[tutorial] edited hands-on practices (#1899)

* Add handson to ColossalAI.

* Change names of handsons and edit sequence parallel example.

* Edit wrong folder name

* resolve conflict

* delete readme

examples/tutorial/sequence_parallel/model/layers/dropout.py

+import torch
+
+def bias_dropout_add(x, bias, residual, prob, training):
+    # type: (Tensor, Tensor, Tensor, float, bool) -> Tensor
+    out = torch.nn.functional.dropout(x + bias, p=prob, training=training)
+    out = residual + out
+    return out
+
+
+def get_bias_dropout_add(training):
+    def _bias_dropout_add(x, bias, residual, prob):
+        return bias_dropout_add(x, bias, residual, prob, training)
+    return _bias_dropout_add
\ No newline at end of file

examples/tutorial/sequence_parallel/model/layers/embedding.py

+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.nn.init as init
+
+
+class VocabEmbedding(torch.nn.Module):
+
+    def __init__(self, num_embeddings, embedding_dim):
+        super(VocabEmbedding, self).__init__()
+        # Keep the input dimensions.
+        self.num_embeddings = num_embeddings
+        self.embedding_dim = embedding_dim
+        self.padding_idx = None
+        self.max_norm = None
+        self.norm_type = 2.
+        self.scale_grad_by_freq = False
+        self.sparse = False
+        self._weight = None
+
+        # Allocate weights and initialize.
+        self.weight = nn.Parameter(torch.empty(
+            self.num_embeddings, self.embedding_dim))
+        init.xavier_uniform_(self.weight)
+
+    def forward(self, hidden_state):
+        output = F.embedding(hidden_state, self.weight,
+                             self.padding_idx, self.max_norm,
+                             self.norm_type, self.scale_grad_by_freq,
+                             self.sparse)
+        return output
+
+    def __repr__(self):
+        return f'VocabEmbedding(num_embeddings={self.num_embeddings}, ' \
+            f'embedding_dim={self.embedding_dim})'
+
+
+class Embedding(nn.Module):
+    """Language model embeddings.
+    Arguments:
+        hidden_size: hidden size
+        vocab_size: vocabulary size
+        max_sequence_length: maximum size of sequence. This
+                             is used for positional embedding
+        embedding_dropout_prob: dropout probability for embeddings
+        init_method: weight initialization method
+        num_tokentypes: size of the token-type embeddings. 0 value
+                        will ignore this embedding
+    """
+
+    def __init__(self,
+                 hidden_size,
+                 vocab_size,
+                 max_sequence_length,
+                 embedding_dropout_prob,
+                 num_tokentypes):
+        super(Embedding, self).__init__()
+
+        self.hidden_size = hidden_size
+        self.num_tokentypes = num_tokentypes
+
+        self.word_embeddings = VocabEmbedding(vocab_size, self.hidden_size)
+
+        # Position embedding (serial).
+        self.position_embeddings = torch.nn.Embedding(
+            max_sequence_length, self.hidden_size)
+
+        # Token type embedding.
+        # Add this as an optional field that can be added through
+        # method call so we can load a pretrain model without
+        # token types and add them as needed.
+        if self.num_tokentypes > 0:
+            self.tokentype_embeddings = torch.nn.Embedding(self.num_tokentypes,
+                                                           self.hidden_size)
+        else:
+            self.tokentype_embeddings = None
+
+        # Embeddings dropout
+        self.embedding_dropout = torch.nn.Dropout(embedding_dropout_prob)
+
+    @property
+    def word_embedding_weight(self):
+        return self.word_embeddings.weight
+
+    def forward(self, input_ids, position_ids, tokentype_ids=None):
+        # Embeddings.
+        words_embeddings = self.word_embeddings(input_ids)
+        position_embeddings = self.position_embeddings(position_ids)
+        embeddings = words_embeddings + position_embeddings
+        if tokentype_ids is not None and self.tokentype_embeddings is not None:
+            embeddings = embeddings + self.tokentype_embeddings(tokentype_ids)
+
+        # Dropout.
+        embeddings = self.embedding_dropout(embeddings)
+
+        return embeddings

examples/tutorial/sequence_parallel/model/layers/head.py

+import colossalai
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from .pooler import Pooler
+from .linear import Linear
+from .embedding import VocabEmbedding
+from colossalai.core import global_context as gpc
+from colossalai.context import ParallelMode
+from colossalai.kernel import LayerNorm
+from loss_func.cross_entropy import vocab_cross_entropy
+
+
+class BertLMHead(nn.Module):
+    """Masked LM head for Bert
+    Arguments:
+        hidden_size: hidden size
+        init_method: init method for weight initialization
+        layernorm_epsilon: tolerance for layer norm divisions
+    """
+
+    def __init__(self,
+                 vocab_size,
+                 hidden_size,
+                 ):
+
+        super(BertLMHead, self).__init__()
+        self.bias = torch.nn.Parameter(torch.zeros(vocab_size))
+
+        self.dense = Linear(hidden_size, hidden_size)
+        self.layernorm = LayerNorm(hidden_size)
+        self.gelu = torch.nn.functional.gelu
+
+    def forward(self, hidden_states, word_embeddings_weight, lm_labels):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.gelu(hidden_states)
+        hidden_states = self.layernorm(hidden_states)
+
+        output = F.linear(hidden_states, word_embeddings_weight, self.bias)
+        lm_loss = vocab_cross_entropy(output, lm_labels)
+
+        return lm_loss
+
+
+class BertBinaryHead(nn.Module):
+
+    def __init__(self, hidden_size):
+        super().__init__()
+        self.pooler = Pooler(hidden_size)
+        self.dense = Linear(hidden_size, 2)
+
+    def forward(self, hidden_states):
+        if gpc.get_local_rank(ParallelMode.SEQUENCE) == 0:
+            output = self.pooler(hidden_states)
+            output = self.dense(output)
+        else:
+            output = None
+        return output
+
+
+class BertDualHead(nn.Module):
+
+    def __init__(self, hidden_size, vocab_size, add_binary_head):
+        super().__init__()
+        self.lm_head = BertLMHead(vocab_size, hidden_size)
+        self.add_binary_head = add_binary_head
+        if add_binary_head:
+            self.binary_head = BertBinaryHead(hidden_size)
+        else:
+            self.binary_head = None
+
+    def forward(self, hidden_states, word_embeddings_weight, lm_labels):
+        if self.add_binary_head:
+            binary_output = self.binary_head(hidden_states)
+        else:
+            binary_output = None
+        lm_loss = self.lm_head(hidden_states, word_embeddings_weight, lm_labels)
+        return lm_loss, binary_output

examples/tutorial/sequence_parallel/model/layers/init_method.py

+import torch
+import math
+
+def init_normal(tensor, sigma):
+    """Init method based on N(0, sigma)."""
+    torch.nn.init.normal_(tensor, mean=0.0, std=sigma)
+
+
+def output_init_normal(tensor, sigma, num_layers):
+    """Init method based on N(0, sigma/sqrt(2*num_layers)."""
+    std = sigma / math.sqrt(2.0 * num_layers)
+    torch.nn.init.normal_(tensor, mean=0.0, std=std)

examples/tutorial/sequence_parallel/model/layers/linear.py

+import torch
+import torch.nn as nn
+from torch.nn import Parameter
+import torch.nn.functional as F
+import torch.nn.init as init
+
+
+class Linear(nn.Module):
+    """Linear layer with column parallelism.
+    The linear layer is defined as Y = XA + b. A is parallelized along
+    its second dimension as A = [A_1, ..., A_p].
+    Arguments:
+        input_size: first dimension of matrix A.
+        output_size: second dimension of matrix A.
+        bias: If true, add bias
+        init_method: method to initialize weights. Note that bias is always set
+                     to zero.
+        stride: For the strided linear layers.
+        keep_master_weight_for_test: This was added for testing and should be
+                                     set to False. It returns the master weights
+                                     used for initialization.
+        skip_bias_add: This was added to enable performance optimations where bias
+                       can be fused with other elementwise operations. we skip
+                       adding bias but instead return it.
+    """
+
+    def __init__(self,
+                 input_size,
+                 output_size,
+                 bias=True,
+                 skip_bias_add=False):
+        super(Linear, self).__init__()
+
+        # Keep input parameters
+        self.input_size = input_size
+        self.output_size = output_size
+        self.skip_bias_add = skip_bias_add
+
+        self.weight = Parameter(torch.empty(self.output_size,
+                                            self.input_size,
+                                            ))
+        init.normal_(self.weight)
+        if bias:
+            self.bias = Parameter(torch.empty(self.output_size))
+            # Always initialize bias to zero.
+            with torch.no_grad():
+                self.bias.zero_()
+        else:
+            self.register_parameter('bias', None)
+
+    def forward(self, input_):
+        # Matrix multiply.
+        bias = self.bias if not self.skip_bias_add else None
+        output = F.linear(input_, self.weight, bias)
+
+        if self.skip_bias_add:
+            return output, self.bias
+        else:
+            return output
+
+    def __repr__(self):
+        return f'Linear(in_features={self.input_size}, out_features={self.output_size}, ' + \
+            f'bias={self.bias is not None}, skip_bias_add={self.skip_bias_add})'

