FCN/Setup.py at master · wangying-ict/FCN · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
import Inference
import numpy as np
from utils.utils import image_process, image_comparison
import argparse
import tensorflow as tf


class SplitDeconvolution:
    def __init__(self, network_name, inference_mode):
        self.current_networks = network_name
        self.inference_mode = inference_mode

    def run_sim(self):
        network_parameters_file = open("./networks_configuration/" + self.current_networks + ".csv", 'r')
        first_row_name = True
        latest_result = np.load("./raw_data/" + self.current_networks + "/input/Input_Data.npy")

        for row in network_parameters_file:
            if first_row_name:
                first_row_name = False
                continue

            layer_parameters = row.strip().split(',')
            if len(layer_parameters) < 10:
                continue

            name = layer_parameters[0]
            print("Running for " + name + "\n")

            weight_file_path = "./raw_data/" + self.current_networks + "/weight/" + name + "_weight.npy"
            bias_file_path = "./raw_data/" + self.current_networks + "/bias/" + name + "_bias.npy"

            # network parameters
            stride = int(layer_parameters[1])
            output_x_length = int(layer_parameters[2])
            output_y_length = int(layer_parameters[3])
            input_pad = [int(layer_parameters[4]), int(layer_parameters[5]),
                         int(layer_parameters[6]), int(layer_parameters[7])]
            op_mode = layer_parameters[8]
            act_mode = layer_parameters[9]
            # out = int(layer_parameters[10])

            running = Inference.Computation(stride, output_x_length, output_y_length, input_pad, op_mode, act_mode,
                                            latest_result, weight_file_path, bias_file_path)
            # split deconvolution
            if self.inference_mode == "split_deconv":
                latest_result = running.operation_executive()
                # save_f = latest_result.reshape((out, output_x_length * output_y_length))
                # np.savetxt(name + ".txt", save_f, '%.7f')
            elif self.inference_mode == "tf_deconv":
                latest_result = running.tf_executive()

        if self.inference_mode == "tf_deconv":
            with tf.Session() as sess:
                sess.run(tf.global_variables_initializer())
                latest_result = sess.run(latest_result)
                latest_result = latest_result[0, :, :, :]
        image_process(latest_result, "image_" + self.inference_mode)

        print("Inference Done")


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        '--model', help='Network Name', required=True)
    parser.add_argument(
        '--mode', help='Select Inference Mode [tf_deconv, split_deconv, verify]', required=True)
    args = parser.parse_args()
    SD = SplitDeconvolution(args.model, args.mode)

    if args.mode == "verify":
        image_comparison()
    else:
        SD.run_sim()


if __name__ == '__main__':
    main()