aboutsummaryrefslogtreecommitdiffstats
path: root/ldm/modules/encoders
diff options
context:
space:
mode:
Diffstat (limited to 'ldm/modules/encoders')
-rw-r--r--ldm/modules/encoders/__init__.py0
-rw-r--r--ldm/modules/encoders/modules.py234
-rw-r--r--ldm/modules/encoders/xlmr.py137
3 files changed, 371 insertions, 0 deletions
diff --git a/ldm/modules/encoders/__init__.py b/ldm/modules/encoders/__init__.py
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/ldm/modules/encoders/__init__.py
diff --git a/ldm/modules/encoders/modules.py b/ldm/modules/encoders/modules.py
new file mode 100644
index 00000000..ededbe43
--- /dev/null
+++ b/ldm/modules/encoders/modules.py
@@ -0,0 +1,234 @@
+import torch
+import torch.nn as nn
+from functools import partial
+import clip
+from einops import rearrange, repeat
+from transformers import CLIPTokenizer, CLIPTextModel
+import kornia
+
+from ldm.modules.x_transformer import Encoder, TransformerWrapper # TODO: can we directly rely on lucidrains code and simply add this as a reuirement? --> test
+
+
+class AbstractEncoder(nn.Module):
+ def __init__(self):
+ super().__init__()
+
+ def encode(self, *args, **kwargs):
+ raise NotImplementedError
+
+
+
+class ClassEmbedder(nn.Module):
+ def __init__(self, embed_dim, n_classes=1000, key='class'):
+ super().__init__()
+ self.key = key
+ self.embedding = nn.Embedding(n_classes, embed_dim)
+
+ def forward(self, batch, key=None):
+ if key is None:
+ key = self.key
+ # this is for use in crossattn
+ c = batch[key][:, None]
+ c = self.embedding(c)
+ return c
+
+
+class TransformerEmbedder(AbstractEncoder):
+ """Some transformer encoder layers"""
+ def __init__(self, n_embed, n_layer, vocab_size, max_seq_len=77, device="cuda"):
+ super().__init__()
+ self.device = device
+ self.transformer = TransformerWrapper(num_tokens=vocab_size, max_seq_len=max_seq_len,
+ attn_layers=Encoder(dim=n_embed, depth=n_layer))
+
+ def forward(self, tokens):
+ tokens = tokens.to(self.device) # meh
+ z = self.transformer(tokens, return_embeddings=True)
+ return z
+
+ def encode(self, x):
+ return self(x)
+
+
+class BERTTokenizer(AbstractEncoder):
+ """ Uses a pretrained BERT tokenizer by huggingface. Vocab size: 30522 (?)"""
+ def __init__(self, device="cuda", vq_interface=True, max_length=77):
+ super().__init__()
+ from transformers import BertTokenizerFast # TODO: add to reuquirements
+ self.tokenizer = BertTokenizerFast.from_pretrained("bert-base-uncased")
+ self.device = device
+ self.vq_interface = vq_interface
+ self.max_length = max_length
+
+ def forward(self, text):
+ batch_encoding = self.tokenizer(text, truncation=True, max_length=self.max_length, return_length=True,
+ return_overflowing_tokens=False, padding="max_length", return_tensors="pt")
+ tokens = batch_encoding["input_ids"].to(self.device)
+ return tokens
+
+ @torch.no_grad()
+ def encode(self, text):
+ tokens = self(text)
+ if not self.vq_interface:
+ return tokens
+ return None, None, [None, None, tokens]
+
+ def decode(self, text):
+ return text
+
+
+class BERTEmbedder(AbstractEncoder):
+ """Uses the BERT tokenizr model and add some transformer encoder layers"""
+ def __init__(self, n_embed, n_layer, vocab_size=30522, max_seq_len=77,
+ device="cuda",use_tokenizer=True, embedding_dropout=0.0):
+ super().__init__()
+ self.use_tknz_fn = use_tokenizer
+ if self.use_tknz_fn:
+ self.tknz_fn = BERTTokenizer(vq_interface=False, max_length=max_seq_len)
+ self.device = device
+ self.transformer = TransformerWrapper(num_tokens=vocab_size, max_seq_len=max_seq_len,
+ attn_layers=Encoder(dim=n_embed, depth=n_layer),
+ emb_dropout=embedding_dropout)
+
+ def forward(self, text):
+ if self.use_tknz_fn:
+ tokens = self.tknz_fn(text)#.to(self.device)
+ else:
+ tokens = text
+ z = self.transformer(tokens, return_embeddings=True)
+ return z
+
+ def encode(self, text):
+ # output of length 77
+ return self(text)
+
+
+class SpatialRescaler(nn.Module):
+ def __init__(self,
+ n_stages=1,
+ method='bilinear',
+ multiplier=0.5,
+ in_channels=3,
+ out_channels=None,
+ bias=False):
+ super().__init__()
+ self.n_stages = n_stages
+ assert self.n_stages >= 0
+ assert method in ['nearest','linear','bilinear','trilinear','bicubic','area']
+ self.multiplier = multiplier
+ self.interpolator = partial(torch.nn.functional.interpolate, mode=method)
+ self.remap_output = out_channels is not None
+ if self.remap_output:
+ print(f'Spatial Rescaler mapping from {in_channels} to {out_channels} channels after resizing.')
