Merge branch 'master' into gradient-clipping

modules/hypernetworks/hypernetwork.py

@@ -12,7 +12,7 @@ import torch
 import tqdm
 from einops import rearrange, repeat
 from ldm.util import default
-from modules import devices, processing, sd_models, shared
+from modules import devices, processing, sd_models, shared, sd_samplers
 from modules.textual_inversion import textual_inversion
 from modules.textual_inversion.learn_schedule import LearnRateScheduler
 from torch import einsum
@@ -38,7 +38,7 @@ class HypernetworkModule(torch.nn.Module):
     activation_dict.update({cls_name.lower(): cls_obj for cls_name, cls_obj in inspect.getmembers(torch.nn.modules.activation) if inspect.isclass(cls_obj) and cls_obj.__module__ == 'torch.nn.modules.activation'})
 
     def __init__(self, dim, state_dict=None, layer_structure=None, activation_func=None, weight_init='Normal',
-                 add_layer_norm=False, use_dropout=False, activate_output=False, last_layer_dropout=True):
+                 add_layer_norm=False, use_dropout=False, activate_output=False, last_layer_dropout=False):
         super().__init__()
 
         assert layer_structure is not None, "layer_structure must not be None"
@@ -154,16 +154,28 @@ class Hypernetwork:
                 HypernetworkModule(size, None, self.layer_structure, self.activation_func, self.weight_init,
                                    self.add_layer_norm, self.use_dropout, self.activate_output, last_layer_dropout=self.last_layer_dropout),
             )
+        self.eval_mode()
 
     def weights(self):
         res = []
+        for k, layers in self.layers.items():
+            for layer in layers:
+                res += layer.parameters()
+        return res
 
+    def train_mode(self):
         for k, layers in self.layers.items():
             for layer in layers:
                 layer.train()
-                res += layer.trainables()
+                for param in layer.parameters():
+                    param.requires_grad = True
 
-        return res
+    def eval_mode(self):
+        for k, layers in self.layers.items():
+            for layer in layers:
+                layer.eval()
+                for param in layer.parameters():
+                    param.requires_grad = False
 
     def save(self, filename):
         state_dict = {}
@@ -265,7 +277,7 @@ def load_hypernetwork(filename):
             print(traceback.format_exc(), file=sys.stderr)
     else:
         if shared.loaded_hypernetwork is not None:
-            print(f"Unloading hypernetwork")
+            print("Unloading hypernetwork")
 
         shared.loaded_hypernetwork = None
 
@@ -366,19 +378,44 @@ def report_statistics(loss_info:dict):
             print(e)
 
 
+def create_hypernetwork(name, enable_sizes, overwrite_old, layer_structure=None, activation_func=None, weight_init=None, add_layer_norm=False, use_dropout=False):
+    # Remove illegal characters from name.
+    name = "".join( x for x in name if (x.isalnum() or x in "._- "))
 
-def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log_directory, training_width, training_height, steps, clip_grad_mode, clip_grad_value, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
+    fn = os.path.join(shared.cmd_opts.hypernetwork_dir, f"{name}.pt")
+    if not overwrite_old:
+        assert not os.path.exists(fn), f"file {fn} already exists"
+
+    if type(layer_structure) == str:
+        layer_structure = [float(x.strip()) for x in layer_structure.split(",")]
+
+    hypernet = modules.hypernetworks.hypernetwork.Hypernetwork(
+        name=name,
+        enable_sizes=[int(x) for x in enable_sizes],
+        layer_structure=layer_structure,
+        activation_func=activation_func,
+        weight_init=weight_init,
+        add_layer_norm=add_layer_norm,
+        use_dropout=use_dropout,
+    )
+    hypernet.save(fn)
+
+    shared.reload_hypernetworks()
+
+
+def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
     # images allows training previews to have infotext. Importing it at the top causes a circular import problem.
     from modules import images
 
     save_hypernetwork_every = save_hypernetwork_every or 0
     create_image_every = create_image_every or 0
-    textual_inversion.validate_train_inputs(hypernetwork_name, learn_rate, batch_size, data_root, template_file, steps, save_hypernetwork_every, create_image_every, log_directory, name="hypernetwork")
+    textual_inversion.validate_train_inputs(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_hypernetwork_every, create_image_every, log_directory, name="hypernetwork")
 
     path = shared.hypernetworks.get(hypernetwork_name, None)
     shared.loaded_hypernetwork = Hypernetwork()
     shared.loaded_hypernetwork.load(path)
 
+    shared.state.job = "train-hypernetwork"
     shared.state.textinfo = "Initializing hypernetwork training..."
     shared.state.job_count = steps
 
@@ -403,38 +440,37 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log
     hypernetwork = shared.loaded_hypernetwork
     checkpoint = sd_models.select_checkpoint()
 
