minor updates

app.py

@@ -1,26 +1,12 @@
+import gradio as gr
 from matplotlib import cm 
 import matplotlib.pyplot as plt
 from mpl_toolkits.axes_grid1 import make_axes_locatable
-
 import numpy as np
-
-# import onnx
 import onnxruntime as ort
-# from onnx import helper
-# from optimum.onnxruntime import ORTModel
-
-# import pandas as pd
 from PIL import Image
-
 from scipy import special
 
-# import torch
-# import torch.utils.data
-
-import gradio as gr
-# from transformers import pipeline
-
-
 # model_path = 'chlab/planet_detection_models/'
 model_path = './models/'
 
@@ -64,7 +50,6 @@ def get_activations(intermediate_model, image: list,
     
     input_name = intermediate_model.get_inputs()[0].name
     outputs = intermediate_model.run(None, {input_name: image})
-    # outputs = intermediate_model(image)
     
     output_1 = outputs[1]
     output_2 = outputs[2]
@@ -72,19 +57,9 @@ def get_activations(intermediate_model, image: list,
     output = outputs[0][0]
     output = special.softmax(output)
     
-    # origin = 'lower'
-    
-    # plt.rcParams['xtick.labelsize'] = ticks
-    # plt.rcParams['ytick.labelsize'] = ticks
-    
-    # fig, axs = plt.subplots(nrows=1, ncols=3, figsize=(28, 8))
-    
-    # ax1, ax2, ax3 = axs[0], axs[1], axs[2]
-    
     in_image = np.sum(image[0, :, :, :], axis=0)
     in_image = normalize_array(in_image)
-    
-    # im1 = ax1.imshow(in_image, cmap=cmap, vmin=0, vmax=vmax, origin=origin)
+
     if layer is None:
         activation_1 = np.sum(output_1[0, :, :, :], axis=0)
         activation_2 = np.sum(output_2[0, :, :, :], axis=0)
@@ -99,22 +74,6 @@ def get_activations(intermediate_model, image: list,
         activation_2 -= np.mean(activation_2)
         activation_2 = np.abs(activation_2)
     
-    
-    # im2 = ax2.imshow(activation_1, cmap=cmap, #vmin=0, vmax=1, 
-    #                  origin=origin)
-    # im3 = ax3.imshow(activation_2, cmap=cmap, #vmin=0, vmax=1, 
-    #                  origin=origin) 
-    # ims = [im1, im2, im3]
-        
-    # for (i, ax) in enumerate(axs):
-    #     divider = make_axes_locatable(ax)
-    #     cax = divider.append_axes('right', size='5%', pad=0.05)
-    #     fig.colorbar(ims[i], cax=cax, orientation='vertical')
-        
-    # ax1.set_title('Input', fontsize=titles)
-    
-    # plt.show()
-    
     return output, in_image, activation_1, activation_2
 
 
@@ -128,22 +87,10 @@ def predict_and_analyze(model_name, num_channels, dim, image):
     
     num_channels = int(num_channels)
     W = int(dim)
-    
-    # image = image.read()
-    
-    # with open(image, 'rb') as f:
-    #     im = f.readlines()
-    # image = np.frombuffer(image)
-    
+
     print("Loading data")
     image = np.load(image.name, allow_pickle=True)
     
-    # image = image.reshape((num_channels, W, W))
-    
-    # W = int(np.sqrt(image.shape[1]))
-    
-    # image = image.reshape((num_channels, W, W))
-    
     if len(image.shape) != 4:
         image = image[np.newaxis, :, :, :]
         
@@ -182,9 +129,9 @@ def predict_and_analyze(model_name, num_channels, dim, image):
     
     ax1, ax2 = axs[0], axs[1]
     
-    im1 = ax1.imshow(activation_1, cmap=cmap, #vmin=0, vmax=1, 
+    im1 = ax1.imshow(activation_1, cmap=cmap,
                      origin=origin)
-    im2 = ax2.imshow(activation_2, cmap=cmap, #vmin=0, vmax=1, 
+    im2 = ax2.imshow(activation_2, cmap=cmap, 
                      origin=origin) 
     
     ims = [im1, im2]
@@ -224,7 +171,6 @@ if __name__ == "__main__":
                 #  gr.Image(label="Activation 1", show_label=True), 
                 #  gr.Image(label="Actication 2", show_label=True)],
                 gr.Plot(label="Activations", show_label=True) 
-                #  gr.Plot(label="Actication 2", show_label=True)],
                 ],
         title="Kinematic Planet Detector"
     )