+ self.channel_mapper = nn.Conv2d(in_channels,out_channels,1,bias=bias)
+
+ def forward(self,x):
+ for stage in range(self.n_stages):
+ x = self.interpolator(x, scale_factor=self.multiplier)
+
+
+ if self.remap_output:
+ x = self.channel_mapper(x)
+ return x
+
+ def encode(self, x):
+ return self(x)
+
+class FrozenCLIPEmbedder(AbstractEncoder):
+ """Uses the CLIP transformer encoder for text (from Hugging Face)"""
+ def __init__(self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77):
+ super().__init__()
+ self.tokenizer = CLIPTokenizer.from_pretrained(version)
+ self.transformer = CLIPTextModel.from_pretrained(version)
+ self.device = device
+ self.max_length = max_length
+ self.freeze()
+
+ def freeze(self):
+ self.transformer = self.transformer.eval()
+ for param in self.parameters():
+ param.requires_grad = False
+
+ def forward(self, text):
+ batch_encoding = self.tokenizer(text, truncation=True, max_length=self.max_length, return_length=True,
+ return_overflowing_tokens=False, padding="max_length", return_tensors="pt")
+ tokens = batch_encoding["input_ids"].to(self.device)
+ outputs = self.transformer(input_ids=tokens)
+
+ z = outputs.last_hidden_state
+ return z
+
+ def encode(self, text):
+ return self(text)
+
+
+class FrozenCLIPTextEmbedder(nn.Module):
+ """
+ Uses the CLIP transformer encoder for text.
+ """
+ def __init__(self, version='ViT-L/14', device="cuda", max_length=77, n_repeat=1, normalize=True):
+ super().__init__()
+ self.model, _ = clip.load(version, jit=False, device="cpu")
+ self.device = device
+ self.max_length = max_length
+ self.n_repeat = n_repeat
+ self.normalize = normalize
+
+ def freeze(self):
+ self.model = self.model.eval()
+ for param in self.parameters():
+ param.requires_grad = False
+
+ def forward(self, text):
+ tokens = clip.tokenize(text).to(self.device)
+ z = self.model.encode_text(tokens)
+ if self.normalize:
+ z = z / torch.linalg.norm(z, dim=1, keepdim=True)
+ return z
+
+ def encode(self, text):
+ z = self(text)
+ if z.ndim==2:
+ z = z[:, None, :]
+ z = repeat(z, 'b 1 d -> b k d', k=self.n_repeat)
+ return z
+
+
+class FrozenClipImageEmbedder(nn.Module):
+ """
+ Uses the CLIP image encoder.
+ """
+ def __init__(
+ self,
+ model,
+ jit=False,
+ device='cuda' if torch.cuda.is_available() else 'cpu',
+ antialias=False,
+ ):
+ super().__init__()
+ self.model, _ = clip.load(name=model, device=device, jit=jit)
+
+ self.antialias = antialias
+
+ self.register_buffer('mean', torch.Tensor([0.48145466, 0.4578275, 0.40821073]), persistent=False)
+ self.register_buffer('std', torch.Tensor([0.26862954, 0.26130258, 0.27577711]), persistent=False)
+
+ def preprocess(self, x):
+ # normalize to [0,1]
+ x = kornia.geometry.resize(x, (224, 224),
+ interpolation='bicubic',align_corners=True,
+ antialias=self.antialias)
+ x = (x + 1.) / 2.
