Python ProtonetClassifier Examples

Example #1

File: solver-byte.py Project: SolitaryKnife/pckt_simnet

iscxvpn2016_loader = iscxvpn2016(pcap_dir="D://Datasets/ISCXVPN2016/", h5_dir="D://Datasets/packets-50k/", max_packets_on_cache=50000, as_bit=False, verbose=True)
ustctfc2016_loader = ustctfc2016(pcap_dir="D://Datasets/USTC-TFC2016/", h5_dir="D://Datasets/USTC-TFC2016-packets-50k/", max_packets_on_cache=50000, as_bit=False, verbose=True)

classifiers = {
    # "relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),
    # "reg_relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),

    # "protonet_1": M.ProtonetClassifier(in_channels=52, mid_channels=[], out_channels=32),
    # "protonet_2": M.ProtonetClassifier(in_channels=52, mid_channels=[64], out_channels=32),
    # "protonet_3": M.ProtonetClassifier(in_channels=52, mid_channels=[128, 64], out_channels=32),
    # "protonet_4": M.ProtonetClassifier(in_channels=52, mid_channels=[256, 128, 64], out_channels=32),
    # "protonet_bottleneck_end": M.ProtonetClassifier(in_channels=52, mid_channels=[256, 128, 64], out_channels=10),
    # "protonet_bottleneck_mid": M.ProtonetClassifier(in_channels=52, mid_channels=[128, 32, 128], out_channels=32),


    "protonet_mini_byte_1": M.ProtonetClassifier(in_channels=52, mid_channels=[66], out_channels=10),
    "protonet_mini_byte_2": M.ProtonetClassifier(in_channels=52, mid_channels=[26], out_channels=10),

    # "simnet_simple": M.SimnetClassifier(in_channels=52, channels=[10]),
    # "simnet_1": M.SimnetClassifier(in_channels=52, channels=[32]),
    # "simnet_2": M.SimnetClassifier(in_channels=52, channels=[64, 32]),
    # "simnet_3": M.SimnetClassifier(in_channels=52, channels=[128, 64, 32]),

    # "reg_protonet_1": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=32),
    # "reg_protonet_2": M.ProtonetClassifier(in_channels=416, mid_channels=[64], out_channels=32),
    # "reg_protonet_3": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 64], out_channels=32),
    # "reg_protonet_4": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=32),
    # "reg_protonet_bottleneck_end": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=10),
    # "reg_protonet_bottleneck_mid": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 32, 128], out_channels=32),

    # "reg_simnet_simple": M.SimnetClassifier(in_channels=416, channels=[10]),

Example #2

                           shuffle=True,
                           generator=generator)


load = dataloader.iscxvpn2016(pcap_dir="D://Datasets/ISCXVPN2016/",
                              h5_dir="D://Datasets/packets-15k/",
                              as_bit=True)
datasets = {
    "a": load("youtube"),
    "b": load("youtube"),
    "c": load("youtube"),
    "d": load("youtube")
}
confmat = plmc.ConfusionMatrix(num_classes=4)

network = model.ProtonetClassifier(416, 10)
dl = build_dataloader(datasets, n_support=50, n_queries=10)

step = 0
for queries, labels, *supports in dl:

    # print(a.size())
    # print(b.size())
    # print(len(supports))
    # for i, s in enumerate(supports):
    #     print(f"support {i}", s.size())

    logits = network(queries, *supports)

    print(confmat(logits, labels))
    step += 1

Example #3

File: solver.py Project: SolitaryKnife/pckt_simnet

    h5_dir="D://Datasets/USTC-TFC2016-packets-50k/",
    max_packets_on_cache=50000,
    as_bit=True,
    verbose=True)

classifiers = {
    # "relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),
    # "reg_relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),
    # "relation_2": M.RelationNetClassifier(in_channels=416, feature_channels=[32, 10], simnet_channels=[32, 16, 4]),
    # "reg_relation_2": M.RelationNetClassifier(in_channels=416, feature_channels=[32, 10], simnet_channels=[32, 16, 4]),
    # "relation_3": M.RelationNetClassifier(in_channels=416, feature_channels=[64, 32], simnet_channels=[64, 32, 16, 4]),
    # "reg_relation_3": M.RelationNetClassifier(in_channels=416, feature_channels=[64, 32], simnet_channels=[64, 32, 16, 4]),

    # "relpnet_1": M.RelationNetClassifier_Protonet1(simnet_channels=[128, 64, 32]),
    "protonet_mini_bit":
    M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=10),

    # "protonet_1": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=32),
    # "protonet_2": M.ProtonetClassifier(in_channels=416, mid_channels=[64], out_channels=32),
    # "protonet_3": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 64], out_channels=32),
    # "protonet_4": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=32),
    # "protonet_bottleneck_end": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=10),
    # "protonet_bottleneck_mid": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 32, 128], out_channels=32),

    # "simnet_simple": M.SimnetClassifier(in_channels=416, channels=[10]),
    # "simnet_1": M.SimnetClassifier(in_channels=416, channels=[32]),
    # "simnet_2": M.SimnetClassifier(in_channels=416, channels=[64, 32]),
    # "simnet_3": M.SimnetClassifier(in_channels=416, channels=[128, 64, 32]),

    # "reg_protonet_1": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=32),
    # "reg_protonet_2": M.ProtonetClassifier(in_channels=416, mid_channels=[64], out_channels=32),

Example #4

import torchvision.transforms.functional as T


iscxvpn2016_loader = iscxvpn2016(pcap_dir="D://Datasets/ISCXVPN2016/", h5_dir="D://Datasets/packets-50k/", max_packets_on_cache=50000, as_bit=True, verbose=True)
ustctfc2016_loader = ustctfc2016(pcap_dir="D://Datasets/USTC-TFC2016/", h5_dir="D://Datasets/USTC-TFC2016-packets-50k/", max_packets_on_cache=50000, as_bit=True, verbose=True)


classifiers = {

    # "relpnet_1": M.RelationNetClassifier_Protonet1(simnet_channels=[128, 64, 32]),
    # "protonet_mini_bit": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=10),

    # "relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),
    # "reg_relation_1": M.RelationNetClassifier(in_channels=416, feature_channels=[10], simnet_channels=[32, 16, 4]),

    "protonet_1": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=32),
    "protonet_2": M.ProtonetClassifier(in_channels=416, mid_channels=[64], out_channels=32),
    "protonet_3": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 64], out_channels=32),
    "protonet_4": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=32),
    # "protonet_bottleneck_end": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=10),
    # "protonet_bottleneck_mid": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 32, 128], out_channels=32),

    # "simnet_simple": M.SimnetClassifier(in_channels=416, channels=[10]),
    # "simnet_1": M.SimnetClassifier(in_channels=416, channels=[32]),
    # "simnet_2": M.SimnetClassifier(in_channels=416, channels=[64, 32]),
    # "simnet_3": M.SimnetClassifier(in_channels=416, channels=[128, 64, 32]),

    # "reg_protonet_1": M.ProtonetClassifier(in_channels=416, mid_channels=[], out_channels=32),
    # "reg_protonet_2": M.ProtonetClassifier(in_channels=416, mid_channels=[64], out_channels=32),
    # "reg_protonet_3": M.ProtonetClassifier(in_channels=416, mid_channels=[128, 64], out_channels=32),
    # "reg_protonet_4": M.ProtonetClassifier(in_channels=416, mid_channels=[256, 128, 64], out_channels=32),