examples/tutorial/sequence_parallel/model/layers/mlp.py

+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from .linear import Linear
+from colossalai.kernel.jit import bias_gelu_impl
+
+
+class TransformerMLP(nn.Module):
+    """MLP.
+    MLP will take the input with h hidden state, project it to 4*h
+    hidden dimension, perform nonlinear transformation, and project the
+    state back into h hidden dimension. At the end, dropout is also
+    applied.
+    """
+
+    def __init__(self, hidden_size, mlp_ratio, fuse_gelu=True):
+        super(TransformerMLP, self).__init__()
+
+        # Project to 4h.
+        self.dense_h_to_4h = Linear(
+            hidden_size,
+            int(hidden_size*mlp_ratio),
+            skip_bias_add=True)
+
+        self.bias_gelu_fusion = fuse_gelu
+        self.activation_func = F.gelu
+
+        # Project back to h.
+        self.dense_4h_to_h = Linear(
+            int(hidden_size*mlp_ratio),
+            hidden_size,
+            skip_bias_add=True)
+
+    def forward(self, hidden_states):
+        # hidden states should be in the shape of [s, b, h]
+        # it will be projects into [s, b, 4h]
+        # and projected back to [s, b, h]
+        intermediate_parallel, bias_parallel = self.dense_h_to_4h(hidden_states)
+
+        if self.bias_gelu_fusion:
+            intermediate_parallel = \
+                bias_gelu_impl(intermediate_parallel, bias_parallel)
+        else:
+            intermediate_parallel = \
+                self.activation_func(intermediate_parallel + bias_parallel)
+
+        # [s, b, h]
+        output, output_bias = self.dense_4h_to_h(intermediate_parallel)
+        return output, output_bias

examples/tutorial/sequence_parallel/model/layers/pooler.py

+import torch
+import torch.nn as nn
+from .linear import Linear
+
+
+class Pooler(nn.Module):
+    """Pooler layer.
+
+    Pool hidden states of a specific token (for example start of the
+    sequence) and add a linear transformation followed by a tanh.
+
+    Arguments:
+        hidden_size: hidden size
+        init_method: weight initialization method for the linear layer.
+            bias is set to zero.
+    """
+
+    def __init__(self, hidden_size):
+        super(Pooler, self).__init__()
+        self.dense = Linear(hidden_size, hidden_size)
+
+    def forward(self, hidden_states, sequence_index=0):
+        # hidden_states: [b, s, h]
+        # sequence_index: index of the token to pool.
+        pooled = hidden_states[:, sequence_index, :]
+        pooled = self.dense(pooled)
+        pooled = torch.tanh(pooled)
+        return pooled

examples/tutorial/sequence_parallel/model/layers/preprocess.py

+from colossalai.context.parallel_mode import ParallelMode
+import torch
+import torch.nn as nn
+from colossalai.core import global_context as gpc
+
+
+class PreProcessor(nn.Module):
+
+    def __init__(self, sub_seq_length):
+        super().__init__()
+        self.sub_seq_length = sub_seq_length
+
+    def bert_position_ids(self, token_ids):
+        # Create position ids
+        seq_length = token_ids.size(1)
+        local_rank = gpc.get_local_rank(ParallelMode.SEQUENCE)
+        position_ids = torch.arange(seq_length*local_rank,
+                                    seq_length * (local_rank+1),
+                                    dtype=torch.long,
+                                    device=token_ids.device)
+        position_ids = position_ids.unsqueeze(0).expand_as(token_ids)
+
+        return position_ids
+
+    def bert_extended_attention_mask(self, attention_mask):
+        local_rank = gpc.get_local_rank(ParallelMode.SEQUENCE)
+        start_index = local_rank * self.sub_seq_length
+        end_index = (local_rank + 1) * self.sub_seq_length
+
+        # We create a 3D attention mask from a 2D tensor mask.
+        # [b, 1, s]
+        attention_mask_b1s = attention_mask.unsqueeze(1)
+        # [b, s, 1]
+        attention_mask_bs1 = attention_mask.unsqueeze(2)
+        # [b, s/D, s]
+        attention_mask_bss = attention_mask_b1s * attention_mask_bs1
+
+        attention_mask_bss = attention_mask_bss[:, start_index:end_index, :]
+
+        # [b, 1, s/D, s]
+        extended_attention_mask = attention_mask_bss.unsqueeze(1)
+
+        # Convert attention mask to binary:
+        extended_attention_mask = (extended_attention_mask < 0.5)
+
+        return extended_attention_mask
+
+    def forward(self, input_ids=None, attention_mask=None):
+        if attention_mask is not None:
+            extended_attention_mask = self.bert_extended_attention_mask(attention_mask)
+        else:
+            extended_attention_mask = None
+
+        if input_ids is not None:
+            position_ids = self.bert_position_ids(input_ids)
+        else:
+            position_ids = None
+        return position_ids, extended_attention_mask