+ # renormalize according to clip
+ x = kornia.enhance.normalize(x, self.mean, self.std)
+ return x
+
+ def forward(self, x):
+ # x is assumed to be in range [-1,1]
+ return self.model.encode_image(self.preprocess(x))
+
+
+if __name__ == "__main__":
+ from ldm.util import count_params
+ model = FrozenCLIPEmbedder()
+ count_params(model, verbose=True) \ No newline at end of file
diff --git a/ldm/modules/encoders/xlmr.py b/ldm/modules/encoders/xlmr.py
new file mode 100644
index 00000000..beab3fdf
--- /dev/null
+++ b/ldm/modules/encoders/xlmr.py
@@ -0,0 +1,137 @@
+from transformers import BertPreTrainedModel,BertModel,BertConfig
+import torch.nn as nn
+import torch
+from transformers.models.xlm_roberta.configuration_xlm_roberta import XLMRobertaConfig
+from transformers import XLMRobertaModel,XLMRobertaTokenizer
+from typing import Optional
+
+class BertSeriesConfig(BertConfig):
+ def __init__(self, vocab_size=30522, hidden_size=768, num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, hidden_act="gelu", hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, max_position_embeddings=512, type_vocab_size=2, initializer_range=0.02, layer_norm_eps=1e-12, pad_token_id=0, position_embedding_type="absolute", use_cache=True, classifier_dropout=None,project_dim=512, pooler_fn="average",learn_encoder=False,model_type='bert',**kwargs):
+
+ super().__init__(vocab_size, hidden_size, num_hidden_layers, num_attention_heads, intermediate_size, hidden_act, hidden_dropout_prob, attention_probs_dropout_prob, max_position_embeddings, type_vocab_size, initializer_range, layer_norm_eps, pad_token_id, position_embedding_type, use_cache, classifier_dropout, **kwargs)
+ self.project_dim = project_dim
+ self.pooler_fn = pooler_fn
+ self.learn_encoder = learn_encoder
+
+class RobertaSeriesConfig(XLMRobertaConfig):
+ def __init__(self, pad_token_id=1, bos_token_id=0, eos_token_id=2,project_dim=512,pooler_fn='cls',learn_encoder=False, **kwargs):
+ super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
+ self.project_dim = project_dim
+ self.pooler_fn = pooler_fn
+ self.learn_encoder = learn_encoder
+
+
+class BertSeriesModelWithTransformation(BertPreTrainedModel):
+
+ _keys_to_ignore_on_load_unexpected = [r"pooler"]
+ _keys_to_ignore_on_load_missing = [r"position_ids", r"predictions.decoder.bias"]
+ config_class = BertSeriesConfig
+
+ def __init__(self, config=None, **kargs):
+ # modify initialization for autoloading
+ if config is None:
+ config = XLMRobertaConfig()
+ config.attention_probs_dropout_prob= 0.1
+ config.bos_token_id=0
+ config.eos_token_id=2
+ config.hidden_act='gelu'
+ config.hidden_dropout_prob=0.1
+ config.hidden_size=1024
+ config.initializer_range=0.02
+ config.intermediate_size=4096
+ config.layer_norm_eps=1e-05
+ config.max_position_embeddings=514
+
+ config.num_attention_heads=16
+ config.num_hidden_layers=24
+ config.output_past=True
+ config.pad_token_id=1
+ config.position_embedding_type= "absolute"
+
+ config.type_vocab_size= 1
+ config.use_cache=True
+ config.vocab_size= 250002
+ config.project_dim = 768
+ config.learn_encoder = False
+ super().__init__(config)
+ self.roberta = XLMRobertaModel(config)
+ self.transformation = nn.Linear(config.hidden_size,config.project_dim)
+ self.pre_LN=nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.tokenizer = XLMRobertaTokenizer.from_pretrained('xlm-roberta-large')
+ self.pooler = lambda x: x[:,0]
+ self.post_init()
+
+ def encode(self,c):
+ device = next(self.parameters()).device
+ text = self.tokenizer(c,
+ truncation=True,
+ max_length=77,
+ return_length=False,
+ return_overflowing_tokens=False,
+ padding="max_length",
+ return_tensors="pt")
+ text["input_ids"] = torch.tensor(text["input_ids"]).to(device)
+ text["attention_mask"] = torch.tensor(
+ text['attention_mask']).to(device)
+ features = self(**text)
+ return features['projection_state']
+
+ def forward(
+ self,
+ input_ids: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ token_type_ids: Optional[torch.Tensor] = None,
+ position_ids: Optional[torch.Tensor] = None,
+ head_mask: Optional[torch.Tensor] = None,
+ inputs_embeds: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ output_attentions: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ ) :
+ r"""
+ """
+
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+
+ outputs = self.roberta(
+ input_ids=input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ output_attentions=output_attentions,
+ output_hidden_states=True,
+ return_dict=return_dict,
+ )
+
+ # last module outputs
+ sequence_output = outputs[0]
+
+
+ # project every module
+ sequence_output_ln = self.pre_LN(sequence_output)
+
+ # pooler
+ pooler_output = self.pooler(sequence_output_ln)
+ pooler_output = self.transformation(pooler_output)
+ projection_state = self.transformation(outputs.last_hidden_state)
+
+ return {
+ 'pooler_output':pooler_output,
+ 'last_hidden_state':outputs.last_hidden_state,
+ 'hidden_states':outputs.hidden_states,
+ 'attentions':outputs.attentions,
+ 'projection_state':projection_state,
+ 'sequence_out': sequence_output
+ }
+
+
+class RobertaSeriesModelWithTransformation(BertSeriesModelWithTransformation):
+ base_model_prefix = 'roberta'
+ config_class= RobertaSeriesConfig \ No newline at end of file
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-29 00:47:26 +0000