diff --git a/Dockerfile b/Dockerfile index 4604199..de613aa 100644 --- a/Dockerfile +++ b/Dockerfile @@ -7,7 +7,7 @@ RUN --mount=type=cache,target=/root/.cache/pip pip install --upgrade pip WORKDIR /app RUN git clone https://github.com/TIGER-AI-Lab/Mantis.git --single-branch /app/Mantis RUN git clone https://github.com/togethercomputer/Dragonfly --single-branch /app/Dragonfly -RUN git clone https://github.com/baaivision/Emu3 --single-branch /app/emu3 +RUN git clone https://github.com/baaivision/Emu3 --single-branch /app/Emu3 COPY requirements.txt . ARG VERSION=latest diff --git a/README.md b/README.md index 0c863d5..6477151 100644 --- a/README.md +++ b/README.md @@ -31,6 +31,7 @@ Can't decide which to use? See the [OpenVLM Leaderboard](https://huggingface.co/ - - [X] [BAAI/Bunny-v1_1-Llama-3-8B-V](https://huggingface.co/BAAI/Bunny-v1_1-Llama-3-8B-V) - - [X] [Bunny-Llama-3-8B-V](https://huggingface.co/BAAI/Bunny-Llama-3-8B-V) - - [X] [Emu2-Chat](https://huggingface.co/BAAI/Emu2-Chat) (may need the --max-memory option to GPU split, slow to load) +- - [X] [Emu3-Chat](https://huggingface.co/BAAI/Emu3-Chat) - [X] [cognitivecomputations](https://huggingface.co/cognitivecomputations) - - [X] [dolphin-vision-72b](https://huggingface.co/cognitivecomputations/dolphin-vision-72b) (alternate docker only) - - [X] [dolphin-vision-7b](https://huggingface.co/cognitivecomputations/dolphin-vision-7b) (alternate docker only) @@ -40,7 +41,9 @@ Can't decide which to use? See the [OpenVLM Leaderboard](https://huggingface.co/ - [X] [failspy](https://huggingface.co/failspy) - - [X] [Phi-3-vision-128k-instruct-abliterated-alpha](https://huggingface.co/failspy/Phi-3-vision-128k-instruct-abliterated-alpha) - [X] [falcon-11B-vlm](https://huggingface.co/tiiuae/falcon-11B-vlm) (alternate docker only) -- [X] [fancyfeast/joy-caption-pre-alpha](https://huggingface.co/spaces/fancyfeast/joy-caption-pre-alpha) (caption only) +- [X] [fancyfeast](https://huggingface.co/fancyfeast) +- - [X] [joy-caption-alpha-two](https://huggingface.co/spaces/fancyfeast/joy-caption-alpha-two) (with experimental multi-image support) +- - [X] [joy-caption-pre-alpha](https://huggingface.co/spaces/fancyfeast/joy-caption-pre-alpha) (caption only) - [X] [fuyu-8b](https://huggingface.co/adept/fuyu-8b) [pretrain] - [X] [HuggingFaceM4/idefics2](https://huggingface.co/HuggingFaceM4) - - [X] [idefics2-8b](https://huggingface.co/HuggingFaceM4/idefics2-8b) (wont gpu split, alternate docker only) @@ -154,6 +157,11 @@ If you can't find your favorite model, you can [open a new issue](https://github ## Recent updates +Version 0.36.0 + +- new model support: BAAI/Emu3-Chat +- Experimental support for fancyfeast/joy-caption-alpha-two with multiple images (see: [backend](backend/joy-caption-latest.py) for more details) + Version 0.35.0 - Update Molmo (tensorflow-cpu no longer required), and add autocast for faster, smaller types than float32. diff --git a/backend/emu3.py b/backend/emu3.py index 028dbaf..3596178 100644 --- a/backend/emu3.py +++ b/backend/emu3.py @@ -1,26 +1,26 @@ from transformers import AutoTokenizer, AutoModel, AutoImageProcessor, AutoModelForCausalLM from transformers.generation.configuration_utils import GenerationConfig +from Emu3.emu3.mllm.processing_emu3 import Emu3Processor from vision_qna import * -# WIP - -# BAAI/Emu3-Gen # BAAI/Emu3-Chat VQ_HUB = "BAAI/Emu3-VisionTokenizer" class VisionQnA(VisionQnABase): model_name: str = "emu3" - format: str = "emu3" + format: str = "vicuna" visual_layers: List[str] = [] def __init__(self, model_id: str, device: str, device_map: str = 'auto', extra_params = {}, format = None): super().