examples/tutorial/sequence_parallel/train.py

+import colossalai
+from colossalai.context.parallel_mode import ParallelMode
+from colossalai.core import global_context as gpc
+from data import build_train_valid_test_data_iterators
+from data.tokenizer import initialize_tokenizer, get_padded_vocab_size
+from data.bert_helper import get_batch_for_sequence_parallel, SequenceParallelDataIterator
+from colossalai.amp import AMP_TYPE
+from colossalai.logging import get_dist_logger
+from colossalai.utils import MultiTimer, is_using_pp
+from model.bert import BertForPretrain
+from lr_scheduler import AnnealingLR
+from loss_func.bert_loss import BertLoss
+import torch
+from colossalai.engine.schedule import PipelineSchedule
+from colossalai.amp import AMP_TYPE
+from colossalai.nn.optimizer import FusedAdam
+from colossalai.kernel import LayerNorm
+from model.bert import build_pipeline_bert
+
+
+def process_batch_data(batch_data):
+    tokens, types, sentence_order, loss_mask, lm_labels, padding_mask = batch_data
+    if gpc.is_first_rank(ParallelMode.PIPELINE):
+        data = dict(input_ids=tokens, attention_masks=padding_mask, tokentype_ids=types, lm_labels=lm_labels)
+    else:
+        data = dict(attention_masks=padding_mask, tokentype_ids=types, lm_labels=lm_labels)
+    label = dict(loss_mask=loss_mask, sentence_order=sentence_order)
+    return data, label
+
+
+def main():
+    # initialize
+    colossalai.launch_from_torch(config='./config.py', seed=1234, backend='nccl')
+
+    logger = get_dist_logger()
+
+    # build dataloader
+    initialize_tokenizer(gpc.config.VOCAB_FILE_PATH, tokenizer_type='BertWordPieceLowerCase')
+    VOCAB_SIZE = get_padded_vocab_size()
+    trainloader, validloader, testloader = build_train_valid_test_data_iterators(
+        train_iters=gpc.config.TRAIN_ITERS,
+        global_batch_size=gpc.config.GLOBAL_BATCH_SIZE,
+        eval_interval=gpc.config.EVAL_INTERVAL,
+        eval_iters=gpc.config.EVAL_ITERS,
+        data_prefix=[gpc.config.DATA_PATH],
+        data_impl='mmap',
+        splits_string='949,50,1',
+        max_seq_length=gpc.config.SEQ_LENGTH,
+        masked_lm_prob=0.15,
+        short_seq_prob=0.1,
+        seed=1234,
+        skip_warmup=True,
+        binary_head=False,
+    )
+
+    logger.info("Dataloaders are built", ranks=[0])
+
+    # build model
+    if hasattr(gpc.config, 'fp16') and gpc.config.fp16.get('mode') == AMP_TYPE.NAIVE:
+        is_naive_fp16 = True
+    else:
+        is_naive_fp16 = False
+
+    use_pipeline = is_using_pp()
+    kwargs = dict(vocab_size=VOCAB_SIZE,
+                  hidden_size=gpc.config.HIDDEN_SIZE,
+                  max_sequence_length=gpc.config.SEQ_LENGTH,
+                  num_attention_heads=gpc.config.NUM_ATTENTION_HEADS,
+                  convert_fp16_to_fp32_in_softmax=True,
+                  is_naive_fp16=is_naive_fp16,
+                  add_binary_head=gpc.config.ADD_BINARY_HEAD)
+
+    if use_pipeline:
+        model = build_pipeline_bert(num_layers=gpc.config.DEPTH, num_chunks=1, **kwargs)
+    else:
+        model = BertForPretrain(num_layers=gpc.config.DEPTH, **kwargs)
+
+    model = model.half()
+    model.reset_parameters()
+    logger.info(f"Model is built with softmax in fp32 = {is_naive_fp16}", ranks=[0])
+
+    total_numel = 0
+    for p in model.parameters():
+        total_numel += p.numel()
+    logger.info(f"This model has {total_numel} parameters")
+
+    # build criterion
+    criterion = BertLoss()
+    logger.info("Criterion is built", ranks=[0])
+
+    # layernorm and bias has no weight decay
+    weight_decay_params = {'params': []}
+    no_weight_decay_params = {'params': [], 'weight_decay': 0.0}
+    for module_ in model.modules():
+        if isinstance(module_, LayerNorm):
+            no_weight_decay_params['params'].extend([p for p in list(module_._parameters.values()) if p is not None])
+        else:
+            weight_decay_params['params'].extend(
+                [p for n, p in list(module_._parameters.items()) if p is not None and n != 'bias'])
+            no_weight_decay_params['params'].extend(
+                [p for n, p in list(module_._parameters.items()) if p is not None and n == 'bias'])
+
+    logger.info(
+        f"without weight decay param: {len(no_weight_decay_params['params'])}, with weight decay param: {len(weight_decay_params['params'])}"
+    )
+    # optimizer
+    optimizer = FusedAdam((weight_decay_params, no_weight_decay_params),
+                          lr=gpc.config.LR,
+                          weight_decay=gpc.config.WEIGHT_DECAY)
+    logger.info("Optimizer is built", ranks=[0])
+
+    # lr scheduler
+    # follow Megatron-LM setting
+    warmup_steps = int(gpc.config.DECAY_ITERS * gpc.config.WARMUP_FRACTION)
+    lr_scheduler = AnnealingLR(optimizer=optimizer,
+                               max_lr=gpc.config.LR,
+                               min_lr=gpc.config.MIN_LR,
+                               warmup_steps=warmup_steps,
+                               decay_steps=gpc.config.DECAY_ITERS,
+                               decay_style='linear')
+    logger.info(f"LR Scheduler is built with {warmup_steps} warmup steps and {gpc.config.DECAY_ITERS} decay steps")
+
+    # # init
+    engine, *dummy = colossalai.initialize(
+        model,
+        optimizer,
+        criterion,
+    )
+
+    # build timer
+    timer = MultiTimer()
+    skip_iters = 0
+
+    # build loss tracker
+    accumulated_train_loss = torch.zeros(1, dtype=torch.float32).cuda()
+    accumulated_eval_loss = torch.zeros(1, dtype=torch.float32).cuda()
+
+    # build data iters for pipeline parallel
+    if use_pipeline:
+        train_data_iter = SequenceParallelDataIterator(trainloader)
+        valid_data_iter = SequenceParallelDataIterator(validloader)
+
+    for step in range(1, gpc.config.TRAIN_ITERS + 1):
+        timer.start('train-iterations')
+        engine.train()
+        if use_pipeline:
+            engine.zero_grad()
+            _, _, train_loss = engine.execute_schedule(train_data_iter, return_output_label=False)
+            engine.step()
+        else:
+            tokens, types, sentence_order, loss_mask, lm_labels, padding_mask = get_batch_for_sequence_parallel(
+                trainloader)
+            engine.zero_grad()
+            lm_loss, sop_output = engine(tokens, padding_mask, types, lm_labels)
+            train_loss = engine.criterion(lm_loss, sop_output, loss_mask, sentence_order)
+            engine.backward(train_loss)
+            engine.step()
+        timer.stop('train-iterations', keep_in_history=True)
+
+        if not gpc.is_initialized(ParallelMode.PIPELINE) or gpc.is_last_rank(ParallelMode.PIPELINE):
+            accumulated_train_loss += train_loss
+
+        lr_scheduler.step()
+
+        if step % gpc.config.EVAL_INTERVAL == 0:
+            engine.eval()
+
+            for j in range(gpc.config.EVAL_ITERS):
+                with torch.no_grad():
+                    if use_pipeline:
+                        _, _, eval_loss = engine.execute_schedule(valid_data_iter,
+                                                                  forward_only=True,
+                                                                  return_output_label=False)
+                    else:
+                        tokens, types, sentence_order, loss_mask, lm_labels, padding_mask = get_batch_for_sequence_parallel(
+                            validloader)
+                        lm_loss, sop_output = engine(tokens, padding_mask, types, lm_labels)
+                        eval_loss = engine.criterion(lm_loss, sop_output, loss_mask, sentence_order)
+
+                    if not gpc.is_initialized(ParallelMode.PIPELINE) or gpc.is_last_rank(ParallelMode.PIPELINE):
+                        accumulated_eval_loss += eval_loss
+
+            if not gpc.is_initialized(ParallelMode.PIPELINE) or gpc.is_last_rank(ParallelMode.PIPELINE):
+                accumulated_eval_loss /= gpc.config.EVAL_ITERS
+                accumulated_train_loss /= gpc.config.EVAL_INTERVAL
+
+            timer_string = []
+            for n, t in timer:
+                timer_string.append(f"{n}: {t.get_history_mean()*1000:.5f}")
+            timer_string = ' | '.join(timer_string)
+            lr = list(engine.optimizer.param_groups)[0]['lr']
+            loss_scale = engine.optimizer.optim.loss_scale.item()
+
+            if gpc.is_initialized(ParallelMode.PIPELINE):
+                ranks = [gpc.get_ranks_in_group(ParallelMode.PIPELINE)[-1]]
+            else:
+                ranks = [0]
+            logger.info(f'Step {step} / {gpc.config.TRAIN_ITERS} | Train Loss: {accumulated_train_loss.item():.5g} ' +
+                        f'| Eval Loss: {accumulated_eval_loss.item():.5g} ' + f'| Loss Scale: {loss_scale}' +
+                        f"| Learning rate: {lr} | " + timer_string,
+                        ranks=ranks)
+
+            for n, t in timer:
+                t.reset()
+            accumulated_eval_loss.zero_()
+            accumulated_train_loss.zero_()
+
+
+if __name__ == '__main__':
+    main()

