def run_probe(args):
device = torch.device(
"cuda:" + str(args.cuda_id) if torch.cuda.is_available() else "cpu")
collect_mode = args.collect_mode if args.method != 'pretrained-rl-agent' else "pretrained_representations"
tr_eps, val_eps, tr_labels, val_labels, test_eps, test_labels = get_episodes(
args, device, collect_mode=collect_mode, train_mode="probe")
print("got episodes!")
observation_shape = tr_eps[0][0].shape
wandb.config.update(vars(args))
if args.train_encoder and args.method in train_encoder_methods:
print("Training encoder from scratch")
encoder = train_encoder(args)
encoder.probing = True
encoder.eval()
else:
if args.encoder_type == "Nature":
encoder = NatureCNN(observation_shape[0], args)
elif args.encoder_type == "Impala":
encoder = ImpalaCNN(observation_shape[0], args)
if args.weights_path == "None":
if args.method not in probe_only_methods:
sys.stderr.write(
"Probing without loading in encoder weights! Are sure you want to do that??"
)
else:
print(
"Print loading in encoder weights from probe of type {} from the following path: {}"
.format(args.method, args.weights_path))
encoder.load_state_dict(torch.load(args.weights_path))
encoder.eval()
torch.set_num_threads(1)
if args.method == 'majority':
test_acc, test_f1score = majority_baseline(tr_labels, test_labels,
wandb)
else:
trainer = ProbeTrainer(encoder,
wandb,
epochs=args.epochs,
sample_label=tr_labels[0][0],
lr=args.probe_lr,
batch_size=args.batch_size,
device=device,
patience=args.patience,
log=False)
trainer.train(tr_eps, val_eps, tr_labels, val_labels)
test_acc, test_f1score = trainer.test(test_eps, test_labels)
print(test_acc)
print(test_f1score)
wandb.log(test_acc)
wandb.log(test_f1score)
def run_probe(args):
wandb.config.update(vars(args))
tr_eps, val_eps, tr_labels, val_labels, test_eps, test_labels = get_episodes(
steps=args.probe_steps,
env_name=args.env_name,
seed=args.seed,
num_processes=args.num_processes,
num_frame_stack=args.num_frame_stack,
downsample=not args.no_downsample,
color=args.color,
entropy_threshold=args.entropy_threshold,
collect_mode=args.probe_collect_mode,
train_mode="probe",
checkpoint_index=args.checkpoint_index,
min_episode_length=args.batch_size)
print("got episodes!")
if args.train_encoder and args.method in train_encoder_methods:
print("Training encoder from scratch")
encoder = train_encoder(args)
encoder.probing = True
encoder.eval()
elif args.method in ["pretrained-rl-agent", "majority"]:
encoder = None
else:
observation_shape = tr_eps[0][0].shape
if args.encoder_type == "Nature":
encoder = NatureCNN(observation_shape[0], args)
elif args.encoder_type == "Impala":
encoder = ImpalaCNN(observation_shape[0], args)
if args.weights_path == "None":
if args.method not in probe_only_methods:
sys.stderr.write(
"Probing without loading in encoder weights! Are sure you want to do that??"
)
else:
print(
"Print loading in encoder weights from probe of type {} from the following path: {}"
.format(args.method, args.weights_path))
encoder.load_state_dict(torch.load(args.weights_path))
encoder.eval()
torch.set_num_threads(1)
if args.method == 'majority':
test_acc, test_f1score = majority_baseline(tr_labels, test_labels,
wandb)
else:
trainer = ProbeTrainer(encoder=encoder,
epochs=args.epochs,
method_name=args.method,
lr=args.probe_lr,
batch_size=args.batch_size,
patience=args.patience,
wandb=wandb,
fully_supervised=(args.method == "supervised"),
save_dir=wandb.run.dir)
trainer.train(tr_eps, val_eps, tr_labels, val_labels)
test_acc, test_f1score = trainer.test(test_eps, test_labels)
# trainer = SKLearnProbeTrainer(encoder=encoder)
# test_acc, test_f1score = trainer.train_test(tr_eps, val_eps, tr_labels, val_labels,
# test_eps, test_labels)
print(test_acc, test_f1score)
wandb.log(test_acc)
wandb.log(test_f1score)
from src.dim_baseline import DIMTrainer
from src.encoders import ImpalaCNN, NatureCNN, NatureOneCNN
from src.global_infonce_stdim import GlobalInfoNCESpatioTemporalTrainer
from src.global_local_infonce import GlobalLocalInfoNCESpatioTemporalTrainer
from src.infonce_spatio_temporal import InfoNCESpatioTemporalTrainer
from src.no_action_feedforward_predictor import NaFFPredictorTrainer
from src.spatio_temporal import SpatioTemporalTrainer
from src.utils import get_argparser
from src.vae import VAETrainer
import torch
import wandb
if __name__ == "__main__":
parser = get_argparser()
print("1")
args = parser.parse_args()
print("2")
tags = ["pretraining-only"]
print("3")
wandb.init(project=args.wandb_proj, entity="curl-atari", tags=tags)
print("4")
config = {}
print("5")
config.update(vars(args))
print("6")
wandb.config.update(config)
print("7")
index_array = torch.randperm(823)
print("8")
train_encoder(args, index_array)