From: François Fleuret Date: Thu, 13 Jul 2023 22:34:50 +0000 (+0200) Subject: Update. X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=commitdiff_plain;h=4b7407bbbd9636b89f663a6a9124e078a16aaef8;p=picoclvr.git Update. --- diff --git a/world.py b/world.py index d32d545..a93684b 100755 --- a/world.py +++ b/world.py @@ -10,6 +10,8 @@ import cairo class Box: + nb_rgb_levels = 10 + def __init__(self, x, y, w, h, r, g, b): self.x = x self.y = y @@ -47,7 +49,12 @@ def scene2tensor(xh, yh, scene, size): ctx.rel_line_to(0, b.h * size) ctx.rel_line_to(-b.w * size, 0) ctx.close_path() - ctx.set_source_rgba(b.r, b.g, b.b, 1.0) + ctx.set_source_rgba( + b.r / (Box.nb_rgb_levels - 1), + b.g / (Box.nb_rgb_levels - 1), + b.b / (Box.nb_rgb_levels - 1), + 1.0, + ) ctx.fill() hs = size * 0.1 @@ -59,17 +66,28 @@ def scene2tensor(xh, yh, scene, size): ctx.close_path() ctx.fill() - return pixel_map[None, :, :, :3].flip(-1).permute(0, 3, 1, 2).float() / 255 + return ( + pixel_map[None, :, :, :3] + .flip(-1) + .permute(0, 3, 1, 2) + .long() + .mul(Box.nb_rgb_levels) + .floor_divide(256) + ) def random_scene(): scene = [] colors = [ - (1.00, 0.00, 0.00), - (0.00, 1.00, 0.00), - (0.60, 0.60, 1.00), - (1.00, 1.00, 0.00), - (0.75, 0.75, 0.75), + ((Box.nb_rgb_levels - 1), 0, 0), + (0, (Box.nb_rgb_levels - 1), 0), + (0, 0, (Box.nb_rgb_levels - 1)), + ((Box.nb_rgb_levels - 1), (Box.nb_rgb_levels - 1), 0), + ( + (Box.nb_rgb_levels * 2) // 3, + (Box.nb_rgb_levels * 2) // 3, + (Box.nb_rgb_levels * 2) // 3, + ), ] for k in range(10): @@ -85,7 +103,7 @@ def random_scene(): return scene -def generate_sequence(nb_steps=10, all_frames=False, size=64): +def generate_episode(nb_steps=10, size=64): delta = 0.1 effects = [ (False, 0, 0), @@ -137,10 +155,6 @@ def generate_sequence(nb_steps=10, all_frames=False, size=64): if xh < 0 or xh > 1 or yh < 0 or yh > 1: xh, yh = x, y - if all_frames: - frames.append(scene2tensor(xh, yh, scene, size=size)) - - if not all_frames: frames.append(scene2tensor(xh, yh, scene, size=size)) if change: @@ -231,8 +245,8 @@ def patchify(x, factor, invert_size=None): class Normalizer(nn.Module): def __init__(self, mu, std): super().__init__() - self.mu = nn.Parameter(mu) - self.log_var = nn.Parameter(2 * torch.log(std)) + self.register_buffer("mu", mu) + self.register_buffer("log_var", 2 * torch.log(std)) def forward(self, x): return (x - self.mu) / torch.exp(self.log_var / 2.0) @@ -254,55 +268,68 @@ class SignSTE(nn.Module): def train_encoder( train_input, - dim_hidden=64, - block_size=16, - nb_bits_per_block=10, + test_input, + depth=2, + dim_hidden=48, + nb_bits_per_token=10, lr_start=1e-3, - lr_end=1e-5, + lr_end=1e-4, nb_epochs=10, batch_size=25, device=torch.device("cpu"), ): - mu, std = train_input.mean(), train_input.std() + mu, std = train_input.float().mean(), train_input.float().std() + + def encoder_core(depth, dim): + l = [ + [ + nn.Conv2d( + dim * 2**k, dim * 2**k, kernel_size=5, stride=1, padding=2 + ), + nn.ReLU(), + nn.Conv2d(dim * 2**k, dim * 2 ** (k + 1), kernel_size=2, stride=2), + nn.ReLU(), + ] + for k in range(depth) + ] + + return nn.Sequential(*[x for m in l for x in m]) + + def