see the following output:

Output

1 Detected peaks at: [156, 176, 199, 223, 244, 267, 289, 311, 333, 356, 378, 400, 424, 445, 468, 490]

v1 is typically more accurate than v0 for larger sweep sizes of one hundred or more points.

Data Requirements

One-dimensional Current Data

Shape: (n, )
1D array of current values

Important Notes for Voltage-Current Data

The array must be of shape (n, ) where n is the number of current values in the measurement.
Analysis is outputted in terms of indices of the input array.
The models automatically handle scaling and normalization internally.
For best results, ensure your data has sufficient resolution in regions of interest.