examples/tutorial/stable_diffusion/LICENSE

+Copyright (c) 2022 Robin Rombach and Patrick Esser and contributors
+
+CreativeML Open RAIL-M
+dated August 22, 2022
+
+Section I: PREAMBLE
+
+Multimodal generative models are being widely adopted and used, and have the potential to transform the way artists, among other individuals, conceive and benefit from AI or ML technologies as a tool for content creation.
+
+Notwithstanding the current and potential benefits that these artifacts can bring to society at large, there are also concerns about potential misuses of them, either due to their technical limitations or ethical considerations.
+
+In short, this license strives for both the open and responsible downstream use of the accompanying model. When it comes to the open character, we took inspiration from open source permissive licenses regarding the grant of IP rights. Referring to the downstream responsible use, we added use-based restrictions not permitting the use of the Model in very specific scenarios, in order for the licensor to be able to enforce the license in case potential misuses of the Model may occur. At the same time, we strive to promote open and responsible research on generative models for art and content generation.
+
+Even though downstream derivative versions of the model could be released under different licensing terms, the latter will always have to include - at minimum - the same use-based restrictions as the ones in the original license (this license). We believe in the intersection between open and responsible AI development; thus, this License aims to strike a balance between both in order to enable responsible open-science in the field of AI.
+
+This License governs the use of the model (and its derivatives) and is informed by the model card associated with the model.
+
+NOW THEREFORE, You and Licensor agree as follows:
+
+1. Definitions
+
+- "License" means the terms and conditions for use, reproduction, and Distribution as defined in this document.
+- "Data" means a collection of information and/or content extracted from the dataset used with the Model, including to train, pretrain, or otherwise evaluate the Model. The Data is not licensed under this License.
+- "Output" means the results of operating a Model as embodied in informational content resulting therefrom.
+- "Model" means any accompanying machine-learning based assemblies (including checkpoints), consisting of learnt weights, parameters (including optimizer states), corresponding to the model architecture as embodied in the Complementary Material, that have been trained or tuned, in whole or in part on the Data, using the Complementary Material.
+- "Derivatives of the Model" means all modifications to the Model, works based on the Model, or any other model which is created or initialized by transfer of patterns of the weights, parameters, activations or output of the Model, to the other model, in order to cause the other model to perform similarly to the Model, including - but not limited to - distillation methods entailing the use of intermediate data representations or methods based on the generation of synthetic data by the Model for training the other model.
+- "Complementary Material" means the accompanying source code and scripts used to define, run, load, benchmark or evaluate the Model, and used to prepare data for training or evaluation, if any. This includes any accompanying documentation, tutorials, examples, etc, if any.
+- "Distribution" means any transmission, reproduction, publication or other sharing of the Model or Derivatives of the Model to a third party, including providing the Model as a hosted service made available by electronic or other remote means - e.g. API-based or web access.
+- "Licensor" means the copyright owner or entity authorized by the copyright owner that is granting the License, including the persons or entities that may have rights in the Model and/or distributing the Model.
+- "You" (or "Your") means an individual or Legal Entity exercising permissions granted by this License and/or making use of the Model for whichever purpose and in any field of use, including usage of the Model in an end-use application - e.g. chatbot, translator, image generator.
+- "Third Parties" means individuals or legal entities that are not under common control with Licensor or You.
+- "Contribution" means any work of authorship, including the original version of the Model and any modifications or additions to that Model or Derivatives of the Model thereof, that is intentionally submitted to Licensor for inclusion in the Model by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner. For the purposes of this definition, "submitted" means any form of electronic, verbal, or written communication sent to the Licensor or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Model, but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as "Not a Contribution."
+- "Contributor" means Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Model.
+
+Section II: INTELLECTUAL PROPERTY RIGHTS
+
+Both copyright and patent grants apply to the Model, Derivatives of the Model and Complementary Material. The Model and Derivatives of the Model are subject to additional terms as described in Section III.
+
+2. Grant of Copyright License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare, publicly display, publicly perform, sublicense, and distribute the Complementary Material, the Model, and Derivatives of the Model.
+3. Grant of Patent License. Subject to the terms and conditions of this License and where and as applicable, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this paragraph) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Model and the Complementary Material, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Model to which such Contribution(s) was submitted. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Model and/or Complementary Material or a Contribution incorporated within the Model and/or Complementary Material constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for the Model and/or Work shall terminate as of the date such litigation is asserted or filed.
+
+Section III: CONDITIONS OF USAGE, DISTRIBUTION AND REDISTRIBUTION
+
+4. Distribution and Redistribution. You may host for Third Party remote access purposes (e.g. software-as-a-service), reproduce and distribute copies of the Model or Derivatives of the Model thereof in any medium, with or without modifications, provided that You meet the following conditions:
+Use-based restrictions as referenced in paragraph 5 MUST be included as an enforceable provision by You in any type of legal agreement (e.g. a license) governing the use and/or distribution of the Model or Derivatives of the Model, and You shall give notice to subsequent users You Distribute to, that the Model or Derivatives of the Model are subject to paragraph 5. This provision does not apply to the use of Complementary Material.
+You must give any Third Party recipients of the Model or Derivatives of the Model a copy of this License;
+You must cause any modified files to carry prominent notices stating that You changed the files;
+You must retain all copyright, patent, trademark, and attribution notices excluding those notices that do not pertain to any part of the Model, Derivatives of the Model.
+You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions - respecting paragraph 4.a. - for use, reproduction, or Distribution of Your modifications, or for any such Derivatives of the Model as a whole, provided Your use, reproduction, and Distribution of the Model otherwise complies with the conditions stated in this License.
+5. Use-based restrictions. The restrictions set forth in Attachment A are considered Use-based restrictions. Therefore You cannot use the Model and the Derivatives of the Model for the specified restricted uses. You may use the Model subject to this License, including only for lawful purposes and in accordance with the License. Use may include creating any content with, finetuning, updating, running, training, evaluating and/or reparametrizing the Model. You shall require all of Your users who use the Model or a Derivative of the Model to comply with the terms of this paragraph (paragraph 5).
+6. The Output You Generate. Except as set forth herein, Licensor claims no rights in the Output You generate using the Model. You are accountable for the Output you generate and its subsequent uses. No use of the output can contravene any provision as stated in the License.
+
+Section IV: OTHER PROVISIONS
+
+7. Updates and Runtime Restrictions. To the maximum extent permitted by law, Licensor reserves the right to restrict (remotely or otherwise) usage of the Model in violation of this License, update the Model through electronic means, or modify the Output of the Model based on updates. You shall undertake reasonable efforts to use the latest version of the Model.
+8. Trademarks and related. Nothing in this License permits You to make use of Licensors’ trademarks, trade names, logos or to otherwise suggest endorsement or misrepresent the relationship between the parties; and any rights not expressly granted herein are reserved by the Licensors.
+9. Disclaimer of Warranty. Unless required by applicable law or agreed to in writing, Licensor provides the Model and the Complementary Material (and each Contributor provides its Contributions) on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied, including, without limitation, any warranties or conditions of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE. You are solely responsible for determining the appropriateness of using or redistributing the Model, Derivatives of the Model, and the Complementary Material and assume any risks associated with Your exercise of permissions under this License.
+10. Limitation of Liability. In no event and under no legal theory, whether in tort (including negligence), contract, or otherwise, unless required by applicable law (such as deliberate and grossly negligent acts) or agreed to in writing, shall any Contributor be liable to You for damages, including any direct, indirect, special, incidental, or consequential damages of any character arising as a result of this License or out of the use or inability to use the Model and the Complementary Material (including but not limited to damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses), even if such Contributor has been advised of the possibility of such damages.
+11. Accepting Warranty or Additional Liability. While redistributing the Model, Derivatives of the Model and the Complementary Material thereof, You may choose to offer, and charge a fee for, acceptance of support, warranty, indemnity, or other liability obligations and/or rights consistent with this License. However, in accepting such obligations, You may act only on Your own behalf and on Your sole responsibility, not on behalf of any other Contributor, and only if You agree to indemnify, defend, and hold each Contributor harmless for any liability incurred by, or claims asserted against, such Contributor by reason of your accepting any such warranty or additional liability.
+12. If any provision of this License is held to be invalid, illegal or unenforceable, the remaining provisions shall be unaffected thereby and remain valid as if such provision had not been set forth herein.
+
+END OF TERMS AND CONDITIONS
+
+
+
+
+Attachment A
+
+Use Restrictions
+
+You agree not to use the Model or Derivatives of the Model:
+- In any way that violates any applicable national, federal, state, local or international law or regulation;
+- For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
+- To generate or disseminate verifiably false information and/or content with the purpose of harming others;
+- To generate or disseminate personal identifiable information that can be used to harm an individual;
+- To defame, disparage or otherwise harass others;
+- For fully automated decision making that adversely impacts an individual’s legal rights or otherwise creates or modifies a binding, enforceable obligation;
+- For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics;
+- To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
+- For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories;
+- To provide medical advice and medical results interpretation;
+- To generate or disseminate information for the purpose to be used for administration of justice, law enforcement, immigration or asylum processes, such as predicting an individual will commit fraud/crime commitment (e.g. by text profiling, drawing causal relationships between assertions made in documents, indiscriminate and arbitrarily-targeted use).