-    ititial_step = hypernetwork.step or 0
-    if ititial_step >= steps:
-        shared.state.textinfo = f"Model has already been trained beyond specified max steps"
+    initial_step = hypernetwork.step or 0
+    if initial_step >= steps:
+        shared.state.textinfo = "Model has already been trained beyond specified max steps"
         return hypernetwork, filename
 
-    scheduler = LearnRateScheduler(learn_rate, steps, ititial_step)
+    scheduler = LearnRateScheduler(learn_rate, steps, initial_step)
     
-    clip_grad = torch.nn.utils.clip_grad_value_ if clip_grad_mode == "value" else \
-        torch.nn.utils.clip_grad_norm_ if clip_grad_mode == "norm" else \
-        None
+    clip_grad = torch.nn.utils.clip_grad_value_ if clip_grad_mode == "value" else torch.nn.utils.clip_grad_norm_ if clip_grad_mode == "norm" else None
     if clip_grad:
-        clip_grad_sched = LearnRateScheduler(clip_grad_value, steps, ititial_step, verbose=False)
-    
+        clip_grad_sched = LearnRateScheduler(clip_grad_value, steps, initial_step, verbose=False)
+
     # dataset loading may take a while, so input validations and early returns should be done before this
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
-    with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, device=devices.device, template_file=template_file, include_cond=True, batch_size=batch_size)
+
+    pin_memory = shared.opts.pin_memory
+
+    ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, cond_model=shared.sd_model.cond_stage_model, device=devices.device, template_file=template_file, include_cond=True, batch_size=batch_size, gradient_step=gradient_step, shuffle_tags=shuffle_tags, tag_drop_out=tag_drop_out, latent_sampling_method=latent_sampling_method)
+    
+    latent_sampling_method = ds.latent_sampling_method
+
+    dl = modules.textual_inversion.dataset.PersonalizedDataLoader(ds, latent_sampling_method=latent_sampling_method, batch_size=ds.batch_size, pin_memory=pin_memory)
+
+    old_parallel_processing_allowed = shared.parallel_processing_allowed
 
     if unload:
+        shared.parallel_processing_allowed = False
         shared.sd_model.cond_stage_model.to(devices.cpu)
         shared.sd_model.first_stage_model.to(devices.cpu)
 
-    size = len(ds.indexes)
-    loss_dict = defaultdict(lambda : deque(maxlen = 1024))
-    losses = torch.zeros((size,))
-    previous_mean_losses = [0]
-    previous_mean_loss = 0
-    print("Mean loss of {} elements".format(size))
-
     weights = hypernetwork.weights()
-    for weight in weights:
-        weight.requires_grad = True
+    hypernetwork.train_mode()
 
     # Here we use optimizer from saved HN, or we can specify as UI option.
     if hypernetwork.optimizer_name in optimizer_dict:
@@ -452,145 +488,172 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log
             print("Cannot resume from saved optimizer!")
             print(e)
 
+    scaler = torch.cuda.amp.GradScaler()
+    
+    batch_size = ds.batch_size
+    gradient_step = ds.gradient_step
+    # n steps = batch_size * gradient_step * n image processed
+    steps_per_epoch = len(ds) // batch_size // gradient_step
+    max_steps_per_epoch = len(ds) // batch_size - (len(ds) // batch_size) % gradient_step
+    loss_step = 0
+    _loss_step = 0 #internal
+    # size = len(ds.indexes)
+    # loss_dict = defaultdict(lambda : deque(maxlen = 1024))
+    # losses = torch.zeros((size,))
+    # previous_mean_losses = [0]
+    # previous_mean_loss = 0
+    # print("Mean loss of {} elements".format(size))
+
     steps_without_grad = 0
 
     last_saved_file = "<none>"
     last_saved_image = "<none>"
     forced_filename = "<none>"
 
-    pbar = tqdm.tqdm(enumerate(ds), total=steps - ititial_step)
-    for i, entries in pbar:
-        hypernetwork.step = i + ititial_step
-        if len(loss_dict) > 0:
-            previous_mean_losses = [i[-1] for i in loss_dict.values()]
-            previous_mean_loss = mean(previous_mean_losses)
-            
-        scheduler.apply(optimizer, hypernetwork.step)
-        if scheduler.finished:
-            break
+    pbar = tqdm.tqdm(total=steps - initial_step)
+    try:
+        for i in range((steps-initial_step) * gradient_step):
+            if scheduler.finished:
+                break
+            if shared.state.interrupted:
+                break
+            for j, batch in enumerate(dl):
+                # works as a drop_last=True for gradient accumulation
+                if j == max_steps_per_epoch:
+                    break
+                scheduler.apply(optimizer, hypernetwork.step)
+                if scheduler.finished:
+                    break
+                if shared.state.interrupted:
+                    break
 