__init__(model_id, device, device_map, extra_params, format) - - self.processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=self.params.get('trust_remote_code', False)) - self.model = AutoModel.from_pretrained(**self.params).eval() + self.model = AutoModelForCausalLM.from_pretrained(**self.params).eval() + tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=self.params.get('trust_remote_code', False)) + image_processor = AutoImageProcessor.from_pretrained(VQ_HUB, trust_remote_code=True) + image_tokenizer = AutoModel.from_pretrained(VQ_HUB, device_map=self.params['device_map'], trust_remote_code=self.params.get('trust_remote_code', False)).eval() + self.processor = Emu3Processor(image_processor, image_tokenizer, tokenizer) # bitsandbytes already moves the model to the device, so we don't need to do it again. if not (extra_params.get('load_in_4bit', False) or extra_params.get('load_in_8bit', False)): @@ -29,21 +29,32 @@ def __init__(self, model_id: str, device: str, device_map: str = 'auto', extra_p self.loaded_banner() async def stream_chat_with_images(self, request: ImageChatRequest) -> AsyncGenerator[str, None]: - images, prompt = await prompt_from_messages(request.messages, self.format) + image = None + text = '' + + for m in request.messages: + if m.role == 'user': + for c in m.content: + if c.type == 'image_url': + image = await url_to_image(c.image_url.url) + break + + text = "".join([t.text for t in request.messages[-1].content if t.text]) - inputs = self.processor(images=images, text=prompt, return_tensors="pt").to(self.model.device) + inputs = self.processor(text=text, image=image, mode='U', padding_side="left", padding="longest", return_tensors="pt") - default_params = { - 'do_sample': False, -# 'eos_token_id': self.processor.tokenizer.eos_token_id, -# 'pad_token_id': self.processor.tokenizer.eos_token_id, - } + default_params = dict( + max_new_tokens=320, + pad_token_id=self.processor.tokenizer.pad_token_id, + bos_token_id=self.processor.tokenizer.bos_token_id, + eos_token_id=self.processor.tokenizer.eos_token_id, + ) params = self.get_generation_params(request, default_params=default_params) generation_kwargs = dict( - **inputs, - **params, + input_ids=inputs.input_ids.to(self.device), + generation_config=GenerationConfig(**params), ) for new_text in threaded_streaming_generator(generate=self.model.generate, tokenizer=self.processor.tokenizer, generation_kwargs=generation_kwargs): @@ -53,22 +64,3 @@ async def stream_chat_with_images(self, request: ImageChatRequest) -> AsyncGener else: yield new_text[:end] break - - async def chat_with_images(self, request: ImageChatRequest) -> str: - images, prompt = await prompt_from_messages(request.messages, self.format) - - inputs = self.processor(images=images, text=prompt, return_tensors="pt").to(self.model.device) - - default_params = { - 'do_sample': False, -# 'eos_token_id': self.processor.tokenizer.eos_token_id, -# 'pad_token_id': self.processor.tokenizer.eos_token_id, - } - - params = self.get_generation_params(request, default_params=default_params) - - output = self.model.generate(**inputs, **params) - response = self.processor.tokenizer.decode(output[0][inputs.input_ids.size(1):].cpu(), skip_special_tokens=True) - - return response - diff --git a/backend/joy-caption-latest.py b/backend/joy-caption-latest.py new file mode 100644 index 0000000..a1b1150 --- /dev/null +++ b/backend/joy-caption-latest.py @@ -0,0 +1,239 @@ +from PIL import Image +from huggingface_hub import snapshot_download +from pathlib import Path +from transformers import AutoModel, AutoProcessor, AutoTokenizer, AutoModelForCausalLM +import torch +import torch.nn as nn +import torch.amp.autocast_mode +import torchvision.transforms.functional as TVF + +from vision_qna import * + +# https://huggingface.co/spaces/fancyfeast/joy-caption-alpha-two + +# fancyfeast/joy-caption-alpha-two + +LATEST="cgrkzexw-599808/" +LATEST_NAME="JoyCaption Alpha Two (2024-09-26a)" +CLIP_PATH = "google/siglip-so400m-patch14-384" + +# This is the expected conversation format, but others work too. +SYSTEM_MSG="You are a helpful image captioner." +CAPTION_TYPE_MAP = { + "Descriptive": [ + "Write a descriptive caption for this image in a formal tone.", + "Write a descriptive caption for this image in a formal tone within {word_count} words.", + "Write a {length} descriptive caption for this image in a formal tone.", + ], + "Descriptive (Informal)": [ + "Write a descriptive caption for this image in a casual tone.", + "Write a descriptive caption for this image in a casual tone within {word_count} words.", + "Write a {length} descriptive caption for this image in a casual tone.", + ], + "Training Prompt": [ + "Write