examples/tutorial/stable_diffusion/README.md

+# Handson 6: Acceleration of Stable Diffusion
+
+*[Colosssal-AI](https://github.com/hpcaitech/ColossalAI) provides a faster and lower cost solution for pretraining and
+fine-tuning for AIGC (AI-Generated Content) applications such as the model [stable-diffusion](https://github.com/CompVis/stable-diffusion) from [Stability AI](https://stability.ai/).*
+
+We take advantage of [Colosssal-AI](https://github.com/hpcaitech/ColossalAI) to exploit multiple optimization strategies
+, e.g. data parallelism, tensor parallelism, mixed precision & ZeRO, to scale the training to multiple GPUs.
+
+## Stable Diffusion
+[Stable Diffusion](https://huggingface.co/CompVis/stable-diffusion) is a latent text-to-image diffusion
+model.
+Thanks to a generous compute donation from [Stability AI](https://stability.ai/) and support from [LAION](https://laion.ai/), we were able to train a Latent Diffusion Model on 512x512 images from a subset of the [LAION-5B](https://laion.ai/blog/laion-5b/) database.
+Similar to Google's [Imagen](https://arxiv.org/abs/2205.11487),
+this model uses a frozen CLIP ViT-L/14 text encoder to condition the model on text prompts.
+
+<p id="diffusion_train" align="center">
+<img src="https://raw.githubusercontent.com/hpcaitech/public_assets/main/colossalai/img/diffusion_train.png" width=800/>
+</p>
+
+[Stable Diffusion with Colossal-AI](https://github.com/hpcaitech/ColossalAI/tree/main/examples/images/diffusion) provides **6.5x faster training and pretraining cost saving, the hardware cost of fine-tuning can be almost 7X cheaper** (from RTX3090/4090 24GB to RTX3050/2070 8GB).
+
+<p id="diffusion_demo" align="center">
+<img src="https://raw.githubusercontent.com/hpcaitech/public_assets/main/colossalai/img/diffusion_demo.png" width=800/>
+</p>
+
+## Requirements
+A suitable [conda](https://conda.io/) environment named `ldm` can be created
+and activated with:
+
+```
+conda env create -f environment.yaml
+conda activate ldm
+```
+
+You can also update an existing [latent diffusion](https://github.com/CompVis/latent-diffusion) environment by running
+
+```
+conda install pytorch torchvision -c pytorch
+pip install transformers==4.19.2 diffusers invisible-watermark
+pip install -e .
+```
+
+### Install [Colossal-AI v0.1.10](https://colossalai.org/download/) From Our Official Website
+```
+pip install colossalai==0.1.10+torch1.11cu11.3 -f https://release.colossalai.org
+```
+
+### Install [Lightning](https://github.com/Lightning-AI/lightning)
+We use the Sep. 2022 version with commit id as `b04a7aa`.
+```
+git clone https://github.com/Lightning-AI/lightning && cd lightning && git reset --hard b04a7aa
+pip install -r requirements.txt && pip install .
+```
+
+> The specified version is due to the interface incompatibility caused by the latest update of [Lightning](https://github.com/Lightning-AI/lightning), which will be fixed in the near future.
+
+## Dataset
+The DataSet is from [LAION-5B](https://laion.ai/blog/laion-5b/), the subset of [LAION](https://laion.ai/),
+you should the change the `data.file_path` in the `config/train_colossalai.yaml`
+
+## Training
+
+we provide the script `train.sh` to run the training task , and two Stategy in `configs`:`train_colossalai.yaml`, `train_ddp.yaml`
+
+for example, you can run the training from colossalai by
+```
+python main.py --logdir /tmp -t --postfix test -b config/train_colossalai.yaml
+```
+
+- you can change the `--logdir` the save the log information and the last checkpoint
+
+### Training config
+you can change the trainging config in the yaml file
+
+- accelerator: acceleratortype, default 'gpu'
+- devices: device number used for training, default 4
+- max_epochs: max training epochs
+- precision: usefp16 for training or not, default 16, you must use fp16 if you want to apply colossalai
+
+
+## Comments
+
+- Our codebase for the diffusion models builds heavily on [OpenAI's ADM codebase](https://github.com/openai/guided-diffusion)
+, [lucidrains](https://github.com/lucidrains/denoising-diffusion-pytorch),
+[Stable Diffusion](https://github.com/CompVis/stable-diffusion), [Lightning](https://github.com/Lightning-AI/lightning) and [Hugging Face](https://huggingface.co/CompVis/stable-diffusion).
+Thanks for open-sourcing!
+
+- The implementation of the transformer encoder is from [x-transformers](https://github.com/lucidrains/x-transformers) by [lucidrains](https://github.com/lucidrains?tab=repositories).
+
+- The implementation of [flash attention](https://github.com/HazyResearch/flash-attention) is from [HazyResearch](https://github.com/HazyResearch).
+
+## BibTeX
+
+```
+@article{bian2021colossal,
+  title={Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training},
+  author={Bian, Zhengda and Liu, Hongxin and Wang, Boxiang and Huang, Haichen and Li, Yongbin and Wang, Chuanrui and Cui, Fan and You, Yang},
+  journal={arXiv preprint arXiv:2110.14883},
+  year={2021}
+}
+@misc{rombach2021highresolution,
+  title={High-Resolution Image Synthesis with Latent Diffusion Models},
+  author={Robin Rombach and Andreas Blattmann and Dominik Lorenz and Patrick Esser and Björn Ommer},
+  year={2021},
+  eprint={2112.10752},
+  archivePrefix={arXiv},
+  primaryClass={cs.CV}
+}
+@article{dao2022flashattention,
+  title={FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness},
+  author={Dao, Tri and Fu, Daniel Y. and Ermon, Stefano and Rudra, Atri and R{\'e}, Christopher},
+  journal={arXiv preprint arXiv:2205.14135},
+  year={2022}
+}
+```

examples/tutorial/stable_diffusion/configs/train_colossalai.yaml

+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: image
+    cond_stage_key: caption
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1.e-4 ]
+        f_min: [ 1.e-10 ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 32 # unused
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/unet/diffusion_pytorch_model.bin'
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: False
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/vae/diffusion_pytorch_model.bin'
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
+      params:
+        use_fp16: True
+
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 64
+    wrap: False
+    train:
+      target: ldm.data.base.Txt2ImgIterableBaseDataset
+      params:
+        file_path: "/data/scratch/diffuser/laion_part0/"
+        world_size: 1
+        rank: 0
+
+lightning:
+  trainer:
+    accelerator: 'gpu' 
+    devices: 4
+    log_gpu_memory: all
+    max_epochs: 2
+    precision: 16
+    auto_select_gpus: False
+    strategy:
+      target: pytorch_lightning.strategies.ColossalAIStrategy
+      params:
+        use_chunk: False
+        enable_distributed_storage: True,
+        placement_policy: cuda
+        force_outputs_fp32: False
+
+    log_every_n_steps: 2
+    logger: True
+    default_root_dir: "/tmp/diff_log/"
+    profiler: pytorch
+
+  logger_config:
+    wandb:
+      target: pytorch_lightning.loggers.WandbLogger
+      params:
+          name: nowname
+          save_dir: "/tmp/diff_log/"
+          offline: opt.debug
+          id: nowname
\ No newline at end of file

examples/tutorial/stable_diffusion/configs/train_ddp.yaml

+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: image
+    cond_stage_key: caption
+    image_size: 32
+    channels: 4
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 100 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1.e-4 ]
+        f_min: [ 1.e-10  ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 32 # unused
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/unet/diffusion_pytorch_model.bin'
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: False
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/vae/diffusion_pytorch_model.bin'
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
+      params:
+        use_fp16: True
+
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 64
+    wrap: False
+    train:
+      target: ldm.data.base.Txt2ImgIterableBaseDataset
+      params:
+        file_path: "/data/scratch/diffuser/laion_part0/"
+        world_size: 1
+        rank: 0
+
+lightning:
+  trainer:
+    accelerator: 'gpu' 
+    devices: 4
+    log_gpu_memory: all
+    max_epochs: 2
+    precision: 16
+    auto_select_gpus: False
+    strategy:
+      target: pytorch_lightning.strategies.DDPStrategy
+      params:
+        find_unused_parameters: False
+    log_every_n_steps: 2
+#    max_steps: 6o
+    logger: True
+    default_root_dir: "/tmp/diff_log/"
+    # profiler: pytorch
+
+  logger_config:
+    wandb:
+      target: pytorch_lightning.loggers.WandbLogger
+      params:
+          name: nowname
+          save_dir: "/tmp/diff_log/"
+          offline: opt.debug
+          id: nowname
\ No newline at end of file

examples/tutorial/stable_diffusion/configs/train_pokemon.yaml

+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: image
+    cond_stage_key: caption
+    image_size: 32
+    channels: 4
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+    check_nan_inf: False
+
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1.e-4 ]
+        f_min: [ 1.e-10  ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 32 # unused
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/unet/diffusion_pytorch_model.bin'
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: False
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        from_pretrained: '/data/scratch/diffuser/stable-diffusion-v1-4/vae/diffusion_pytorch_model.bin'
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
+      params:
+        use_fp16: True
+
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 32
+    wrap: False
+    train:
+      target: ldm.data.pokemon.PokemonDataset
+      # params:
+        # file_path: "/data/scratch/diffuser/laion_part0/"
+        # world_size: 1
+        # rank: 0
+
+lightning:
+  trainer:
+    accelerator: 'gpu' 
+    devices: 4
+    log_gpu_memory: all
+    max_epochs: 2
+    precision: 16
+    auto_select_gpus: False
+    strategy:
+      target: pytorch_lightning.strategies.ColossalAIStrategy
+      params:
+        use_chunk: False
+        enable_distributed_storage: True,
+        placement_policy: cuda
+        force_outputs_fp32: False
+        initial_scale: 65536
+        min_scale: 1
+        max_scale: 65536
+        # max_scale: 4294967296
+
+    log_every_n_steps: 2
+    logger: True
+    default_root_dir: "/tmp/diff_log/"
+    profiler: pytorch
+
+  logger_config:
+    wandb:
+      target: pytorch_lightning.loggers.WandbLogger
+      params:
+          name: nowname
+          save_dir: "/tmp/diff_log/"
+          offline: opt.debug
+          id: nowname
\ No newline at end of file

examples/tutorial/stable_diffusion/environment.yaml

+name: ldm
+channels:
+  - pytorch
+  - defaults
+dependencies:
+  - python=3.9.12
+  - pip=20.3
+  - cudatoolkit=11.3
+  - pytorch=1.11.0
+  - torchvision=0.12.0
+  - numpy=1.19.2
+  - pip:
+    - albumentations==0.4.3
+    - diffusers
+    - opencv-python==4.6.0.66
+    - pudb==2019.2
+    - invisible-watermark
+    - imageio==2.9.0
+    - imageio-ffmpeg==0.4.2
+    - pytorch-lightning==1.4.2
+    - omegaconf==2.1.1
+    - test-tube>=0.7.5
+    - streamlit>=0.73.1
+    - einops==0.3.0
+    - torch-fidelity==0.3.0
+    - transformers==4.19.2
+    - torchmetrics==0.6.0
+    - kornia==0.6
+    - prefetch_generator
+    - -e git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers
+    - -e git+https://github.com/openai/CLIP.git@main#egg=clip
+    - -e .

