Spaces:
Sleeping
Sleeping
File size: 16,037 Bytes
c6baa40 5fa72db c6baa40 5fa72db c6baa40 | 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 | import gradio as gr
import sys
shared_theme = gr.themes.Base()
sys.path.append('Engineering-Clinic-Emerging-AI-Design-Interface/Interface_Dependencies')
sys.path.append('Engineering-Clinic-Emerging-AI-Design-Interface/Interface')
sys.path.append('Engineering-Clinic-Emerging-AI-Design-Interface/yolov7-main')
sys.path.append('Engineering-Clinic-Emerging-AI-Design-Interface')
sys.path.append('Interface_Dependencies')
sys.path.append('Interface')
from run_methods_train import run_all_files, interface_login
#from train import interface_login, train
def build_train_interface():
with gr.Blocks(theme=shared_theme) as demo:
gr.Markdown(
"""
# Training Interface for YOLOv7 (Work in progress...)
Train your own YOLOv7 model!
""")
with gr.Row() as file_settings:
# Allows choice for uploading training files or using default official files [official files included in github]
file_type = gr.Radio(label="File Type",info="Choose 'Default' if you are using default values uploaded by server, Choose 'Upload' if you are uploading your own files from your computer",
choices=['Default','Upload'],value='Default',interactive=True,visible=True)
# Allows choice of source and device type
source_type = gr.Radio(label="Source Type",info="'Computer' is set by default",
choices=['Computer'],value='Computer',interactive=True,visible=True)
dev_type = gr.Radio(label="Device Type",info="Choose 'CPU' if you have no gpu, Choose 'Cuda' if you have a cuda enabled gpu",
choices=['CPU','Cuda'],value='CPU',interactive=True,visible=True)
#loggers
with gr.Accordion("Logger Options") as login_accordion:
use_logger = gr.Checkbox(label="Use Logger",info="Check this box if you want to use a logger",visible=True,interactive=True,value=True)
logger_type = gr.Radio(choices=['WANDB', 'ClearML', 'Tensorboard'],value='WANDB',interactive=True,visible=True,
label="Logger",info="Choose which logger to use")
key = gr.Textbox(label="WANDB Key", placeholder="Input your WANDB project key here...", interactive=True)
placeholder = gr.Textbox(label="NULL",placeholder="Not implemented yet, skips loggers process", interactive=False)
response = gr.Textbox(label="NULL",placeholder="Skips loggers process", interactive=False)
login_but = gr.Button(value="Login")
clear_but = gr.ClearButton(value="Clear key", components=[key],
interactive=True,visible=True)
#login_but.click(fn=interface_login,inputs=[logger_type, key],outputs=[])
#Input parameters
with gr.Row("Inputs") as inputs:
is_finetune = gr.Checkbox(label="Finetune",info="Check this box if you want to finetune a model", interactive=True)
custom_pretrained_file = gr.File(label='Weights File',file_count='single', type='file', visible=True,interactive=True)
pretrained_official = gr.File(label='Official Weight File',type='file',file_count='single',file_types=["pt"],value="weights/yolov7.pt", visible=True,interactive=True)
custom_dataset_file = gr.File(label="Custom Dataset",file_count='single',type='file',
file_types=['.zip'],visible=True,interactive=True)
dataset_official = gr.File(label="Official dataset", type='file',file_count='single',file_types=["zip"], value="coco128.zip", visible=True,interactive=True)
custom_data_file = gr.File(label="Custom Data file",file_count='single',type='file',
file_types=['.yaml'],visible=True,interactive=True)
datafile_official = gr.File(label="Official data file", type='file',file_count='single',file_types=["yaml"], value="yolov7-main/data/coco128.yaml", visible=True,interactive=True)
custom_hyp_file = gr.File(label="Custom hyp file",file_count='single',type='file',
file_types=['.yaml'],visible=True,interactive=True)