a stable diffusion prompt for this image.", + "Write a stable diffusion prompt for this image within {word_count} words.", + "Write a {length} stable diffusion prompt for this image.", + ], + "MidJourney": [ + "Write a MidJourney prompt for this image.", + "Write a MidJourney prompt for this image within {word_count} words.", + "Write a {length} MidJourney prompt for this image.", + ], + "Booru tag list": [ + "Write a list of Booru tags for this image.", + "Write a list of Booru tags for this image within {word_count} words.", + "Write a {length} list of Booru tags for this image.", + ], + "Booru-like tag list": [ + "Write a list of Booru-like tags for this image.", + "Write a list of Booru-like tags for this image within {word_count} words.", + "Write a {length} list of Booru-like tags for this image.", + ], + "Art Critic": [ + "Analyze this image like an art critic would with information about its composition, style, symbolism, the use of color, light, any artistic movement it might belong to, etc.", + "Analyze this image like an art critic would with information about its composition, style, symbolism, the use of color, light, any artistic movement it might belong to, etc. Keep it within {word_count} words.", + "Analyze this image like an art critic would with information about its composition, style, symbolism, the use of color, light, any artistic movement it might belong to, etc. Keep it {length}.", + ], + "Product Listing": [ + "Write a caption for this image as though it were a product listing.", + "Write a caption for this image as though it were a product listing. Keep it under {word_count} words.", + "Write a {length} caption for this image as though it were a product listing.", + ], + "Social Media Post": [ + "Write a caption for this image as if it were being used for a social media post.", + "Write a caption for this image as if it were being used for a social media post. Limit the caption to {word_count} words.", + "Write a {length} caption for this image as if it were being used for a social media post.", + ], +} +extra_options=[ + "If there is a person/character in the image you must refer to them as {name}.", + "Do NOT include information about people/characters that cannot be changed (like ethnicity, gender, etc), but do still include changeable attributes (like hair style).", + "Include information about lighting.", + "Include information about camera angle.", + "Include information about whether there is a watermark or not.", + "Include information about whether there are JPEG artifacts or not.", + "If it is a photo you MUST include information about what camera was likely used and details such as aperture, shutter speed, ISO, etc.", + "Do NOT include anything sexual; keep it PG.", + "Do NOT mention the image's resolution.", + "You MUST include information about the subjective aesthetic quality of the image from low to very high.", + "Include information on the image's composition style, such as leading lines, rule of thirds, or symmetry.", + "Do NOT mention any text that is in the image.", + "Specify the depth of field and whether the background is in focus or blurred.", + "If applicable, mention the likely use of artificial or natural lighting sources.", + "Do NOT use any ambiguous language.", + "Include whether the image is sfw, suggestive, or nsfw.", + "ONLY describe the most important elements of the image." +] + +class ImageAdapter(nn.Module): + def __init__(self, input_features: int, output_features: int, ln1: bool, pos_emb: bool, num_image_tokens: int, deep_extract: bool): + super().