examples/tutorial/stable_diffusion/ldm/data/base.py

+import math
+from abc import abstractmethod
+
+import torch
+from torch.utils.data import Dataset, ConcatDataset, ChainDataset, IterableDataset
+import os
+import numpy as np
+import cv2
+
+class Txt2ImgIterableBaseDataset(IterableDataset):
+    '''
+    Define an interface to make the IterableDatasets for text2img data chainable
+    '''
+    def __init__(self, file_path: str, rank, world_size):
+        super().__init__()
+        self.file_path = file_path
+        self.folder_list = []
+        self.file_list = []
+        self.txt_list = []
+        self.info = self._get_file_info(file_path)
+        self.start = self.info['start']
+        self.end = self.info['end']
+        self.rank = rank
+
+        self.world_size = world_size
+        # self.per_worker = int(math.floor((self.end - self.start) / float(self.world_size)))
+        # self.iter_start = self.start + self.rank * self.per_worker
+        # self.iter_end = min(self.iter_start + self.per_worker, self.end)
+        # self.num_records = self.iter_end - self.iter_start
+        # self.valid_ids = [i for i in range(self.iter_end)]
+        self.num_records = self.end - self.start
+        self.valid_ids = [i for i in range(self.end)]
+
+        print(f'{self.__class__.__name__} dataset contains {self.__len__()} examples.')
+
+    def __len__(self):
+        # return self.iter_end - self.iter_start
+        return self.end - self.start
+
+    def __iter__(self):
+        sample_iterator = self._sample_generator(self.start, self.end)
+        # sample_iterator = self._sample_generator(self.iter_start, self.iter_end)
+        return sample_iterator
+
+    def _sample_generator(self, start, end):
+        for idx in range(start, end):
+            file_name = self.file_list[idx]
+            txt_name = self.txt_list[idx]
+            f_ = open(txt_name, 'r')
+            txt_ = f_.read()
+            f_.close()
+            image = cv2.imdecode(np.fromfile(file_name, dtype=np.uint8), 1)
+            image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+            image = torch.from_numpy(image) / 255
+            yield {"caption": txt_, "image":image}
+
+
+    def _get_file_info(self, file_path):
+        info = \
+        {
+            "start": 1,
+            "end": 0,
+        }
+        self.folder_list = [file_path + i for i in os.listdir(file_path) if '.' not in i]
+        for folder in self.folder_list:
+            files = [folder + '/' + i for i in os.listdir(folder) if 'jpg' in i]
+            txts = [k.replace('jpg', 'txt') for k in files]
+            self.file_list.extend(files)
+            self.txt_list.extend(txts)
+        info['end'] = len(self.file_list)
+        # with open(file_path, 'r') as fin:
+        #     for _ in enumerate(fin):
+        #         info['end'] += 1
+        # self.txt_list = [k.replace('jpg', 'txt') for k in self.file_list]
+        return info
\ No newline at end of file

