Multimodal-RAG / conversation.py
Tile's picture
first commit
4f234f9
raw
history blame
10.2 kB
import dataclasses
from enum import auto, Enum
from typing import List, Tuple
import os
class SeparatorStyle(Enum):
"""Different separator style."""
SINGLE = auto()
TWO = auto()
MPT = auto()
PLAIN = auto()
LLAMA_2 = auto()
MISTRAL = auto()
# video_helper_map = {
# # 'Chips Making Deal Video' : {'path' : '/data/videos/ChipmakingDeal/sub-videos/', 'prefix' : 'ChipmakingDeal_split'},
# 'Keynote 2023' : {'path' : '/data/videos/PatsKeynote23/sub-videos/', 'prefix' : 'keynotes23_split'},
# 'Intel Behind the Bell' : {'path' : '/data/videos/BehindTheBell/sub-videos/', 'prefix' : 'Behind the Bell Intel_split'},
# 'CEOs Talk' : {'path' : '/data/videos/SamPatTalkAI/sub-videos/', 'prefix' : 'Sam Altman and Pat Gelsinger Talk Artificial Intelligence_split'},
# 'Chips Act Funding Announcement' : {'path' : '/data/videos/IntelChipsFundingAnnounce/sub-videos/', 'prefix' : 'Intel Celebrates CHIPS and Science Act Direct Funding Announcement (Replay)_split'},
# '22nm-Chip Technology' : {'path' : '/data/videos/MarkBohrExplains22nm/sub-videos/', 'prefix' : 'Video Animation Mark Bohr Gets Small 22nm Explained Intel_split'},
# '14nm-Chip Technology' : {'path' : '/data/videos/MarkBohrExplains14nm/sub-videos/', 'prefix' : 'Explanation of Intels 14nm Process_split'},
# }
video_helper_map = {
# 'Chips Making Deal Video' : {'path' : '/data/videos/ChipmakingDeal/sub-videos/', 'prefix' : 'ChipmakingDeal_split'},
'Innovation-2023' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/PatsKeynote23/sub-videos/', 'prefix' : 'keynotes23_split'},
'Behind-the-Bell-Intel' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/BehindTheBell/sub-videos/', 'prefix' : 'Behind the Bell Intel_split'},
'Foundry-Connect' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/SamPatTalkAI/sub-videos/', 'prefix' : 'Sam Altman and Pat Gelsinger Talk Artificial Intelligence_split'},
'Chips Act Funding Announcement' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/IntelChipsFundingAnnounce/sub-videos/', 'prefix' : 'Intel Celebrates CHIPS and Science Act Direct Funding Announcement (Replay)_split'},
'22nm-transistor-animation' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/MarkBohrExplains22nm/sub-videos/', 'prefix' : 'Video Animation Mark Bohr Gets Small 22nm Explained Intel_split'},
'14nm-transistor-animation' : {'path' : '/data1/tile_gh/Multimodal-RAG/videos/MarkBohrExplains14nm/sub-videos/', 'prefix' : 'Explanation of Intels 14nm Process_split'},
}
@dataclasses.dataclass
class Conversation:
"""A class that keeps all conversation history."""
system: str
roles: List[str]
messages: List[List[str]]
offset: int
sep_style: SeparatorStyle = SeparatorStyle.SINGLE
sep: str = "\n"
sep2: str = None
version: str = "Unknown"
path_to_img: str = None
video_title: str = None
caption: str = None
skip_next: bool = False
def _template_caption(self):
out = ""
if self.caption is not None:
out = f"The caption associated with the image is '{self.caption}'. "
return out
def get_prompt(self):
messages = self.messages
if len(messages) > 0 and messages[1][1] is not None and "<image>" not in messages[0][1]:
# if there is a history message and <image> is not yet in the first message of user
# then add <image>\n to the beginning
messages = self.messages.copy()
init_role, init_msg = messages[0].copy()
messages[0] = (init_role, "<image>\n" + self._template_caption() + init_msg)
if len(messages) > 1 and messages[1][1] is None:
#Need to do RAG. prompt is the query only
ret = messages[0][1]
else:
if self.sep_style == SeparatorStyle.SINGLE:
ret = ""
for role, message in messages:
if message:
ret += role + ": " + message + self.sep
else:
ret += role + ":"
elif self.sep_style == SeparatorStyle.LLAMA_2:
wrap_sys = lambda msg: f"<<SYS>>\n{msg}\n<</SYS>>\n\n" if len(msg) > 0 else msg
wrap_inst = lambda msg: f"[INST] {msg} [/INST]"
ret = ""
for i, (role, message) in enumerate(messages):
if i == 0:
assert message, "first message should not be none"
assert role == self.roles[0], "first message should come from user"
if message:
if type(message) is tuple:
message, _, _ = message
if i == 0: message = wrap_sys(self.system) + message
if i % 2 == 0:
message = wrap_inst(message)
ret += self.sep + message
else:
ret += " " + message + " " + self.sep2
else:
ret += ""
ret = ret.lstrip(self.sep)
else:
raise ValueError(f"Invalid style: {self.sep_style}")
return ret
def append_message(self, role, message):
self.messages.append([role, message])
def get_images(self, return_pil=False):
images = []
if self.path_to_img is not None:
path_to_image = self.path_to_img
images.append(path_to_image)
# import base64
# from io import BytesIO
# from PIL import Image
# image = Image.open(path_to_image)
# max_hw, min_hw = max(image.size), min(image.size)
# aspect_ratio = max_hw / min_hw
# max_len, min_len = 800, 400
# shortest_edge = int(min(max_len / aspect_ratio, min_len, min_hw))
# longest_edge = int(shortest_edge * aspect_ratio)
# W, H = image.size
# if longest_edge != max(image.size):
# if H > W:
# H, W = longest_edge, shortest_edge
# else:
# H, W = shortest_edge, longest_edge
# image = image.resize((W, H))
# if return_pil:
# images.append(image)
# else:
# # buffered = BytesIO()
# # # image.save(buffered, format="PNG")
# # img_b64_str = base64.b64encode(buffered.getvalue()).decode()
# images.append(path_to_image)
return images
def to_gradio_chatbot(self):
ret = []
for i, (role, msg) in enumerate(self.messages[self.offset:]):
if i % 2 == 0:
if type(msg) is tuple:
import base64
from io import BytesIO
msg, image, image_process_mode = msg
max_hw, min_hw = max(image.size), min(image.size)
aspect_ratio = max_hw / min_hw
max_len, min_len = 800, 400
shortest_edge = int(min(max_len / aspect_ratio, min_len, min_hw))
longest_edge = int(shortest_edge * aspect_ratio)
W, H = image.size
if H > W:
H, W = longest_edge, shortest_edge
else:
H, W = shortest_edge, longest_edge
image = image.resize((W, H))
buffered = BytesIO()
image.save(buffered, format="JPEG")
img_b64_str = base64.b64encode(buffered.getvalue()).decode()
img_str = f'<img src="data:image/png;base64,{img_b64_str}" alt="user upload image" />'
msg = img_str + msg.replace('<image>', '').strip()
ret.append([msg, None])
else:
ret.append([msg, None])
else:
ret[-1][-1] = msg
return ret
def copy(self):
return Conversation(
system=self.system,
roles=self.roles,
messages=[[x, y] for x, y in self.messages],
offset=self.offset,
sep_style=self.sep_style,
sep=self.sep,
sep2=self.sep2,
version=self.version,)
def dict(self):
return {
"system": self.system,
"roles": self.roles,
"messages": self.messages,
"offset": self.offset,
"sep": self.sep,
"sep2": self.sep2,
"path_to_img": self.path_to_img,
"video_title" : self.video_title,
"caption" : self.caption,
}
def get_path_to_subvideos(self):
print(f"self.video_title {self.video_title}")
print(f"self.path_to_image {self.path_to_img}")
return None
if self.video_title is not None and self.path_to_img is not None:
info = video_helper_map[self.video_title]
path = info['path']
prefix = info['prefix']
vid_index = self.path_to_img.split('/')[-1]
vid_index = vid_index.split('_')[-1]
vid_index = vid_index.replace('.jpg', '')
ret = f"{prefix}{vid_index}.mp4"
ret = os.path.join(path, ret)
return ret
elif self.path_to_img is not None:
return self.path_to_img
return None
multimodal_rag = Conversation(
system="",
roles=("USER", "ASSISTANT"),
messages=(),
offset=0,
sep_style=SeparatorStyle.SINGLE,
sep="\n",
path_to_img=None,
video_title=None,
caption=None,
)
conv_mistral_instruct = Conversation(
system="",
roles=("USER", "ASSISTANT"),
version="llama_v2",
messages=(),
offset=0,
sep_style=SeparatorStyle.LLAMA_2,
sep="",
sep2="</s>",
path_to_img=None,
video_title=None,
caption=None,
)
default_conversation = multimodal_rag
conv_templates = {
"default": multimodal_rag,
"multimodal_rag" : multimodal_rag,
"llavamed_rag" : conv_mistral_instruct,
}
if __name__ == "__main__":
print(default_conversation.get_prompt())