__init__() + self.deep_extract = deep_extract + + if self.deep_extract: + input_features = input_features * 5 + + self.linear1 = nn.Linear(input_features, output_features) + self.activation = nn.GELU() + self.linear2 = nn.Linear(output_features, output_features) + self.ln1 = nn.Identity() if not ln1 else nn.LayerNorm(input_features) + self.pos_emb = None if not pos_emb else nn.Parameter(torch.zeros(num_image_tokens, input_features)) + + # Other tokens (<|image_start|>, <|image_end|>, <|eot_id|>) + self.other_tokens = nn.Embedding(3, output_features) + self.other_tokens.weight.data.normal_(mean=0.0, std=0.02) # Matches HF's implementation of llama3 + + def forward(self, vision_outputs: torch.Tensor): + if self.deep_extract: + x = torch.concat(( + vision_outputs[-2], + vision_outputs[3], + vision_outputs[7], + vision_outputs[13], + vision_outputs[20], + ), dim=-1) + assert len(x.shape) == 3, f"Expected 3, got {len(x.shape)}" # batch, tokens, features + assert x.shape[-1] == vision_outputs[-2].shape[-1] * 5, f"Expected {vision_outputs[-2].shape[-1] * 5}, got {x.shape[-1]}" + else: + x = vision_outputs[-2] + + x = self.ln1(x) + + if self.pos_emb is not None: + assert x.shape[-2:] == self.pos_emb.shape, f"Expected {self.pos_emb.shape}, got {x.shape[-2:]}" + x = x + self.pos_emb + + x = self.linear1(x) + x = self.activation(x) + x = self.linear2(x) + + # <|image_start|>, IMAGE, <|image_end|> + other_tokens = self.other_tokens(torch.tensor([0, 1], device=self.other_tokens.weight.device).expand(x.shape[0], -1)) + assert other_tokens.shape == (x.shape[0], 2, x.shape[2]), f"Expected {(x.shape[0], 2, x.shape[2])}, got {other_tokens.shape}" + x = torch.cat((other_tokens[:, 0:1], x, other_tokens[:, 1:2]), dim=1) + + return x + + def get_eot_embedding(self): + return self.other_tokens(torch.tensor([2], device=self.other_tokens.weight.device)).squeeze(0) + + +class VisionQnA(VisionQnABase): + model_name: str = "joy-caption-alpha-two" + format: str = "llama3" + visual_layers: List[str] = [] + + def __init__(self, model_id: str, device: str, device_map: str = 'auto', extra_params = {}, format = None): + super().__init__(model_id, device, device_map, extra_params, format) + + self.clip_processor = AutoProcessor.from_pretrained(CLIP_PATH) + self.clip_model = AutoModel.from_pretrained(CLIP_PATH).vision_model + + CHECKPOINT_PATH = Path(snapshot_download(repo_id="fancyfeast/joy-caption-alpha-two", repo_type="space", allow_patterns=LATEST)) / LATEST + checkpoint = torch.load(CHECKPOINT_PATH / "clip_model.pt", map_location='cpu', weights_only=True) + checkpoint = {k.replace("_orig_mod.module.", ""): v for k, v in checkpoint.items()} + self.clip_model.load_state_dict(checkpoint) + del checkpoint + + self.clip_model.eval() + self.clip_model.requires_grad_(False) + self.clip_model.to(self.device) + + self.tokenizer = AutoTokenizer.from_pretrained(CHECKPOINT_PATH / "text_model", use_fast=True) + + del self.params['pretrained_model_name_or_path'] + self.model = AutoModelForCausalLM.from_pretrained(CHECKPOINT_PATH / "text_model", **self.params) + self.model.eval() + + self.image_adapter = ImageAdapter(self.clip_model.config.hidden_size, self.model.config.hidden_size, False, False, 38, False) + self.image_adapter.load_state_dict(torch.load(CHECKPOINT_PATH / "image_adapter.pt", map_location="cpu", weights_only=True)) + self.image_adapter.eval() + self.image_adapter.to(self.device) + + self.loaded_banner() + + async def stream_chat_with_images(self, request: ImageChatRequest) -> AsyncGenerator[str, None]: + if request.messages[0].role != 'system': + request.messages = [Message(role='system', content=[Content(type='text', text=SYSTEM_MSG)])] + request.messages + images, prompt = await prompt_from_messages(request.messages, self.format) + + def to_pixel_values(img): + image = img.resize((384, 384), Image.LANCZOS) + pixel_values = TVF.pil_to_tensor(image).unsqueeze(0) / 255.0 + pixel_values = TVF.normalize(pixel_values, [0.5], [0.5]) + return pixel_values.to(self.device) + + def to_image_embed(img): + pixel_values = to_pixel_values(img) + + with torch.amp.autocast_mode.autocast('cuda', enabled=True): + vision_outputs = self.clip_model(pixel_values=pixel_values, output_hidden_states=True) + embedded_images = self.image_adapter(vision_outputs.hidden_states) + return embedded_images.to(device=self.device, dtype=self.model.dtype) + + if images: + image_embeds = [ to_image_embed(img) for img in images ] + else: + image_embeds = [ await url_to_image(black_pixel_url) ] + + split_tokens = [ self.tokenizer.encode(c, return_tensors="pt", add_special_tokens=False, truncation=False) for c in prompt.split('') ] + split_embeds = [ self.model.model.embed_tokens(tok_ids.to(self.device)) for tok_ids in split_tokens ] + + input_ids = [split_tokens[0]] + inputs_embeds = [ split_embeds[0] ] + + for im, tok, emb in zip(image_embeds, split_tokens[1:], split_embeds[1:]): + input_ids.extend([torch.zeros((1, im.shape[1]), dtype=torch.long), tok]) + inputs_embeds.extend([ im, emb ]) + + input_ids = torch.cat(input_ids, dim=1).to(self.device) + attention_mask = torch.ones_like(input_ids).to(self.device) + inputs_embeds = torch.cat(inputs_embeds, dim=1).to(self.device) + + inputs = dict( + input_ids=input_ids, + inputs_embeds=inputs_embeds, + attention_mask=attention_mask, + ) + + default_params = dict( + max_new_tokens=512, + do_sample=True, + suppress_tokens=None, + ) + + params = self.get_generation_params(request, default_params=default_params) + + generation_kwargs = dict( + **inputs, + **params, + ) + + for new_text in threaded_streaming_generator(generate=self.model.generate, tokenizer=self.tokenizer, generation_kwargs=generation_kwargs): + end = new_text.find(self.tokenizer.eos_token) + if end == -1: + yield new_text + else: + yield new_text[:end] + break diff --git a/backend/molmo.py b/backend/molmo.py index 68aa3af..94cc4a2 100644 --- a/backend/molmo.py +++ b/backend/molmo.py @@ -2,10 +2,11 @@ from vision_qna import * -# allenai/MolmoE-1B-0924 -# allenai/Molmo-7B-D-0924 +# allenai/MolmoE-1B-0924 XXX problems with performance and RAM usage +# allenai/Molmo-7B-D-0924 # faster # allenai/Molmo-7B-O-0924 # allenai/Molmo-72B-0924 +# SeanScripts/Molmo-72B-0924-nf4 # cyan2k/molmo-7B-D-bnb-4bit XXX needs tensorflow-cpu # cyan2k/molmo-7B-O-bnb-4bit XXX needs tensorflow-cpu diff --git a/model_conf_tests.json b/model_conf_tests.json index 3f2e469..a3b59cf 100644 --- a/model_conf_tests.json +++ b/model_conf_tests.json @@ -12,6 +12,8 @@ ["BAAI/Bunny-v1_1-4B"], ["BAAI/Emu2-Chat", "--load-in-4bit"], ["BAAI/Emu2-Chat", "--max-memory=0:78GiB,1:20GiB"], + ["BAAI/Emu3-Chat", "--load-in-4bit", "-A", "flash_attention_2"], + ["BAAI/Emu3-Chat", "-A", "flash_attention_2"], ["OpenGVLab/InternVL-Chat-V1-5", "--device-map", "cuda:0", "--load-in-4bit"], ["OpenGVLab/InternVL-Chat-V1-5", "--device-map", "cuda:0", "--max-tiles", "40", "--load-in-4bit"], ["OpenGVLab/InternVL-Chat-V1-5", "--device-map", "cuda:0", "--max-tiles", "40"], @@ -61,6 +63,8 @@ ["echo840/Monkey-Chat"], ["failspy/Phi-3-vision-128k-instruct-abliterated-alpha", "-A", "flash_attention_2", "--load-in-4bit"], ["failspy/Phi-3-vision-128k-instruct-abliterated-alpha", "-A", "flash_attention_2"], + ["fancyfeast/joy-caption-alpha-two", "--load-in-4bit", "-A", "flash_attention_2"], + ["fancyfeast/joy-caption-alpha-two", "-A", "flash_attention_2"], ["fancyfeast/joy-caption-pre-alpha", "--load-in-4bit", "-A", "flash_attention_2"], ["fancyfeast/joy-caption-pre-alpha", "-A", "flash_attention_2"], ["internlm/internlm-xcomposer2d5-7b", "-A", "flash_attention_2", "--device-map", "cuda:0", "--load-in-4bit"], diff --git a/vision_qna.py b/vision_qna.py index 4d8bf60..1bbbce4 100644 --- a/vision_qna.py +++ b/vision_qna.py @@ -103,7 +103,7 @@ def __init__(self, model_id: str, device: str, device_map: str = 'auto', extra_p self.params.update({"trust_remote_code": True }) if format: - self.format = format + self.format = format torch.set_grad_enabled(False) @@ -164,7 +164,7 @@ def get_generation_params(self, request: ImageChatRequest, default_params = {}) return params def threaded_streaming_generator(generate, tokenizer, generation_kwargs): - streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=True, skip_prompt=True, timeout=60) + streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=True, skip_prompt=True, timeout=600) generation_kwargs['streamer'] = streamer @@ -963,6 +963,9 @@ def guess_backend(model_name: str) -> str: if 'mantis' in model_id: return 'mantis' + if 'emu3' in model_id: + return 'emu3' + if 'emu' in model_id: return 'emu' @@ -985,7 +988,10 @@ def guess_backend(model_name: str) -> str: if 'dolphin-vision' in model_id: return 'dv-qwen' - if 'fancyfeast/joy-caption-pre-alpha' in model_id: + if 'joy-caption-alpha-two' in model_id: + return 'joy-caption-latest' + + if 'joy-caption-pre-alpha' in model_id: return 'joy-caption-pre-alpha' if 'hf-internal-testing/pixtral-12b' in model_id: