from one_shot_learning_network import *
from experiment_builder import ExperimentBuilder
import tensorflow.contrib.slim as slim
import data as dataset
import tqdm
from utils.parser_utils import get_args
from utils.storage import save_statistics, build_experiment_folder
tf.reset_default_graph()
args = get_args()
# Experiment builder
data = dataset.FolderDatasetLoader(
num_of_gpus=1,
batch_size=args.batch_size,
image_height=28,
image_width=28,
image_channels=1,
train_val_test_split=(1200 / 1622, 211 / 1622, 211 / 1622),
samples_per_iter=1,
num_workers=4,
data_path="datasets/omniglot_data",
name="omniglot_data",
index_of_folder_indicating_class=-2,
reset_stored_filepaths=False,
num_samples_per_class=args.samples_per_class,
num_classes_per_set=args.classes_per_set,
label_as_int=True)
experiment = ExperimentBuilder(data)
one_shot_omniglot, losses, c_error_opt_op, init = experiment.build_experiment(
new_k = k.replace("ligand_arm.5", "lig_norm2")
elif layer_num == 9:
new_k = k.replace("ligand_arm.9", "lig_norm3")
elif "protein" in k:
if layer_num == 0:
new_k = k.replace("protein_arm.0", "prot_conv1")
elif layer_num == 4:
new_k = k.replace("protein_arm.4", "prot_conv2")
elif layer_num == 8:
new_k = k.replace("protein_arm.8", "prot_conv3")
elif layer_num == 1:
new_k = k.replace("protein_arm.1", "prot_norm1")
elif layer_num == 5:
new_k = k.replace("protein_arm.5", "prot_norm2")
elif layer_num == 9:
new_k = k.replace("protein_arm.9", "prot_norm3")
new_dict[new_k] = v
return new_dict
if __name__ == "__main__":
from utils.parser_utils import get_args
args, device = get_args()
model = DeepDTAv2Meta(args=args, device="cpu")
print("Initial", model.state_dict()["lig_conv1.weight"].mean())
model.load_from_path("DEEPDTAV2_SAMPLE_WEIGHT.pth")
print("After", model.state_dict()["lig_conv1.weight"].mean())