hypfile_official = gr.File(label="Official hyp file", type='file',file_count='single',file_types=["yaml"], value="yolov7-main/data/hyp.scratch.p6.yaml", visible=True,interactive=True)
#training settings
with gr.Accordion("""Model settings""") as model_settings:
inf_size = gr.Number(label='Inference Size (pixels)',value=640,precision=0, interactive=True)
pgt_lr = gr.Number(label='Learning rate for plausibility gradient',value=0.02, interactive=True)
pgt_lr_decay = gr.Number(label='decay learning rate for plausibility gradient',value=0.54, interactive=True)
agnostic_nms = gr.Checkbox(label='Agnostic NMS',value=True, interactive=True)
pgt_lr_decay_step = gr.Number(label='Step for decay learning rate',value=1.0, interactive=True)
epochs = gr.Number(label='# of training loops',value=5, interactive=True)
conf_thres = gr.Number(label='object confidence threshold',value=0.45, interactive=True)
batch_size = gr.Number(label='total batch size for GPU',value=10, interactive=True)
#device type settings
with gr.Accordion("Device Options") as device_settings:
device = gr.Radio(choices=['0', '1', '2','3'],value='Cuda',interactive=True,visible=True,
label="Cuda",info="Choose cuda device number")
default_device = gr.Radio(value='CPU',interactive=True,visible=True,
label="CPU",info="defaults to CPU if no cuda device is present")
#run timer
with gr.Row("Timer") as outputs:
formatted_time = gr.Textbox(label = 'Time to Run in Seconds:', value = "")
#output file
with gr.Row("Outputs") as outputs:
output_weight_file = gr.File(type='file',label="Output weight file",
interactive=False,visible=True)
clear_comp_list_1 = [output_weight_file, formatted_time, is_finetune,
custom_pretrained_file, custom_dataset_file, custom_data_file, custom_hyp_file]
clear_comp_list_2 = [output_weight_file, formatted_time, is_finetune,
custom_pretrained_file, pretrained_official, custom_dataset_file, dataset_official,
custom_data_file, datafile_official, custom_hyp_file, hypfile_official]
# Row for start & clear buttons
with gr.Row() as buttons:
start_but = gr.Button(label="Start")
clear_but_1= gr.ClearButton(value='Clear All', components=clear_comp_list_1,
interactive=True,visible=True)
clear_but_2= gr.ClearButton(value='Clear All', components=clear_comp_list_2,
interactive=True,visible=True)
def change_Loggers_type(logger):
"""
Changes the visible components of the gradio interface
Args:
CPU (bool): If default defice is CPU
CUDA (bool): If default device is cuda
Returns:
Dictionary: Each component of the interface that needs to be updated.
"""
if logger == "WANDB":
return {key: gr.Textbox(visible=True),
placeholder: gr.Textbox(visible=False),
response: gr.Textbox(visible=False)
}
elif logger == "ClearML":
return {key: gr.Textbox(visible=False),
placeholder: gr.Textbox(visible=True),
response: gr.Textbox(visible=False)
}
elif logger == "Tensorboard":
return {key: gr.Textbox(visible=False),
placeholder: gr.Textbox(visible=True),
response: gr.Textbox(visible=False)
}
def change_device_type(dev):
"""
Changes the visible components of the gradio interface
Args:
CPU (bool): If default defice is CPU
CUDA (bool): If default device is cuda
Returns:
Dictionary: Each component of the interface that needs to be updated.
"""
if dev == "CPU":
return {default_device: gr.Checkbox(visible=True),
device: gr.Slider(visible=False)}
elif dev == "Cuda":
return {default_device: gr.Checkbox(visible=False),
device: gr.Slider(visible=True)}
def change_file_type(file, source):
"""
Changes the visible components of the gradio interface
Args:
file (str): Type of the file (uploaded or default)
source (str): If the file is uploaded from computer
Returns:
Dictionary: Each component of the interface that needs to be updated.
"""
if file == "Default":
return {
output_weight_file: gr.File(visible=True),
formatted_time: gr.Textbox(visible=True),
is_finetune: gr.File(visible=True),
custom_pretrained_file: gr.File(label='Weights File',file_count='single', type='file',file_types=["pt"], visible=False,
interactive=True),
pretrained_official: gr.Dropdown(visible=True),
custom_dataset_file: gr.File(label="Custom Dataset",file_count='single',type='file',
file_types=['.zip'], visible=False,interactive=True),
dataset_official: gr.Dropdown(visible=True),
custom_data_file: gr.File(label="Custom Data file",file_count='single',type='file',
file_types=['.yaml'], visible=False, interactive=True),
datafile_official: gr.Dropdown(visible=True),
custom_hyp_file: gr.File(label="Custom hyp file",file_count='single',type='file',
file_types=['.yaml'], visible=False, interactive=True),
hypfile_official: gr.Dropdown(visible=True),
formatted_time: gr.Textbox(visible=True),
clear_but_1: gr.ClearButton(visible=True),
clear_but_2: gr.ClearButton(visible=False)
}
elif file == "Upload":
if source == "Computer":
return {
output_weight_file: gr.File(visible=True),
formatted_time: gr.Textbox(visible=True),
is_finetune: gr.File(visible=True),
custom_pretrained_file: gr.File(label='Weights File',file_count='single', type='file',file_types=["pt"], visible=True,
interactive=True),
pretrained_official: gr.Dropdown(visible=False),
custom_dataset_file: gr.File(label="Custom Dataset",file_count='single',type='file',
file_types=['.zip'], visible=True,interactive=True),
dataset_official: gr.Dropdown(visible=False),
custom_data_file: gr.File(label="Custom Data file",file_count='single',type='file',
file_types=['.yaml'], visible=True,interactive=True),
datafile_official: gr.Dropdown(visible=False),
custom_hyp_file: gr.File(label="Custom hyp file",file_count='single',type='file',
file_types=['.yaml'], visible=True,interactive=True),
hypfile_official: gr.Dropdown(visible=False),
formatted_time: gr.Textbox(visible=True),
clear_but_1: gr.ClearButton(visible=False),
clear_but_2: gr.ClearButton(visible=True)
}
change_comp_list = [output_weight_file, formatted_time, is_finetune,
custom_pretrained_file, pretrained_official, custom_dataset_file, dataset_official, clear_but_1, clear_but_2,
custom_data_file, datafile_official, custom_hyp_file, hypfile_official, device, default_device, key, placeholder, response]
# List of gradio components that are input into the run_all method (when start button is clicked)
run_inputs = [pgt_lr, pgt_lr_decay, pgt_lr_decay_step, conf_thres, epochs, conf_thres,
batch_size, default_device, device, inf_size, agnostic_nms, custom_pretrained_file]
# List of gradio components that are output from the run_all method (when start button is clicked)
run_outputs = [output_weight_file, formatted_time]
# When these settings are changed, the change_file_type method is called
file_type.input(change_file_type, show_progress=True, inputs=[file_type, source_type], outputs=change_comp_list)
source_type.input(change_file_type, show_progress=True, inputs=[file_type, source_type], outputs=change_comp_list)
dev_type.input(change_device_type, show_progress=True, inputs=[dev_type], outputs=change_comp_list)
logger_type.input(change_Loggers_type, show_progress=True, inputs=[logger_type], outputs=change_comp_list)
# When start button is clicked, the run_all method is called
start_but.click(run_all_files, inputs=run_inputs, outputs=run_outputs)
login_but.click(fn=interface_login,inputs=[logger_type, key],outputs=[])
# When the demo is first started, run the change_file_type method to ensure default settings
demo.load(change_device_type, show_progress=True, inputs=[dev_type], outputs=change_comp_list)
demo.load(change_Loggers_type, show_progress=True, inputs=[logger_type], outputs=change_comp_list)
demo.load(change_file_type, show_progress=True, inputs=[file_type, source_type], outputs=change_comp_list)
return demo
if __name__== "__main__" :
demo = build_train_interface() |