examples/tutorial/stable_diffusion/ldm/data/imagenet.py

+import os, yaml, pickle, shutil, tarfile, glob
+import cv2
+import albumentations
+import PIL
+import numpy as np
+import torchvision.transforms.functional as TF
+from omegaconf import OmegaConf
+from functools import partial
+from PIL import Image
+from tqdm import tqdm
+from torch.utils.data import Dataset, Subset
+
+import taming.data.utils as tdu
+from taming.data.imagenet import str_to_indices, give_synsets_from_indices, download, retrieve
+from taming.data.imagenet import ImagePaths
+
+from ldm.modules.image_degradation import degradation_fn_bsr, degradation_fn_bsr_light
+
+
+def synset2idx(path_to_yaml="data/index_synset.yaml"):
+    with open(path_to_yaml) as f:
+        di2s = yaml.load(f)
+    return dict((v,k) for k,v in di2s.items())
+
+
+class ImageNetBase(Dataset):
+    def __init__(self, config=None):
+        self.config = config or OmegaConf.create()
+        if not type(self.config)==dict:
+            self.config = OmegaConf.to_container(self.config)
+        self.keep_orig_class_label = self.config.get("keep_orig_class_label", False)
+        self.process_images = True  # if False we skip loading & processing images and self.data contains filepaths
+        self._prepare()
+        self._prepare_synset_to_human()
+        self._prepare_idx_to_synset()
+        self._prepare_human_to_integer_label()
+        self._load()
+
+    def __len__(self):
+        return len(self.data)
+
+    def __getitem__(self, i):
+        return self.data[i]
+
+    def _prepare(self):
+        raise NotImplementedError()
+
+    def _filter_relpaths(self, relpaths):
+        ignore = set([
+            "n06596364_9591.JPEG",
+        ])
+        relpaths = [rpath for rpath in relpaths if not rpath.split("/")[-1] in ignore]
+        if "sub_indices" in self.config:
+            indices = str_to_indices(self.config["sub_indices"])
+            synsets = give_synsets_from_indices(indices, path_to_yaml=self.idx2syn)  # returns a list of strings
+            self.synset2idx = synset2idx(path_to_yaml=self.idx2syn)
+            files = []
+            for rpath in relpaths:
+                syn = rpath.split("/")[0]
+                if syn in synsets:
+                    files.append(rpath)
+            return files
+        else:
+            return relpaths
+
+    def _prepare_synset_to_human(self):
+        SIZE = 2655750
+        URL = "https://heibox.uni-heidelberg.de/f/9f28e956cd304264bb82/?dl=1"
+        self.human_dict = os.path.join(self.root, "synset_human.txt")
+        if (not os.path.exists(self.human_dict) or
+                not os.path.getsize(self.human_dict)==SIZE):
+            download(URL, self.human_dict)
+
+    def _prepare_idx_to_synset(self):
+        URL = "https://heibox.uni-heidelberg.de/f/d835d5b6ceda4d3aa910/?dl=1"
+        self.idx2syn = os.path.join(self.root, "index_synset.yaml")
+        if (not os.path.exists(self.idx2syn)):
+            download(URL, self.idx2syn)
+
+    def _prepare_human_to_integer_label(self):
+        URL = "https://heibox.uni-heidelberg.de/f/2362b797d5be43b883f6/?dl=1"
+        self.human2integer = os.path.join(self.root, "imagenet1000_clsidx_to_labels.txt")
+        if (not os.path.exists(self.human2integer)):
+            download(URL, self.human2integer)
+        with open(self.human2integer, "r") as f:
+            lines = f.read().splitlines()
+            assert len(lines) == 1000
+            self.human2integer_dict = dict()
+            for line in lines:
+                value, key = line.split(":")
+                self.human2integer_dict[key] = int(value)
+
+    def _load(self):
+        with open(self.txt_filelist, "r") as f:
+            self.relpaths = f.read().splitlines()
+            l1 = len(self.relpaths)
+            self.relpaths = self._filter_relpaths(self.relpaths)
+            print("Removed {} files from filelist during filtering.".format(l1 - len(self.relpaths)))
+
+        self.synsets = [p.split("/")[0] for p in self.relpaths]
+        self.abspaths = [os.path.join(self.datadir, p) for p in self.relpaths]
+
+        unique_synsets = np.unique(self.synsets)
+        class_dict = dict((synset, i) for i, synset in enumerate(unique_synsets))
+        if not self.keep_orig_class_label:
+            self.class_labels = [class_dict[s] for s in self.synsets]
+        else:
+            self.class_labels = [self.synset2idx[s] for s in self.synsets]
+
+        with open(self.human_dict, "r") as f:
+            human_dict = f.read().splitlines()
+            human_dict = dict(line.split(maxsplit=1) for line in human_dict)
+
+        self.human_labels = [human_dict[s] for s in self.synsets]
+
+        labels = {
+            "relpath": np.array(self.relpaths),
+            "synsets": np.array(self.synsets),
+            "class_label": np.array(self.class_labels),
+            "human_label": np.array(self.human_labels),
+        }
+
+        if self.process_images:
+            self.size = retrieve(self.config, "size", default=256)
+            self.data = ImagePaths(self.abspaths,
+                                   labels=labels,
+                                   size=self.size,
+                                   random_crop=self.random_crop,
+                                   )
+        else:
+            self.data = self.abspaths
+
+
+class ImageNetTrain(ImageNetBase):
+    NAME = "ILSVRC2012_train"
+    URL = "http://www.image-net.org/challenges/LSVRC/2012/"
+    AT_HASH = "a306397ccf9c2ead27155983c254227c0fd938e2"
+    FILES = [
+        "ILSVRC2012_img_train.tar",
+    ]
+    SIZES = [
+        147897477120,
+    ]
+
+    def __init__(self, process_images=True, data_root=None, **kwargs):
+        self.process_images = process_images
+        self.data_root = data_root
+        super().__init__(**kwargs)
+
+    def _prepare(self):
+        if self.data_root:
+            self.root = os.path.join(self.data_root, self.NAME)
+        else:
+            cachedir = os.environ.get("XDG_CACHE_HOME", os.path.expanduser("~/.cache"))
+            self.root = os.path.join(cachedir, "autoencoders/data", self.NAME)
+
+        self.datadir = os.path.join(self.root, "data")
+        self.txt_filelist = os.path.join(self.root, "filelist.txt")
+        self.expected_length = 1281167
+        self.random_crop = retrieve(self.config, "ImageNetTrain/random_crop",
+                                    default=True)
+        if not tdu.is_prepared(self.root):
+            # prep
+            print("Preparing dataset {} in {}".format(self.NAME, self.root))
+
+            datadir = self.datadir
+            if not os.path.exists(datadir):
+                path = os.path.join(self.root, self.FILES[0])
+                if not os.path.exists(path) or not os.path.getsize(path)==self.SIZES[0]:
+                    import academictorrents as at
+                    atpath = at.get(self.AT_HASH, datastore=self.root)
+                    assert atpath == path
+
+                print("Extracting {} to {}".format(path, datadir))
+                os.makedirs(datadir, exist_ok=True)
+                with tarfile.open(path, "r:") as tar:
+                    tar.extractall(path=datadir)
+
+                print("Extracting sub-tars.")
+                subpaths = sorted(glob.glob(os.path.join(datadir, "*.tar")))
+                for subpath in tqdm(subpaths):
+                    subdir = subpath[:-len(".tar")]
+                    os.makedirs(subdir, exist_ok=True)
+                    with tarfile.open(subpath, "r:") as tar:
+                        tar.extractall(path=subdir)
+
+            filelist = glob.glob(os.path.join(datadir, "**", "*.JPEG"))
+            filelist = [os.path.relpath(p, start=datadir) for p in filelist]
+            filelist = sorted(filelist)
+            filelist = "\n".join(filelist)+"\n"
+            with open(self.txt_filelist, "w") as f:
+                f.write(filelist)
+
+            tdu.mark_prepared(self.root)
+
+
+class ImageNetValidation(ImageNetBase):
+    NAME = "ILSVRC2012_validation"
+    URL = "http://www.image-net.org/challenges/LSVRC/2012/"
+    AT_HASH = "5d6d0df7ed81efd49ca99ea4737e0ae5e3a5f2e5"
+    VS_URL = "https://heibox.uni-heidelberg.de/f/3e0f6e9c624e45f2bd73/?dl=1"
+    FILES = [
+        "ILSVRC2012_img_val.tar",
+        "validation_synset.txt",
+    ]
+    SIZES = [
+        6744924160,
+        1950000,
+    ]
+
+    def __init__(self, process_images=True, data_root=None, **kwargs):
+        self.data_root = data_root
+        self.process_images = process_images
+        super().__init__(**kwargs)
+
+    def _prepare(self):
+        if self.data_root:
+            self.root = os.path.join(self.data_root, self.NAME)
+        else:
+            cachedir = os.environ.get("XDG_CACHE_HOME", os.path.expanduser("~/.cache"))
+            self.root = os.path.join(cachedir, "autoencoders/data", self.NAME)
+        self.datadir = os.path.join(self.root, "data")
+        self.txt_filelist = os.path.join(self.root, "filelist.txt")
+        self.expected_length = 50000
+        self.random_crop = retrieve(self.config, "ImageNetValidation/random_crop",
+                                    default=False)
+        if not tdu.is_prepared(self.root):
+            # prep
+            print("Preparing dataset {} in {}".format(self.NAME, self.root))
+
+            datadir = self.datadir
+            if not os.path.exists(datadir):
+                path = os.path.join(self.root, self.FILES[0])
+                if not os.path.exists(path) or not os.path.getsize(path)==self.SIZES[0]:
+                    import academictorrents as at
+                    atpath = at.get(self.AT_HASH, datastore=self.root)
+                    assert atpath == path
+
+                print("Extracting {} to {}".format(path, datadir))
+                os.makedirs(datadir, exist_ok=True)
+                with tarfile.open(path, "r:") as tar:
+                    tar.extractall(path=datadir)
+
+                vspath = os.path.join(self.root, self.FILES[1])
+                if not os.path.exists(vspath) or not os.path.getsize(vspath)==self.SIZES[1]:
+                    download(self.VS_URL, vspath)
+
+                with open(vspath, "r") as f:
+                    synset_dict = f.read().splitlines()
+                    synset_dict = dict(line.split() for line in synset_dict)
+
+                print("Reorganizing into synset folders")
+                synsets = np.unique(list(synset_dict.values()))
+                for s in synsets:
+                    os.makedirs(os.path.join(datadir, s), exist_ok=True)
+                for k, v in synset_dict.items():
+                    src = os.path.join(datadir, k)
+                    dst = os.path.join(datadir, v)
+                    shutil.move(src, dst)
+
+            filelist = glob.glob(os.path.join(datadir, "**", "*.JPEG"))
+            filelist = [os.path.relpath(p, start=datadir) for p in filelist]
+            filelist = sorted(filelist)
+            filelist = "\n".join(filelist)+"\n"
+            with open(self.txt_filelist, "w") as f:
+                f.write(filelist)
+
+            tdu.mark_prepared(self.root)
+
+
+
+class ImageNetSR(Dataset):
+    def __init__(self, size=None,
+                 degradation=None, downscale_f=4, min_crop_f=0.5, max_crop_f=1.,
+                 random_crop=True):
+        """
+        Imagenet Superresolution Dataloader
+        Performs following ops in order:
+        1.  crops a crop of size s from image either as random or center crop
+        2.  resizes crop to size with cv2.area_interpolation
+        3.  degrades resized crop with degradation_fn
+
+        :param size: resizing to size after cropping
+        :param degradation: degradation_fn, e.g. cv_bicubic or bsrgan_light
+        :param downscale_f: Low Resolution Downsample factor
+        :param min_crop_f: determines crop size s,
+          where s = c * min_img_side_len with c sampled from interval (min_crop_f, max_crop_f)
+        :param max_crop_f: ""
+        :param data_root:
+        :param random_crop:
+        """
+        self.base = self.get_base()
+        assert size
+        assert (size / downscale_f).is_integer()
+        self.size = size
+        self.LR_size = int(size / downscale_f)
+        self.min_crop_f = min_crop_f
+        self.max_crop_f = max_crop_f
+        assert(max_crop_f <= 1.)
+        self.center_crop = not random_crop
+
+        self.image_rescaler = albumentations.SmallestMaxSize(max_size=size, interpolation=cv2.INTER_AREA)
+
+        self.pil_interpolation = False # gets reset later if incase interp_op is from pillow
+
+        if degradation == "bsrgan":
+            self.degradation_process = partial(degradation_fn_bsr, sf=downscale_f)
+
+        elif degradation == "bsrgan_light":
+            self.degradation_process = partial(degradation_fn_bsr_light, sf=downscale_f)
+
+        else:
+            interpolation_fn = {
+            "cv_nearest": cv2.INTER_NEAREST,
+            "cv_bilinear": cv2.INTER_LINEAR,
+            "cv_bicubic": cv2.INTER_CUBIC,
+            "cv_area": cv2.INTER_AREA,
+            "cv_lanczos": cv2.INTER_LANCZOS4,
+            "pil_nearest": PIL.Image.NEAREST,
+            "pil_bilinear": PIL.Image.BILINEAR,
+            "pil_bicubic": PIL.Image.BICUBIC,
+            "pil_box": PIL.Image.BOX,
+            "pil_hamming": PIL.Image.HAMMING,
+            "pil_lanczos": PIL.Image.LANCZOS,
+            }[degradation]
+
+            self.pil_interpolation = degradation.startswith("pil_")
+
+            if self.pil_interpolation:
+                self.degradation_process = partial(TF.resize, size=self.LR_size, interpolation=interpolation_fn)
+
+            else:
+                self.degradation_process = albumentations.SmallestMaxSize(max_size=self.LR_size,
+                                                                          interpolation=interpolation_fn)
+
+    def __len__(self):
+        return len(self.base)
+
+    def __getitem__(self, i):
+        example = self.base[i]
+        image = Image.open(example["file_path_"])
+
+        if not image.mode == "RGB":
+            image = image.convert("RGB")
+
+        image = np.array(image).astype(np.uint8)
+
+        min_side_len = min(image.shape[:2])
+        crop_side_len = min_side_len * np.random.uniform(self.min_crop_f, self.max_crop_f, size=None)
+        crop_side_len = int(crop_side_len)
+
+        if self.center_crop:
+            self.cropper = albumentations.CenterCrop(height=crop_side_len, width=crop_side_len)
+
+        else:
+            self.cropper = albumentations.RandomCrop(height=crop_side_len, width=crop_side_len)
+
+        image = self.cropper(image=image)["image"]
+        image = self.image_rescaler(image=image)["image"]
+
+        if self.pil_interpolation:
+            image_pil = PIL.Image.fromarray(image)
+            LR_image = self.degradation_process(image_pil)
+            LR_image = np.array(LR_image).astype(np.uint8)
+
+        else:
+            LR_image = self.degradation_process(image=image)["image"]
+
+        example["image"] = (image/127.5 - 1.0).astype(np.float32)
+        example["LR_image"] = (LR_image/127.5 - 1.0).astype(np.float32)
+
+        return example
+
+
+class ImageNetSRTrain(ImageNetSR):
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    def get_base(self):
+        with open("data/imagenet_train_hr_indices.p", "rb") as f:
+            indices = pickle.load(f)
+        dset = ImageNetTrain(process_images=False,)
+        return Subset(dset, indices)
+
+
+class ImageNetSRValidation(ImageNetSR):
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    def get_base(self):
+        with open("data/imagenet_val_hr_indices.p", "rb") as f:
+            indices = pickle.load(f)
+        dset = ImageNetValidation(process_images=False,)
+        return Subset(dset, indices)

