File size: 12,735 Bytes
e2d8169 de0e23f 8ad1496 e2d8169 448cee0 e2d8169 f2d7c4e e2d8169 cb2784c e2d8169 b3d62ee 8ad1496 e2d8169 f2d7c4e e2d8169 e35a4fa e2d8169 950cada e2d8169 0353e7c e2d8169 b3d62ee fcaf436 b3d62ee e2d8169 e35a4fa 6048dff f2d7c4e e2d8169 e35a4fa e2d8169 9b3281c 97b2304 0840021 97b2304 97edf64 97b2304 e2d8169 71493cd e2d8169 71493cd e2d8169 9b3281c e2d8169 5336f00 e2d8169 9b3281c e2d8169 5336f00 e2d8169 2105fb2 950cada 2105fb2 e7dff22 09e7c9c 9fe092c e7dff22 950cada e7dff22 09e7c9c 9fe092c e7dff22 950cada 3db9444 2105fb2 cf26105 09e7c9c 2105fb2 9fe092c 2105fb2 85c4fc8 950cada 2105fb2 cf26105 09e7c9c 2105fb2 9fe092c e7dff22 85c4fc8 950cada 3db9444 e7dff22 09e7c9c 9fe092c e35a4fa e7dff22 85c4fc8 950cada e7dff22 09e7c9c 9fe092c e35a4fa 2105fb2 85c4fc8 950cada e2ffb20 09e7c9c e2ffb20 e35a4fa e2ffb20 2105fb2 c2b864f 2105fb2 e35a4fa |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 |
import os
import time
from typing import List, Tuple, Optional
from pathlib import Path
from moviepy.editor import VideoFileClip, concatenate_videoclips
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">🕹️ Google Gemini Playground 🔥</h1>"""
SUBTITLE = """<h2 align="center">Play with Gemini Text and Vision Model API 🖇️</h2>"""
DUPLICATE = """
<div style="text-align: center; display: flex; justify-content: center; align-items: center;">
<a href="https://huggingface.co/spaces/tsereno/Gemini-Powered-App?logs=container&duplicate=true">
<img src="https://bit.ly/3gLdBN6" alt="Duplicate Space" style="margin-right: 10px;">
</a>
<span>Duplicate the Space and run securely with your
<a href="https://makersuite.google.com/app/apikey">GOOGLE API KEY</a>.
</span>
</div>
"""
IMAGE_WIDTH = 512
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 user(text_prompt: str, chatbot: List[Tuple[str, str]]):
return "", chatbot + [[text_prompt, None]]
from moviepy.editor import VideoFileClip, concatenate_videoclips
def bot(
google_key: str,
model_name: str,
image_prompt: Optional[Image.Image],
video_prompt: List[Path],
temperature: float,
max_output_tokens: int,
stop_sequences: str,
top_k: int,
top_p: float,
text_prompt_component: str,
chatbot: List[Tuple[str, str]]
):
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.")
text_prompt = chatbot[-1][0]
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 video_prompt is not None:
if len(video_prompt) > 1:
video_clips = [VideoFileClip(str(video_path)) for video_path in video_prompt]
merged_video = concatenate_videoclips(video_clips)
video_path = "merged_video.mp4"
merged_video.write_videofile(video_path)
else:
video_path = str(video_prompt[0])
model = genai.GenerativeModel(model_name)
video_file = genai.upload_file(path=video_path)
while video_file.state.name == "PROCESSING":
print('.', end='')
time.sleep(10)
video_file = genai.get_file(video_file.name)
if video_file.state.name == "FAILED":
raise ValueError(video_file.state.name)
response = model.generate_content(
contents=[video_file, text_prompt],
stream=True,
generation_config=generation_config,
request_options={"timeout": 600})
response.resolve()
elif image_prompt is not None:
image_prompt = preprocess_image(image_prompt)
model = genai.GenerativeModel(model_name)
response = model.generate_content(
contents=[text_prompt, image_prompt],
stream=True,
generation_config=generation_config)
response.resolve()
else:
model = genai.GenerativeModel(model_name)
response = model.generate_content(
text_prompt,
stream=True,
generation_config=generation_config)
response.resolve()
# 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
)
image_prompt_component = gr.Image(type="pil", label="Image", scale=1)
video_prompt_component = gr.File(label="Video", file_count="multiple")
model_selection = gr.Dropdown(["gemini-1.5-flash-latest", "gemini-1.5-pro-latest"],label="Select Gemini Model",value="gemini-1.5-pro-latest")
chatbot_component = gr.Chatbot(
label='Gemini',
bubble_full_width=False,
scale=2
)
text_prompt_component = gr.Textbox(
placeholder="Hi there!",
label="Ask me anything and press Enter"
)
run_button_component = gr.Button()
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,
model_selection,
image_prompt_component,
video_prompt_component,
temperature_component,
max_output_tokens_component,
stop_sequences_component,
top_k_component,
top_p_component,
text_prompt_component,
chatbot_component
]
with gr.Blocks() as demo:
gr.HTML(TITLE)
gr.HTML(SUBTITLE)
gr.HTML(DUPLICATE)
with gr.Column():
google_key_component.render()
with gr.Row():
video_prompt_component.render()
chatbot_component.render()
text_prompt_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]
)
gr.Examples(
fn=bot,
inputs=bot_inputs,
outputs=[chatbot_component],
examples=
[
[
"",
"gemini-1.5-pro-latest",
None,
None,
.4,
1024,
"",
32,
1,
"How far is the moon from the earth?",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
None,
None,
.4,
1024,
"",
32,
1,
"What is 2+2?",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
"./example1.webp",
None,
.4,
1024,
"",
32,
1,
"If there is an image of golf simulator screen, list all the stat values displayed and summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing. If there is a video of golf swing, summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing.",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
"./example2.jpg",
None,
.4,
1024,
"",
32,
1,
"If there is an image of golf simulator screen, list all the stat values displayed and summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing. If there is a video of golf swing, summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing.",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
None,
["./example1.mp4"],
.4,
1024,
"",
32,
1,
"If there is an image of golf simulator screen, list all the stat values displayed and summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing. If there is a video of golf swing, summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing.",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
None,
["./example2.mp4"],
.4,
1024,
"",
32,
1,
"If there is an image of golf simulator screen, list all the stat values displayed and summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing. If there is a video of golf swing, summarize faults and provide recommendations in the form of low effort fixes for optimizing the golf swing.",
[("", "")]
],
[
"",
"gemini-1.5-pro-latest",
None,
["./example3.mp4"],
.4,
1024,
"",
32,
1,
"Transcribe",
[("", "")]
]
],
#cache_examples="lazy",
)
demo.queue(max_size=99).launch(debug=False, show_error=True) |