-        if shared.state.interrupted:
-            break
-
-        if clip_grad:
-            clip_grad_sched.step(hypernetwork.step)
-
-        with torch.autocast("cuda"):
-            c = stack_conds([entry.cond for entry in entries]).to(devices.device)
-            # c = torch.vstack([entry.cond for entry in entries]).to(devices.device)
-            x = torch.stack([entry.latent for entry in entries]).to(devices.device)
-            loss = shared.sd_model(x, c)[0]
-            del x
-            del c
-
-            losses[hypernetwork.step % losses.shape[0]] = loss.item()
-            for entry in entries:
-                loss_dict[entry.filename].append(loss.item())
+                if clip_grad:
+                    clip_grad_sched.step(hypernetwork.step)
                 
-            optimizer.zero_grad()
-            weights[0].grad = None
-            loss.backward()
+                with devices.autocast():
+                    x = batch.latent_sample.to(devices.device, non_blocking=pin_memory)
+                    if tag_drop_out != 0 or shuffle_tags:
+                        shared.sd_model.cond_stage_model.to(devices.device)
+                        c = shared.sd_model.cond_stage_model(batch.cond_text).to(devices.device, non_blocking=pin_memory)
+                        shared.sd_model.cond_stage_model.to(devices.cpu)
+                    else:
+                        c = stack_conds(batch.cond).to(devices.device, non_blocking=pin_memory)
+                    loss = shared.sd_model(x, c)[0] / gradient_step
+                    del x
+                    del c
 
-            if weights[0].grad is None:
-                steps_without_grad += 1
-            else:
-                steps_without_grad = 0
-            assert steps_without_grad < 10, 'no gradient found for the trained weight after backward() for 10 steps in a row; this is a bug; training cannot continue'
+                    _loss_step += loss.item()
+                scaler.scale(loss).backward()
+                
+                # go back until we reach gradient accumulation steps
+                if (j + 1) % gradient_step != 0:
+                    continue
 
-            if clip_grad:
-                clip_grad(weights, clip_grad_sched.learn_rate)
+                if clip_grad:
+                    clip_grad(weights, clip_grad_sched.learn_rate)
+                
+                scaler.step(optimizer)
+                scaler.update()
+                hypernetwork.step += 1
+                pbar.update()
+                optimizer.zero_grad(set_to_none=True)
+                loss_step = _loss_step
+                _loss_step = 0
 
-            optimizer.step()
+                steps_done = hypernetwork.step + 1
+                
+                epoch_num = hypernetwork.step // steps_per_epoch
+                epoch_step = hypernetwork.step % steps_per_epoch
 
-        steps_done = hypernetwork.step + 1
+                pbar.set_description(f"[Epoch {epoch_num}: {epoch_step+1}/{steps_per_epoch}]loss: {loss_step:.7f}")
+                if hypernetwork_dir is not None and steps_done % save_hypernetwork_every == 0:
+                    # Before saving, change name to match current checkpoint.
+                    hypernetwork_name_every = f'{hypernetwork_name}-{steps_done}'
+                    last_saved_file = os.path.join(hypernetwork_dir, f'{hypernetwork_name_every}.pt')
+                    hypernetwork.optimizer_name = optimizer_name
+                    if shared.opts.save_optimizer_state:
+                        hypernetwork.optimizer_state_dict = optimizer.state_dict()
+                    save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, last_saved_file)
+                    hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
 
-        if torch.isnan(losses[hypernetwork.step % losses.shape[0]]): 
-            raise RuntimeError("Loss diverged.")
-        
-        if len(previous_mean_losses) > 1:
-            std = stdev(previous_mean_losses)
-        else:
-            std = 0
-        dataset_loss_info = f"dataset loss:{mean(previous_mean_losses):.3f}" + u"\u00B1" + f"({std / (len(previous_mean_losses) ** 0.5):.3f})"
-        pbar.set_description(dataset_loss_info)
+                textual_inversion.write_loss(log_directory, "hypernetwork_loss.csv", hypernetwork.step, steps_per_epoch, {
+                    "loss": f"{loss_step:.7f}",
+                    "learn_rate": scheduler.learn_rate
+                })
 
-        if hypernetwork_dir is not None and steps_done % save_hypernetwork_every == 0:
-            # Before saving, change name to match current checkpoint.
-            hypernetwork_name_every = f'{hypernetwork_name}-{steps_done}'
-            last_saved_file = os.path.join(hypernetwork_dir, f'{hypernetwork_name_every}.pt')
-            hypernetwork.optimizer_name = optimizer_name
-            if shared.opts.save_optimizer_state:
-                hypernetwork.optimizer_state_dict = optimizer.state_dict()
-            save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, last_saved_file)
-            hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
+                if images_dir is not None and steps_done % create_image_every == 0:
+                    forced_filename = f'{hypernetwork_name}-{steps_done}'
+                    last_saved_image = os.path.join(images_dir, forced_filename)
+                    hypernetwork.eval_mode()
+                    shared.sd_model.cond_stage_model.to(devices.device)
+                    shared.sd_model.first_stage_model.to(devices.device)
 