examples/tutorial/stable_diffusion/ldm/data/lsun.py

+import os
+import numpy as np
+import PIL
+from PIL import Image
+from torch.utils.data import Dataset
+from torchvision import transforms
+
+
+class LSUNBase(Dataset):
+    def __init__(self,
+                 txt_file,
+                 data_root,
+                 size=None,
+                 interpolation="bicubic",
+                 flip_p=0.5
+                 ):
+        self.data_paths = txt_file
+        self.data_root = data_root
+        with open(self.data_paths, "r") as f:
+            self.image_paths = f.read().splitlines()
+        self._length = len(self.image_paths)
+        self.labels = {
+            "relative_file_path_": [l for l in self.image_paths],
+            "file_path_": [os.path.join(self.data_root, l)
+                           for l in self.image_paths],
+        }
+
+        self.size = size
+        self.interpolation = {"linear": PIL.Image.LINEAR,
+                              "bilinear": PIL.Image.BILINEAR,
+                              "bicubic": PIL.Image.BICUBIC,
+                              "lanczos": PIL.Image.LANCZOS,
+                              }[interpolation]
+        self.flip = transforms.RandomHorizontalFlip(p=flip_p)
+
+    def __len__(self):
+        return self._length
+
+    def __getitem__(self, i):
+        example = dict((k, self.labels[k][i]) for k in self.labels)
+        image = Image.open(example["file_path_"])
+        if not image.mode == "RGB":
+            image = image.convert("RGB")
+
+        # default to score-sde preprocessing
+        img = np.array(image).astype(np.uint8)
+        crop = min(img.shape[0], img.shape[1])
+        h, w, = img.shape[0], img.shape[1]
+        img = img[(h - crop) // 2:(h + crop) // 2,
+              (w - crop) // 2:(w + crop) // 2]
+
+        image = Image.fromarray(img)
+        if self.size is not None:
+            image = image.resize((self.size, self.size), resample=self.interpolation)
+
+        image = self.flip(image)
+        image = np.array(image).astype(np.uint8)
+        example["image"] = (image / 127.5 - 1.0).astype(np.float32)
+        return example
+
+
+class LSUNChurchesTrain(LSUNBase):
+    def __init__(self, **kwargs):
+        super().__init__(txt_file="data/lsun/church_outdoor_train.txt", data_root="data/lsun/churches", **kwargs)
+
+
+class LSUNChurchesValidation(LSUNBase):
+    def __init__(self, flip_p=0., **kwargs):
+        super().__init__(txt_file="data/lsun/church_outdoor_val.txt", data_root="data/lsun/churches",
+                         flip_p=flip_p, **kwargs)
+
+
+class LSUNBedroomsTrain(LSUNBase):
+    def __init__(self, **kwargs):
+        super().__init__(txt_file="data/lsun/bedrooms_train.txt", data_root="data/lsun/bedrooms", **kwargs)
+
+
+class LSUNBedroomsValidation(LSUNBase):
+    def __init__(self, flip_p=0.0, **kwargs):
+        super().__init__(txt_file="data/lsun/bedrooms_val.txt", data_root="data/lsun/bedrooms",
+                         flip_p=flip_p, **kwargs)
+
+
+class LSUNCatsTrain(LSUNBase):
+    def __init__(self, **kwargs):
+        super().__init__(txt_file="data/lsun/cat_train.txt", data_root="data/lsun/cats", **kwargs)
+
+
+class LSUNCatsValidation(LSUNBase):
+    def __init__(self, flip_p=0., **kwargs):
+        super().__init__(txt_file="data/lsun/cat_val.txt", data_root="data/lsun/cats",
+                         flip_p=flip_p, **kwargs)

examples/tutorial/stable_diffusion/ldm/lr_scheduler.py

+import numpy as np
+
+
+class LambdaWarmUpCosineScheduler:
+    """
+    note: use with a base_lr of 1.0
+    """
+    def __init__(self, warm_up_steps, lr_min, lr_max, lr_start, max_decay_steps, verbosity_interval=0):
+        self.lr_warm_up_steps = warm_up_steps
+        self.lr_start = lr_start
+        self.lr_min = lr_min
+        self.lr_max = lr_max
+        self.lr_max_decay_steps = max_decay_steps
+        self.last_lr = 0.
+        self.verbosity_interval = verbosity_interval
+
+    def schedule(self, n, **kwargs):
+        if self.verbosity_interval > 0:
+            if n % self.verbosity_interval == 0: print(f"current step: {n}, recent lr-multiplier: {self.last_lr}")
+        if n < self.lr_warm_up_steps:
+            lr = (self.lr_max - self.lr_start) / self.lr_warm_up_steps * n + self.lr_start
+            self.last_lr = lr
+            return lr
+        else:
+            t = (n - self.lr_warm_up_steps) / (self.lr_max_decay_steps - self.lr_warm_up_steps)
+            t = min(t, 1.0)
+            lr = self.lr_min + 0.5 * (self.lr_max - self.lr_min) * (
+                    1 + np.cos(t * np.pi))
+            self.last_lr = lr
+            return lr
+
+    def __call__(self, n, **kwargs):
+        return self.schedule(n,**kwargs)
+
+
+class LambdaWarmUpCosineScheduler2:
+    """
+    supports repeated iterations, configurable via lists
+    note: use with a base_lr of 1.0.
+    """
+    def __init__(self, warm_up_steps, f_min, f_max, f_start, cycle_lengths, verbosity_interval=0):
+        assert len(warm_up_steps) == len(f_min) == len(f_max) == len(f_start) == len(cycle_lengths)
+        self.lr_warm_up_steps = warm_up_steps
+        self.f_start = f_start
+        self.f_min = f_min
+        self.f_max = f_max
+        self.cycle_lengths = cycle_lengths
+        self.cum_cycles = np.cumsum([0] + list(self.cycle_lengths))
+        self.last_f = 0.
+        self.verbosity_interval = verbosity_interval
+
+    def find_in_interval(self, n):
+        interval = 0
+        for cl in self.cum_cycles[1:]:
+            if n <= cl:
+                return interval
+            interval += 1
+
+    def schedule(self, n, **kwargs):
+        cycle = self.find_in_interval(n)
+        n = n - self.cum_cycles[cycle]
+        if self.verbosity_interval > 0:
+            if n % self.verbosity_interval == 0: print(f"current step: {n}, recent lr-multiplier: {self.last_f}, "
+                                                       f"current cycle {cycle}")
+        if n < self.lr_warm_up_steps[cycle]:
+            f = (self.f_max[cycle] - self.f_start[cycle]) / self.lr_warm_up_steps[cycle] * n + self.f_start[cycle]
+            self.last_f = f
+            return f
+        else:
+            t = (n - self.lr_warm_up_steps[cycle]) / (self.cycle_lengths[cycle] - self.lr_warm_up_steps[cycle])
+            t = min(t, 1.0)
+            f = self.f_min[cycle] + 0.5 * (self.f_max[cycle] - self.f_min[cycle]) * (
+                    1 + np.cos(t * np.pi))
+            self.last_f = f
+            return f
+
+    def __call__(self, n, **kwargs):
+        return self.schedule(n, **kwargs)
+
+
+class LambdaLinearScheduler(LambdaWarmUpCosineScheduler2):
+
+    def schedule(self, n, **kwargs):
+        cycle = self.find_in_interval(n)
+        n = n - self.cum_cycles[cycle]
+        if self.verbosity_interval > 0:
+            if n % self.verbosity_interval == 0: print(f"current step: {n}, recent lr-multiplier: {self.last_f}, "
+                                                       f"current cycle {cycle}")
+
+        if n < self.lr_warm_up_steps[cycle]:
+            f = (self.f_max[cycle] - self.f_start[cycle]) / self.lr_warm_up_steps[cycle] * n + self.f_start[cycle]
+            self.last_f = f
+            return f
+        else:
+            f = self.f_min[cycle] + (self.f_max[cycle] - self.f_min[cycle]) * (self.cycle_lengths[cycle] - n) / (self.cycle_lengths[cycle])
+            self.last_f = f
+            return f
+