X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=main.py;h=524715a33bf152a7b364d43d0b3d44c17bfe727a;hb=e2c3b8046c3fddef8aacb74cf5f848d42044897e;hp=45fa68c9eeaf0a0c802d0825ad82c4c1833614e6;hpb=9df1bd18f930f6b4a30b94fed6de684d5ceae3b7;p=culture.git diff --git a/main.py b/main.py index 45fa68c..524715a 100755 --- a/main.py +++ b/main.py @@ -12,7 +12,16 @@ from torch import nn from torch.nn import functional as F import ffutils -import mygpt, tasks +import mygpt +import sky, quizz_machine + +# world quizzes vs. culture quizzes + +###################################################################### + +accuracy_to_make_c_quizzes = 0.975 +nb_new_c_quizzes_for_train = 1000 +nb_new_c_quizzes_for_test = 100 ###################################################################### @@ -73,7 +82,7 @@ parser.add_argument("--deterministic_synthesis", action="store_true", default=Fa parser.add_argument("--nb_gpts", type=int, default=5) -parser.add_argument("--check", action="store_true", default=False) +parser.add_argument("--dirty_debug", action="store_true", default=False) ###################################################################### @@ -84,6 +93,13 @@ if args.result_dir is None: ###################################################################### +if args.dirty_debug: + accuracy_to_make_c_quizzes = 0.0 + nb_new_c_quizzes_for_train = 100 + nb_new_c_quizzes_for_test = 10 + +###################################################################### + default_args = { "model": "37M", "batch_size": 100, @@ -182,8 +198,8 @@ for n in vars(args): ###################################################################### -if args.check: - args.nb_train_samples = 500 +if args.dirty_debug: + args.nb_train_samples = 2500 args.nb_test_samples = 100 if args.physical_batch_size is None: @@ -194,7 +210,8 @@ else: assert args.nb_train_samples % args.batch_size == 0 assert args.nb_test_samples % args.batch_size == 0 -task = tasks.World( +quizz_machine = quizz_machine.QuizzMachine( + sky.Sky(height=6, width=8, nb_birds=3, nb_iterations=2), nb_train_samples=args.nb_train_samples, nb_test_samples=args.nb_test_samples, batch_size=args.physical_batch_size, @@ -207,7 +224,7 @@ task = tasks.World( log_string(f"device {device}") -vocabulary_size = task.vocabulary_size() +vocabulary_size = quizz_machine.vocabulary_size() log_string(f"vocabulary_size {vocabulary_size}") @@ -216,8 +233,10 @@ log_string(f"vocabulary_size {vocabulary_size}") # Compute the entropy of the training tokens token_count = 0 -for input in task.batches(split="train", desc="train-entropy"): - token_count += F.one_hot(input, num_classes=task.vocabulary_size()).sum((0, 1)) +for input in quizz_machine.batches(split="train", desc="train-entropy"): + token_count += F.one_hot(input, num_classes=quizz_machine.vocabulary_size()).sum( + (0, 1) + ) token_probas = token_count / token_count.sum() entropy = -torch.xlogy(token_probas, token_probas).sum() train_set_perplexity = math.exp(entropy) @@ -239,11 +258,11 @@ if args.max_percents_of_test_in_train >= 0: nb_test, nb_in_train = 0, 0 for test_subset in subsets_as_tuples( - task.batches(split="test", desc="test-check"), 25000 + quizz_machine.batches(split="test", desc="test-check"), 25000 ): in_train = set() for train_subset in subsets_as_tuples( - task.batches(split="train", desc="train-check"), 25000 + quizz_machine.batches(split="train", desc="train-check"), 25000 ): in_train.update(test_subset.intersection(train_subset)) nb_in_train += len(in_train) @@ -260,14 +279,14 @@ if args.max_percents_of_test_in_train >= 0: ############################## -def one_epoch(model, task): +def one_epoch(model, quizz_machine): optimizer = torch.optim.Adam(model.parameters(), lr=args.learning_rate) model.train() nb_train_samples, acc_train_loss = 0, 0.0 - for input in task.batches(split="train"): + for input in quizz_machine.batches(split="train"): input = input.to(device) if nb_train_samples % args.batch_size == 0: @@ -292,14 +311,14 @@ def one_epoch(model, task): ###################################################################### -def run_tests(model, task, deterministic_synthesis): +def run_tests(model, quizz_machine, deterministic_synthesis): with torch.autograd.no_grad(): model.eval() nb_test_samples, acc_test_loss = 0, 0.0 nb_samples_accumulated = 0 - for input in task.batches(split="test"): + for input in quizz_machine.batches(split="test"): input = input.to(device) bs = model(mygpt.BracketedSequence(input)) @@ -311,7 +330,7 @@ def run_tests(model, task, deterministic_synthesis): nb_test_samples += input.size(0) - main_test_accuracy = task.produce_results( + main_test_accuracy = quizz_machine.produce_results( n_epoch=n_epoch, model=model, result_dir=args.result_dir, @@ -329,43 +348,59 @@ def run_tests(model, task, deterministic_synthesis): ###################################################################### -def create_quizzes( +def create_c_quizzes( model, other_models, - task, + quizz_machine, nb_for_train=1000, nb_for_test=100, + min_ave_seq_logproba=None, ): kept = [] + sum_logits, sum_nb_c_quizzes = 0, 0 + while sum([x.size(0) for x in kept]) < nb_for_train + nb_for_test: - new_quizzes, nb_correct = task.create_new_quizzes( + nb_to_generate = 4 * (nb_for_train + nb_for_test) + + new_c_quizzes, nb_correct, ave_seq_logproba = quizz_machine.create_c_quizzes( n_epoch=n_epoch, result_dir=args.result_dir, logger=log_string, - nb=4 * (nb_for_train + nb_for_test), + nb=nb_to_generate, model=model, other_models=other_models, + min_ave_seq_logproba=min_ave_seq_logproba, ) - print(nb_correct) + sum_logits += new_c_quizzes.size(0) * ave_seq_logproba + sum_nb_c_quizzes += new_c_quizzes.size(0) + + to_keep = new_c_quizzes[nb_correct == len(other_models) - 1] + + if args.dirty_debug: + to_keep = new_c_quizzes + + log_string( + f"keep {to_keep.size(0)}/{new_c_quizzes.size(0)} c_quizzes ({to_keep.size(0)*100/new_c_quizzes.size(0):.02f}%)" + ) - to_keep = new_quizzes[nb_correct == len(other_models) - 1] - log_string(f"keep {to_keep.size(0)} quizzes") kept.append(to_keep) - new_quizzes = torch.cat(kept, dim=0)[: nb_for_train + nb_for_test] + new_c_quizzes = torch.cat(kept, dim=0)[: nb_for_train + nb_for_test] - task.store_new_quizzes(new_quizzes[:nb_for_train], for_train=True) - task.store_new_quizzes(new_quizzes[nb_for_train:], for_train=False) + quizz_machine.store_c_quizzes(new_c_quizzes[:nb_for_train], for_train=True) + quizz_machine.store_c_quizzes(new_c_quizzes[nb_for_train:], for_train=False) - task.save_image( - new_quizzes[:72], + quizz_machine.problem.save_quizzes( + new_c_quizzes[:72], args.result_dir, - f"world_quiz_{n_epoch:04d}_{model.id:02d}.png", + f"culture_c_quiz_{n_epoch:04d}_{model.id:02d}", log_string, ) + return sum_logits / sum_nb_c_quizzes + ###################################################################### @@ -394,17 +429,12 @@ log_string(f"nb_parameters {nb_parameters} ({int(nb_parameters/1e6)}M)") ###################################################################### -accuracy_to_make_quizzes = 0.975 -nb_new_quizzes_for_train = 1000 -nb_new_quizzes_for_test = 100 - -if args.check: - accuracy_to_make_quizzes = 0.0 - nb_new_quizzes_for_train = 10 - nb_new_quizzes_for_test = 10 +min_ave_seq_logproba = None for n_epoch in range(args.nb_epochs): - a = [(model.id, model.main_test_accuracy) for model in models] + log_string(f"--- epoch {n_epoch} ----------------------------------------") + + a = [(model.id, float(model.main_test_accuracy)) for model in models] a.sort(key=lambda p: p[0]) log_string(f"current accuracies {a}") @@ -417,30 +447,45 @@ for n_epoch in range(args.nb_epochs): ) # improve it - one_epoch(model, task) + one_epoch(model, quizz_machine) + + quizz_machine.renew_w_quizzes(args.nb_train_samples // args.nb_gpts) log_string( - f"train_set_composition world {task.nb_batch_samples_world} quizzes {task.nb_batch_samples_quizzes}" + f"train_set_composition w_quizzes {quizz_machine.nb_batch_w_quizzes} c_quizzes {quizz_machine.nb_batch_c_quizzes}" ) # test it - run_tests(model, task, deterministic_synthesis=False) + run_tests(model, quizz_machine, deterministic_synthesis=False) - if model.main_test_accuracy >= accuracy_to_make_quizzes: + log_string( + f"test_set_composition w_quizzes {quizz_machine.nb_batch_w_quizzes} c_quizzes {quizz_machine.nb_batch_c_quizzes}" + ) + + if min([m.main_test_accuracy for m in models]) >= accuracy_to_make_c_quizzes: other_models = models.copy() other_models.remove(model) - create_quizzes( + ave_seq_logproba = create_c_quizzes( model, other_models, - task, - nb_for_train=nb_new_quizzes_for_train, - nb_for_test=nb_new_quizzes_for_test, + quizz_machine, + nb_for_train=nb_new_c_quizzes_for_train, + nb_for_test=nb_new_c_quizzes_for_test, + min_ave_seq_logproba=min_ave_seq_logproba, ) + # We keep the first average logits as a reference + if min_ave_seq_logproba is None: + min_ave_seq_logproba = ave_seq_logproba + else: + log_string( + f"min_ave_seq_logproba {min_ave_seq_logproba} ave_seq_logproba {ave_seq_logproba}" + ) + # We update everyone for model in models: - run_tests(model, task, deterministic_synthesis=False) + run_tests(model, quizz_machine, deterministic_synthesis=False) ######################################################################