Spaces:
Sleeping
Sleeping
File size: 9,707 Bytes
e6b4540 7baf011 80a5f9c e6b4540 80a5f9c e6b4540 85d6203 e6143be 85d6203 e6143be 85d6203 e6143be e6b4540 e80434f 2c66a65 46dbca8 2c66a65 46dbca8 e80434f e6b4540 46dbca8 85d6203 e6b4540 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 |
import os
import time
import uuid
from typing import List, Tuple, Optional, Dict, Union
import google.generativeai as genai
import gradio as gr
from PIL import Image
print("google-generativeai:", genai.__version__)
GOOGLE_API_KEY = os.environ.get("GOOGLE_API_KEY")
TITLE = """<h1 align="center">Gemini Movie Editor Assitant Agent</h1>"""
SUBTITLE = """<h2 align="center">Movie Editing Agent built with Gemini Pro and Gemini Pro Vision API</h2>"""
GETKEY = """
<div style="text-align: center; display: flex; justify-content: center; align-items: center;">
<span>Get an API key
<a href="https://makersuite.google.com/app/apikey">GOOGLE API KEY</a>.
</span>
</div>
"""
AVATAR_IMAGES = (
None,
"https://media.roboflow.com/spaces/gemini-icon.png"
)
IMAGE_CACHE_DIRECTORY = "/tmp"
IMAGE_WIDTH = 512
CHAT_HISTORY = List[Tuple[Optional[Union[Tuple[str], str]], Optional[str]]]
def preprocess_stop_sequences(stop_sequences: str) -> Optional[List[str]]:
if not stop_sequences:
return None
return [sequence.strip() for sequence in stop_sequences.split(",")]
def preprocess_image(image: Image.Image) -> Optional[Image.Image]:
image_height = int(image.height * IMAGE_WIDTH / image.width)
return image.resize((IMAGE_WIDTH, image_height))
def cache_pil_image(image: Image.Image) -> str:
image_filename = f"{uuid.uuid4()}.jpeg"
os.makedirs(IMAGE_CACHE_DIRECTORY, exist_ok=True)
image_path = os.path.join(IMAGE_CACHE_DIRECTORY, image_filename)
image.save(image_path, "JPEG")
return image_path
def preprocess_chat_history(
history: CHAT_HISTORY
) -> List[Dict[str, Union[str, List[str]]]]:
messages = []
for user_message, model_message in history:
if isinstance(user_message, tuple):
pass
elif user_message is not None:
messages.append({'role': 'user', 'parts': [user_message]})
if model_message is not None:
messages.append({'role': 'model', 'parts': [model_message]})
return messages
def upload(files: Optional[List[str]], chatbot: CHAT_HISTORY) -> CHAT_HISTORY:
for file in files:
image = Image.open(file).convert('RGB')
image = preprocess_image(image)
image_path = cache_pil_image(image)
chatbot.append(((image_path,), None))
return chatbot
def user(text_prompt: str, chatbot: CHAT_HISTORY):
if text_prompt:
chatbot.append((text_prompt, None))
return "", chatbot
def bot(
google_key: str,
files: Optional[List[str]],
temperature: float,
max_output_tokens: int,
stop_sequences: str,
top_k: int,
top_p: float,
chatbot: CHAT_HISTORY
):
if len(chatbot) == 0:
return chatbot
google_key = google_key if google_key else GOOGLE_API_KEY
if not google_key:
raise ValueError(
"GOOGLE_API_KEY is not set. "
"Please follow the instructions in the README to set it up.")
