| | import gradio as gr |
| | import numpy as np |
| | from pykrige.ok import OrdinaryKriging |
| |
|
| |
|
| | def fonction_api(lat, lon, debit_min, date): |
| | |
| | echelle = 10 |
| | x = np.array(x) |
| | y = np.array(y) |
| | z = np.array(z) |
| | grid_x_min = x.min() |
| | grid_x_max = x.max() |
| | grid_y_min = y.min() |
| | grid_y_max = y.max() |
| | delta_x, delta_y = (grid_x_max - grid_x_min) / \ |
| | echelle, (grid_y_max - grid_y_min)/echelle |
| | grid_x = np.linspace(grid_x_min, grid_x_max, echelle, dtype="float64") |
| | grid_y = np.linspace(grid_y_min, grid_y_max, echelle, dtype="float64") |
| | OK = OrdinaryKriging(x, y, z, variogram_model="exponential",) |
| | zstar, ss = OK.execute("grid", grid_x, grid_y) |
| | mask = zstar > debit_min |
| | lat_pred = [] |
| | lon_pred = [] |
| | debit_pred = [] |
| | for i in range(echelle): |
| | for j in range(echelle): |
| | if mask[i, j] == True: |
| | lat_pred.append( |
| | ((grid_x_min + (i-1)*delta_x)+(grid_x_min + i*delta_x))/2) |
| | lon_pred.append( |
| | ((grid_y_min + (j-1)*delta_y)+(grid_y_min + j*delta_y))/2) |
| | debit_pred.append(zstar[i, j]) |
| | data = {} |
| | for i in range(len(lat_pred)): |
| | data["point " + str(i)] = [lat_pred[i], lon_pred[i], debit_pred[i]] |
| | return data |
| |
|
| |
|
| | demo = gr.Interface(fn=fonction_api, inputs=[ |
| | "number", "number", "number", "text"], outputs="json") |
| | demo.launch() |
| |
|
