repair DDIM/PLMS/UniPC batches

modules/sd_samplers_timesteps.py

@@ -51,10 +51,9 @@ class CFGDenoiserTimesteps(CFGDenoiser):
         self.alphas = shared.sd_model.alphas_cumprod
 
     def get_pred_x0(self, x_in, x_out, sigma):
-        ts = int(sigma.item())
+        ts = sigma.to(dtype=int)
 
-        s_in = x_in.new_ones([x_in.shape[0]])
-        a_t = self.alphas[ts].item() * s_in
+        a_t = self.alphas[ts][:, None, None, None]
         sqrt_one_minus_at = (1 - a_t).sqrt()
 
         pred_x0 = (x_in - sqrt_one_minus_at * x_out) / a_t.sqrt()

modules/sd_samplers_timesteps_impl.py

@@ -16,16 +16,17 @@ def ddim(model, x, timesteps, extra_args=None, callback=None, disable=None, eta=
     sigmas = eta * np.sqrt((1 - alphas_prev.cpu().numpy()) / (1 - alphas.cpu()) * (1 - alphas.cpu() / alphas_prev.cpu().numpy()))
 
     extra_args = {} if extra_args is None else extra_args
-    s_in = x.new_ones([x.shape[0]])
+    s_in = x.new_ones((x.shape[0]))
+    s_x = x.new_ones((x.shape[0], 1, 1, 1))
     for i in tqdm.trange(len(timesteps) - 1, disable=disable):
         index = len(timesteps) - 1 - i
 
         e_t = model(x, timesteps[index].item() * s_in, **extra_args)
 
-        a_t = alphas[index].item() * s_in
-        a_prev = alphas_prev[index].item() * s_in
-        sigma_t = sigmas[index].item() * s_in
-        sqrt_one_minus_at = sqrt_one_minus_alphas[index].item() * s_in
+        a_t = alphas[index].item() * s_x
+        a_prev = alphas_prev[index].item() * s_x
+        sigma_t = sigmas[index].item() * s_x
+        sqrt_one_minus_at = sqrt_one_minus_alphas[index].item() * s_x
 
         pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt()
         dir_xt = (1. - a_prev - sigma_t ** 2).sqrt() * e_t
@@ -47,13 +48,14 @@ def plms(model, x, timesteps, extra_args=None, callback=None, disable=None):
 
     extra_args = {} if extra_args is None else extra_args
     s_in = x.new_ones([x.shape[0]])
+    s_x = x.new_ones((x.shape[0], 1, 1, 1))
     old_eps = []
 
     def get_x_prev_and_pred_x0(e_t, index):
         # select parameters corresponding to the currently considered timestep
-        a_t = alphas[index].item() * s_in
-        a_prev = alphas_prev[index].item() * s_in
-        sqrt_one_minus_at = sqrt_one_minus_alphas[index].item() * s_in
+        a_t = alphas[index].item() * s_x
+        a_prev = alphas_prev[index].item() * s_x
+        sqrt_one_minus_at = sqrt_one_minus_alphas[index].item() * s_x
 
         # current prediction for x_0
         pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt()