from torch.nn import functional as F
import ffutils
-import mygpt, tasks
+import mygpt, tasks, problems
######################################################################
parser.add_argument("--seed", type=int, default=0)
+parser.add_argument("--max_percents_of_test_in_train", type=int, default=1)
+
+########################################
+
parser.add_argument("--nb_epochs", type=int, default=None)
parser.add_argument("--batch_size", type=int, default=None)
parser.add_argument("--learning_rate_schedule", type=str, default="10: 2e-5,30: 4e-6")
+########################################
+
parser.add_argument("--model", type=str, default="37M")
parser.add_argument("--dim_model", type=int, default=None)
parser.add_argument("--dropout", type=float, default=0.1)
+########################################
+
parser.add_argument("--deterministic_synthesis", action="store_true", default=False)
parser.add_argument("--no_checkpoint", action="store_true", default=False)
##############################
# rpl options
-parser.add_argument("--rpl-nb_starting_values", type=int, default=5)
+parser.add_argument("--rpl_nb_starting_values", type=int, default=5)
-parser.add_argument("--rpl-max_input", type=int, default=9)
+parser.add_argument("--rpl_max_input", type=int, default=9)
-parser.add_argument("--rpl-prog_len", type=int, default=10)
+parser.add_argument("--rpl_prog_len", type=int, default=10)
-parser.add_argument("--rpl-nb_runs", type=int, default=8)
+parser.add_argument("--rpl_nb_runs", type=int, default=8)
+
+parser.add_argument("--rpl_no_prog", action="store_true", default=False)
##############################
# sandbox options
if args.task == "sandbox":
if args.sandbox_level == 0:
- problem = tasks.ProblemLevel0(
+ problem = problems.ProblemLevel0(
nb_sentences=args.sandbox_levels_nb_items,
len_prompt=args.sandbox_levels_len_source,
len_result=args.sandbox_levels_len_result,
)
elif args.sandbox_level == 1:
- problem = tasks.ProblemLevel1(
+ problem = problems.ProblemLevel1(
nb_operators=args.sandbox_levels_nb_items,
len_source=args.sandbox_levels_len_source,
len_result=args.sandbox_levels_len_result,
)
elif args.sandbox_level == 2:
- problem = tasks.ProblemLevel2(
+ problem = problems.ProblemLevel2(
len_source=args.sandbox_levels_len_source,
len_result=args.sandbox_levels_len_result,
)
raise ValueError(f"Unknown sandbox level {args.sandbox_level}")
task = tasks.SandBox(
- problem,
- # tasks.ProblemAddition(zero_padded=False, inverted_result=False),
+ # problem,
+ # problems.ProblemAddition(zero_padded=False, inverted_result=False),
+ problems.ProblemLenId(len_max=args.sandbox_levels_len_source),
nb_train_samples=args.nb_train_samples,
nb_test_samples=args.nb_test_samples,
batch_size=args.batch_size,
max_input=args.rpl_max_input,
prog_len=args.rpl_prog_len,
nb_runs=args.rpl_nb_runs,
+ no_prog=args.rpl_no_prog,
logger=log_string,
device=device,
)
if args.task == "expr" and args.expr_input_file is not None:
task.produce_results(
- nb_epochs_finished,
- model,
- args.result_dir,
- log_string,
- args.deterministic_synthesis,
- args.expr_input_file,
+ n_epoch=nb_epochs_finished,
+ model=model,
+ result_dir=args.result_dir,
+ logger=log_string,
+ deterministic_synthesis=args.deterministic_synthesis,
+ input_file=args.expr_input_file,
)
exit(0)
entropy = -torch.xlogy(token_probas, token_probas).sum()
train_set_perplexity = math.exp(entropy)
-##############################
-
+######################################################################
# A bit of paranoia never hurts
-train_examples = {}
+def subsets_as_tuples(batches, cs):
+ s = set()
+ for batch in batches:
+ for x in batch:
+ s.add(tuple([v.item() for v in x]))
+ if len(s) == cs:
+ yield s
+ s = set()
+ yield s
-for input in task.batches(split="train"):
- assert input.dim() == 2 and input.dtype == torch.int64
- for x in input:
- train_examples[x.sum().item()] = x
-
-nb_total, nb_collisions = 0, 0
-for input in task.batches(split="test"):
- assert input.dim() == 2 and input.dtype == torch.int64
- for x in input:
- nb_total += 1
- y = train_examples.get(x.sum().item())
- if y is not None:
- if x.size() == y.size() and (x - y).abs().sum() == 0:
- nb_collisions += 1
-
-del train_examples
+
+nb_test, nb_in_train = 0, 0
+for test_subset in subsets_as_tuples(task.batches(split="test"), 25000):
+ in_train = set()
+ for train_subset in subsets_as_tuples(task.batches(split="train"), 25000):
+ in_train.update(test_subset.intersection(train_subset))
+ nb_in_train += len(in_train)
+ nb_test += len(test_subset)
log_string(
- f"data_check {nb_collisions*100/nb_total:.02f}% ({nb_collisions}/{nb_total}) of test samples are in the train set"
+ f"data_check {nb_in_train*100/nb_test:.02f}% ({nb_in_train}/{nb_test}) of test samples are in the train set"
)
+assert (
+ nb_in_train <= args.max_percents_of_test_in_train * nb_test / 100
+), f"More than {args.max_percents_of_test_in_train}% of test samples are in the train set"
+
##############################
if args.learning_rate_schedule == "cos":
if nb_epochs_finished >= nb_epochs:
task.produce_results(
- nb_epochs_finished,
- model,
- args.result_dir,
- log_string,
- args.deterministic_synthesis,
+ n_epoch=nb_epochs_finished,
+ model=model,
+ result_dir=args.result_dir,
+ logger=log_string,
+ deterministic_synthesis=args.deterministic_synthesis,
)
for n_epoch in range(nb_epochs_finished, nb_epochs):
)
task.produce_results(
- n_epoch, model, args.result_dir, log_string, args.deterministic_synthesis
+ n_epoch=n_epoch,
+ model=model,
+ result_dir=args.result_dir,
+ logger=log_string,
+ deterministic_synthesis=args.deterministic_synthesis,
)
checkpoint = {