-        textual_inversion.write_loss(log_directory, "hypernetwork_loss.csv", hypernetwork.step, len(ds), {
-            "loss": f"{previous_mean_loss:.7f}",
-            "learn_rate": scheduler.learn_rate
-        })
+                    p = processing.StableDiffusionProcessingTxt2Img(
+                        sd_model=shared.sd_model,
+                        do_not_save_grid=True,
+                        do_not_save_samples=True,
+                    )
 
-        if images_dir is not None and steps_done % create_image_every == 0:
-            forced_filename = f'{hypernetwork_name}-{steps_done}'
-            last_saved_image = os.path.join(images_dir, forced_filename)
+                    if preview_from_txt2img:
+                        p.prompt = preview_prompt
+                        p.negative_prompt = preview_negative_prompt
+                        p.steps = preview_steps
+                        p.sampler_name = sd_samplers.samplers[preview_sampler_index].name
+                        p.cfg_scale = preview_cfg_scale
+                        p.seed = preview_seed
+                        p.width = preview_width
+                        p.height = preview_height
+                    else:
+                        p.prompt = batch.cond_text[0]
+                        p.steps = 20
+                        p.width = training_width
+                        p.height = training_height
 
-            optimizer.zero_grad()
-            shared.sd_model.cond_stage_model.to(devices.device)
-            shared.sd_model.first_stage_model.to(devices.device)
+                    preview_text = p.prompt
 
-            p = processing.StableDiffusionProcessingTxt2Img(
-                sd_model=shared.sd_model,
-                do_not_save_grid=True,
-                do_not_save_samples=True,
-            )
+                    processed = processing.process_images(p)
+                    image = processed.images[0] if len(processed.images) > 0 else None
 
-            if preview_from_txt2img:
-                p.prompt = preview_prompt
-                p.negative_prompt = preview_negative_prompt
-                p.steps = preview_steps
-                p.sampler_index = preview_sampler_index
-                p.cfg_scale = preview_cfg_scale
-                p.seed = preview_seed
-                p.width = preview_width
-                p.height = preview_height
-            else:
-                p.prompt = entries[0].cond_text
-                p.steps = 20
+                    if unload:
+                        shared.sd_model.cond_stage_model.to(devices.cpu)
+                        shared.sd_model.first_stage_model.to(devices.cpu)
+                    hypernetwork.train_mode()
+                    if image is not None:
+                        shared.state.current_image = image
+                        last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
+                        last_saved_image += f", prompt: {preview_text}"
 
-            preview_text = p.prompt
+                shared.state.job_no = hypernetwork.step
 
-            processed = processing.process_images(p)
-            image = processed.images[0] if len(processed.images)>0 else None
-
-            if unload:
-                shared.sd_model.cond_stage_model.to(devices.cpu)
-                shared.sd_model.first_stage_model.to(devices.cpu)
-
-            if image is not None:
-                shared.state.current_image = image
-                last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
-                last_saved_image += f", prompt: {preview_text}"
-
-        shared.state.job_no = hypernetwork.step
-
-        shared.state.textinfo = f"""
+                shared.state.textinfo = f"""
 <p>
-Loss: {previous_mean_loss:.7f}<br/>
-Step: {hypernetwork.step}<br/>
-Last prompt: {html.escape(entries[0].cond_text)}<br/>
+Loss: {loss_step:.7f}<br/>
+Step: {steps_done}<br/>
+Last prompt: {html.escape(batch.cond_text[0])}<br/>
 Last saved hypernetwork: {html.escape(last_saved_file)}<br/>
 Last saved image: {html.escape(last_saved_image)}<br/>
 </p>
 """
-        
-    report_statistics(loss_dict)
+    except Exception:
+        print(traceback.format_exc(), file=sys.stderr)
+    finally:
+        pbar.leave = False
+        pbar.close()
+        hypernetwork.eval_mode()
+        #report_statistics(loss_dict)
 
     filename = os.path.join(shared.cmd_opts.hypernetwork_dir, f'{hypernetwork_name}.pt')
     hypernetwork.optimizer_name = optimizer_name
     if shared.opts.save_optimizer_state:
         hypernetwork.optimizer_state_dict = optimizer.state_dict()
     save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, filename)
+
     del optimizer
     hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
+    shared.sd_model.cond_stage_model.to(devices.device)
+    shared.sd_model.first_stage_model.to(devices.device)
+    shared.parallel_processing_allowed = old_parallel_processing_allowed
+
     return hypernetwork, filename
 
 def save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, filename):