From: François Fleuret Date: Sun, 21 Jan 2024 22:41:09 +0000 (+0100) Subject: Update. X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=commitdiff_plain;h=3d7db5b3c1304fdbd599c2a001b5c31df4df2599;p=mygptrnn.git Update. --- diff --git a/fridge b/fridge index 2cc6d01..82d2b17 100644 --- a/fridge +++ b/fridge @@ -302,3 +302,17 @@ class Calibrator: # G = ( # torch.einsum("ntc,hrc->nhrt", X, self.w_G) + self.b_G[None, :, :, None] # ).softmax(dim=2) + +###################################################################### + +2024 Jan 21 16:55:24 (from main.py) + + with open("test.dat", "a") as f: + for m filter(lambda m: isinstance(m,mygpt.Catenn.Linear),model.modules()): + for p in m.parameters() ] + + + for m in model.modules(): + if isinstance(m, mygpt.Caterpillar): + + diff --git a/mygpt.py b/mygpt.py index b137cdb..040845e 100755 --- a/mygpt.py +++ b/mygpt.py @@ -493,10 +493,8 @@ class Caterpillar(nn.Module): warnings.warn("Caterpillar", RuntimeWarning) - def randw(*d, amplitude=None): - if amplitude is None: - amplitude = 1 / math.sqrt(d[-1]) - return nn.Parameter(amplitude * torch.randn(*d)) + def randw(*d, factor=1): + return nn.Parameter(torch.randn(*d) * factor / math.sqrt(d[-1])) self.caterpillar_length = caterpillar_length self.caterpillar_height = caterpillar_height @@ -508,12 +506,11 @@ class Caterpillar(nn.Module): ###################################################################### - default_bg = -math.log(caterpillar_height - 1) - self.w_G = randw(nb_heads, caterpillar_height, dim_model) - self.b_G = nn.Parameter(torch.full((nb_heads, caterpillar_height), default_bg)) + self.w_G = randw(nb_heads, caterpillar_height, dim_model, factor=1.0) + self.b_G = nn.Parameter(torch.full((nb_heads, caterpillar_height), 0.0)) self.w_K = randw(nb_heads, dim_qk, dim_model) - self.w_V = randw(nb_heads, dim_v, dim_model) + self.w_V = randw(nb_heads, dim_v, dim_model, factor=1) self.w_Q = randw(nb_heads, dim_qk, dim_model) self.w_O = randw(dim_v * nb_heads, dim_model) @@ -586,17 +583,19 @@ class Caterpillar(nn.Module): # Clip the gating to avoid values greater than 1 when several # heads hit the same row - G = G / G.sum(1, keepdim=True).clamp(min=1) + # G = G / G.sum(1, keepdim=True).clamp(min=1) + + H = (1 - G).log().sum(1, keepdim=True).exp() ###################################################################### def recurrence(G, V, K): # We prepare the arguments for the parallel scan - A = 1 - G.sum(1) + A = H - gated_V = torch.einsum("nhrt,nhtd->nrtd", G, V) - gated_K = torch.einsum("nhrt,nhtd->nrtd", G, K) + gated_V = torch.einsum("nhrt,nhtd->nrtd", H * G / (1 - G), V) + gated_K = torch.einsum("nhrt,nhtd->nrtd", H * G / (1 - G), K) # We start from cached values, which matters in inference @@ -653,6 +652,8 @@ class Caterpillar(nn.Module): next_V = next_V.detach() next_K = next_K.detach() + warnings.warn("the rescaling is probably a bad idea", RuntimeWarning) + next_V = next_V + (masked_next_V - masked_next_V.detach()) / ( 1 - self.gate_dropout_proba ) @@ -814,7 +815,7 @@ class MyGPT(nn.Module): causal=False, dropout=0.0, len_max=1e5, - attention_layer="kvrec", + attention_layer="caterpillar", logger=print, args=None, ): @@ -1019,7 +1020,111 @@ class MyGPT(nn.Module): ###################################################################### if __name__ == "__main__": - print("Basic check.") + import argparse + + import numpy as np + import matplotlib.pyplot as plt + import matplotlib.collections as mc + + args = argparse.Namespace( + gate_dropout_proba=0.0, gate_dropout_sync=True, gate_dropout_replace=False + ) + + device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + + dim_model, dim_keys, nb_heads = 512, 64, 1 + dropout = 0.1 + + caterpillar = Caterpillar( + dim_model=dim_model, + dim_qk=dim_keys, + dim_v=dim_model // nb_heads, + nb_heads=nb_heads, + caterpillar_length=16, + caterpillar_height=32, + attention_dropout=dropout, + args=args, + ).to(device) + + qkv = QKVAttention( + dim_model=dim_model, + dim_qk=dim_keys, + dim_v=dim_model // nb_heads, + nb_heads=nb_heads, + causal=True, + attention_dropout=dropout, + args=args, + ).to(device) + + linear = CacheWrapper(nn.Linear(512, 512)).to(device) + + x = torch.randn(1, 256, dim_model) + + x = x.to(device) + x.requires_grad_() + + ###################################################################### + + fig = plt.figure() + fig.set_figheight(6) + fig.set_figwidth(8) + + ax = fig.add_subplot(1, 1, 1) + + # ax.set_xlim(-1.5, 1.5) + # ax.set_ylim(-1.5, 1.5) + # ax.set(aspect=1) + # ax.spines.right.set_visible(False) + # ax.spines.top.set_visible(False) + + # dt = 0.01 + # t = np.arange(dt, 20.0, dt) + # ax.semilogx(t, np.exp(-t / 5.0)) + # ax.grid() + + ###################################################################### + + for label, model in [ + # ("nn.Linear", linear), + ("mygpy.QKVAttention", qkv), + ("mygpt.Caterpillar", caterpillar), + ]: + y = model(BracketedSequence(x, 32, x.size(1) - 32, init_cache=True)).x + + data = [] + for t in range(y.size(1)): + for d in torch.randperm(y.size(2))[:8]: + g = torch.autograd.grad(y[0, t, d], x, retain_graph=True)[0] + sg = g.pow(2).sum().item() + # sg = 0 + # for p in model.parameters(): + # g = torch.autograd.grad(y[0, t, d], p, retain_graph=True)[0] + # sg = sg + g.pow(2).sum().item() + data.append([t, sg]) + + data = torch.tensor(data) + ax.scatter( + data[:, 0], data[:, 1], s=1, label=label + ) # , color='gray', label='Input') + + # ax.legend(frameon=False, loc="top right") + + # Put a legend to the right of the current axis + box = ax.get_position() + ax.set_position([box.x0, box.y0, box.width * 0.8, box.height]) + ax.legend(loc="center left", bbox_to_anchor=(1, 0.5)) + + filename = "plot.pdf" + print(f"saving {filename}") + fig.savefig(filename, bbox_inches="tight") + + # if args.window and hasattr(plt.get_current_fig_manager(), 'window'): + # plt.get_current_fig_manager().window.setGeometry(2, 2, 1024, 768) + # plt.show() + + exit(0) + + ###################################################################### m = Caterpillar( dim_model=4, @@ -1041,8 +1146,6 @@ if __name__ == "__main__": print((y1 - torch.cat([y3a, y3b], dim=1)).abs().max()) exit(0) - device = torch.device("cuda" if torch.cuda.is_available() else "cpu") - vocabulary_size = 128 x = torch.randint(vocabulary_size, (6, 1024))