genai.configure(api_key=google_key)
generation_config = genai.types.GenerationConfig(
temperature=temperature,
max_output_tokens=max_output_tokens,
stop_sequences=preprocess_stop_sequences(stop_sequences=stop_sequences),
top_k=top_k,
top_p=top_p)
if files:
text_prompt = [chatbot[-1][0]] \
if chatbot[-1][0] and isinstance(chatbot[-1][0], str) \
else []
image_prompt = [Image.open(file).convert('RGB') for file in files]
model = genai.GenerativeModel('gemini-pro-vision')
response = model.generate_content(
text_prompt + image_prompt,
stream=True,
generation_config=generation_config)
else:
messages = preprocess_chat_history(chatbot)
model = genai.GenerativeModel('gemini-pro')
response = model.generate_content(
messages,
stream=True,
generation_config=generation_config)
# streaming effect
chatbot[-1][1] = ""
for chunk in response:
for i in range(0, len(chunk.text), 10):
section = chunk.text[i:i + 10]
chatbot[-1][1] += section
time.sleep(0.01)
yield chatbot
google_key_component = gr.Textbox(
label="GOOGLE API KEY",
value="",
type="password",
placeholder="...",
info="You have to provide your own GOOGLE_API_KEY for this app to function properly",
visible=GOOGLE_API_KEY is None
)
chatbot_component = gr.Chatbot(
label='Gemini',
bubble_full_width=False,
avatar_images=AVATAR_IMAGES,
scale=2,
height=400
)
text_prompt_component = gr.Textbox(
placeholder="Hi there! [press Enter]", show_label=False, autofocus=True, scale=8
)
upload_button_component = gr.UploadButton(
label="Upload Images", file_count="multiple", file_types=["image"], scale=1
)
run_button_component = gr.Button(value="Run", variant="primary", scale=1)
temperature_component = gr.Slider(
minimum=0,
maximum=1.0,
value=0.4,
step=0.05,
label="Temperature",
info=(
"Temperature controls the degree of randomness in token selection. Lower "
"temperatures are good for prompts that expect a true or correct response, "
"while higher temperatures can lead to more diverse or unexpected results. "
))
max_output_tokens_component = gr.Slider(
minimum=1,
maximum=2048,
value=1024,
step=1,
label="Token limit",
info=(
"Token limit determines the maximum amount of text output from one prompt. A "
"token is approximately four characters. The default value is 2048."
))
stop_sequences_component = gr.Textbox(
label="Add stop sequence",
value="",
type="text",
placeholder="STOP, END",
info=(
"A stop sequence is a series of characters (including spaces) that stops "
"response generation if the model encounters it. The sequence is not included "
"as part of the response. You can add up to five stop sequences."
))
top_k_component = gr.Slider(
minimum=1,
maximum=40,
value=32,
step=1,
label="Top-K",
info=(
"Top-k changes how the model selects tokens for output. A top-k of 1 means the "
"selected token is the most probable among all tokens in the model’s "
"vocabulary (also called greedy decoding), while a top-k of 3 means that the "
"next token is selected from among the 3 most probable tokens (using "
"temperature)."
))
top_p_component = gr.Slider(
minimum=0,
maximum=1,
value=1,
step=0.01,
label="Top-P",
info=(
"Top-p changes how the model selects tokens for output. Tokens are selected "
"from most probable to least until the sum of their probabilities equals the "
"top-p value. For example, if tokens A, B, and C have a probability of .3, .2, "
"and .1 and the top-p value is .5, then the model will select either A or B as "
"the next token (using temperature). "
))
user_inputs = [
text_prompt_component,
chatbot_component
]
bot_inputs = [
google_key_component,
upload_button_component,
temperature_component,
max_output_tokens_component,
stop_sequences_component,
top_k_component,
top_p_component,
chatbot_component
]
# Hardcoded YouTube URL
youtube_url = "https://www.youtube.com/embed/_q_bfR2Iz4c"
def display_video():
# Embed the YouTube video in an HTML iframe
video_embed = f'<iframe width="560" height="315" src="{youtube_url}" frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>'
return video_embed
# Create a Gradio interface
interface = gr.Interface(
fn=display_video,
outputs="html",
title="YouTube Video Player",
description="This interface displays a hardcoded YouTube video."
)
with gr.Blocks() as demo:
with gr.Tab("Movie Script Writer"):
gr.HTML(TITLE)
with gr.Tab("llamaIndex Image Storage"):
gr.HTML(TITLE)
with gr.Tab("TruEra Gemini Frame Arrranger"):
gr.HTML(TITLE)
gr.HTML(SUBTITLE)
gr.HTML(GETKEY)
with gr.Column():
google_key_component.render()
chatbot_component.render()
with gr.Row():
text_prompt_component.render()
upload_button_component.render()
run_button_component.render()
with gr.Accordion("Parameters", open=False):
temperature_component.render()
max_output_tokens_component.render()
stop_sequences_component.render()
with gr.Accordion("Advanced", open=False):
top_k_component.render()
top_p_component.render()
run_button_component.click(
fn=user,
inputs=user_inputs,
outputs=[text_prompt_component, chatbot_component],
queue=False
).then(
fn=bot, inputs=bot_inputs, outputs=[chatbot_component],
)
text_prompt_component.submit(
fn=user,
inputs=user_inputs,
outputs=[text_prompt_component, chatbot_component],
queue=False
).then(
fn=bot, inputs=bot_inputs, outputs=[chatbot_component],
)
upload_button_component.upload(
fn=upload,
inputs=[upload_button_component, chatbot_component],
outputs=[chatbot_component],
queue=False
)
with gr.Tab("Movie Maker"):
gr.HTML(TITLE)
with gr.Tab("Final Product"):
interface.launch()
demo.queue(max_size=99).launch(debug=False, show_error=True) |