decoder_core(depth, dim): + l = [ + [ + nn.ConvTranspose2d( + dim * 2 ** (k + 1), dim * 2**k, kernel_size=2, stride=2 + ), + nn.ReLU(), + nn.ConvTranspose2d( + dim * 2**k, dim * 2**k, kernel_size=5, stride=1, padding=2 + ), + nn.ReLU(), + ] + for k in range(depth - 1, -1, -1) + ] + + return nn.Sequential(*[x for m in l for x in m]) encoder = nn.Sequential( Normalizer(mu, std), - nn.Conv2d(3, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), + nn.Conv2d(3, dim_hidden, kernel_size=1, stride=1), nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d( - dim_hidden, - nb_bits_per_block, - kernel_size=block_size, - stride=block_size, - padding=0, - ), - SignSTE(), + # 64x64 + encoder_core(depth=depth, dim=dim_hidden), + # 8x8 + nn.Conv2d(dim_hidden * 2**depth, nb_bits_per_token, kernel_size=1, stride=1), ) + quantizer = SignSTE() + decoder = nn.Sequential( - nn.ConvTranspose2d( - nb_bits_per_block, - dim_hidden, - kernel_size=block_size, - stride=block_size, - padding=0, - ), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, dim_hidden, kernel_size=5, stride=1, padding=2), - nn.ReLU(), - nn.Conv2d(dim_hidden, 3, kernel_size=5, stride=1, padding=2), + nn.Conv2d(nb_bits_per_token, dim_hidden * 2**depth, kernel_size=1, stride=1), + # 8x8 + decoder_core(depth=depth, dim=dim_hidden), + # 64x64 + nn.ConvTranspose2d(dim_hidden, 3 * Box.nb_rgb_levels, kernel_size=1, stride=1), ) model = nn.Sequential(encoder, decoder) @@ -313,73 +340,120 @@ def train_encoder( model.to(device) + g5x5 = torch.exp(-torch.tensor([[-2.0, -1.0, 0.0, 1.0, 2.0]]) ** 2 / 2) + g5x5 = (g5x5.t() @ g5x5).view(1, 1, 5, 5) + g5x5 = g5x5 / g5x5.sum() + for k in range(nb_epochs): lr = math.exp( math.log(lr_start) + math.log(lr_end / lr_start) / (nb_epochs - 1) * k ) - print(f"lr {lr}") optimizer = torch.optim.Adam(model.parameters(), lr=lr) - acc_loss, nb_samples = 0.0, 0 - - for input in tqdm.tqdm( - train_input.split(batch_size), - dynamic_ncols=True, - desc="vqae-train", - total=train_input.size(0) // batch_size, - ): - output = model(input) - loss = F.mse_loss(output, input) - acc_loss += loss.item() * input.size(0) - nb_samples += input.size(0) + + acc_train_loss = 0.0 + + for input in train_input.split(batch_size): + z = encoder(input) + zq = z if k < 1 else quantizer(z) + output = decoder(zq) + + output = output.reshape( + output.size(0), -1, 3, output.size(2), output.size(3) + ) + + train_loss = F.cross_entropy(output, input) + + acc_train_loss += train_loss.item() * input.size(0) optimizer.zero_grad() - loss.backward() + train_loss.backward() optimizer.step() - print(f"loss {k} {acc_loss/nb_samples}") + acc_test_loss = 0.0 + + for input in test_input.split(batch_size): + z = encoder(input) + zq = z if k < 1 else quantizer(z) + output = decoder(zq) + + output = output.reshape( + output.size(0), -1, 3, output.size(2), output.size(3) + ) + + test_loss = F.cross_entropy(output, input) + + acc_test_loss += test_loss.item() * input.size(0) + + train_loss = acc_train_loss / train_input.size(0) + test_loss = acc_test_loss / test_input.size(0) + + print(f"train_ae {k} lr {lr} train_loss {train_loss} test_loss {test_loss}") sys.stdout.flush() - return encoder, decoder + return encoder, quantizer, decoder + +def generate_episodes(nb): + all_frames = [] + for n in tqdm.tqdm(range(nb), dynamic_ncols=True, desc="world-data"): + frames, actions = generate_episode(nb_steps=31) + all_frames += [ frames[0], frames[-1] ] + return torch.cat(all_frames, 0).contiguous() + +def create_data_and_processors(nb_train_samples, nb_test_samples): + train_input = generate_episodes(nb_train_samples) + test_input = generate_episodes(nb_test_samples) + encoder, quantizer, decoder = train_encoder(train_input, test_input, nb_epochs=2) + + input = test_input[:64] + + z = encoder(input.float()) + height, width = z.size(2), z.size(3) + zq = quantizer(z).long() + pow2=(2**torch.arange(zq.size(1), device=zq.device))[None,None,:] + seq = (zq.permute(0,2,3,1).clamp(min=0).reshape(zq.size(0),-1,zq.size(1)) * pow2).sum(-1) + print(f"{seq.size()=}") + + ZZ=zq + + zq = ((seq[:,:,None] // pow2)%2)*2-1 + zq = zq.reshape(zq.size(0), height, width, -1).permute(0,3,1,2) + + print(ZZ[0]) + print(zq[0]) + + print("CHECK", (ZZ-zq).abs().sum()) + + results = decoder(zq.float()) + T = 0.1 + results = results.reshape( + results.size(0), -1, 3, results.size(2), results.size(3) + ).permute(0, 2, 3, 4, 1) + results = torch.distributions.categorical.Categorical(logits=results / T).sample() + + + torchvision.utils.save_image( + input.float() / (Box.nb_rgb_levels - 1), "orig.png", nrow=8 + ) + + torchvision.utils.save_image( + results.float() / (Box.nb_rgb_levels - 1), "qtiz.png", nrow=8 + ) ###################################################################### if __name__ == "__main__": - import time + create_data_and_processors(250,100) - all_frames = [] - nb = 25000 - start_time = time.perf_counter() - for n in tqdm.tqdm( - range(nb), - dynamic_ncols=True, - desc="world-data", - ): - frames, actions = generate_sequence(nb_steps=31) - all_frames += frames - end_time = time.perf_counter() - print(f"{nb / (end_time - start_time):.02f} samples per second") - - input = torch.cat(all_frames, 0) - encoder, decoder = train_encoder(input) - - # x = patchify(input, 8) - # y = x.reshape(x.size(0), -1) - # print(f"{x.size()=} {y.size()=}") - # centroids, t = kmeans(y, 4096) - # results = centroids[t] - # results = results.reshape(x.size()) - # results = patchify(results, 8, input.size()) - - z = encoder(input) - results = decoder(z) - - print(f"{input.size()=} {z.size()=} {results.size()=}") - - torchvision.utils.save_image(input[:64], "orig.png", nrow=8) - - torchvision.utils.save_image(results[:64], "qtiz.png", nrow=8) - - # frames, actions = generate_sequence(nb_steps=31, all_frames=True) - # frames = torch.cat(frames, 0) - # torchvision.utils.save_image(frames, "seq.png", nrow=8) + # train_input = generate_episodes(2500) + # test_input = generate_episodes(1000) + + # encoder, quantizer, decoder = train_encoder(train_input, test_input) + + # input = test_input[torch.randperm(test_input.size(0))[:64]] + # z = encoder(input.float()) + # zq = quantizer(z) + # results = decoder(zq) + + # T = 0.1 + # results = torch.distributions.categorical.Categorical(logits=results / T).sample()