Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,10 +1,23 @@
|
|
1 |
import gradio as gr
|
2 |
import os
|
3 |
import subprocess
|
|
|
4 |
from huggingface_hub import snapshot_download
|
5 |
|
6 |
hf_token = os.environ.get("HF_TOKEN")
|
7 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
def set_accelerate_default_config():
|
10 |
try:
|
@@ -13,7 +26,7 @@ def set_accelerate_default_config():
|
|
13 |
except subprocess.CalledProcessError as e:
|
14 |
print(f"An error occurred: {e}")
|
15 |
|
16 |
-
def train_dreambooth_lora_sdxl(instance_data_dir, lora_trained_xl_folder, instance_prompt, max_train_steps, checkpoint_steps):
|
17 |
|
18 |
script_filename = "train_dreambooth_lora_sdxl.py" # Assuming it's in the same folder
|
19 |
|
@@ -47,15 +60,38 @@ def train_dreambooth_lora_sdxl(instance_data_dir, lora_trained_xl_folder, instan
|
|
47 |
try:
|
48 |
subprocess.run(command, check=True)
|
49 |
print("Training is finished!")
|
|
|
|
|
50 |
except subprocess.CalledProcessError as e:
|
51 |
print(f"An error occurred: {e}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
def main(dataset_id,
|
54 |
lora_trained_xl_folder,
|
55 |
instance_prompt,
|
56 |
max_train_steps,
|
57 |
-
checkpoint_steps
|
|
|
|
|
|
|
|
|
|
|
58 |
|
|
|
|
|
|
|
|
|
|
|
59 |
dataset_repo = dataset_id
|
60 |
|
61 |
# Automatically set local_dir based on the last part of dataset_repo
|
@@ -81,12 +117,36 @@ def main(dataset_id,
|
|
81 |
gr.Info("Training begins ...")
|
82 |
|
83 |
instance_data_dir = repo_parts[-1]
|
84 |
-
train_dreambooth_lora_sdxl(instance_data_dir, lora_trained_xl_folder, instance_prompt, max_train_steps, checkpoint_steps)
|
85 |
|
86 |
return f"Done, your trained model has been stored in your models library: your_user_name/{lora-trained-xl-folder}"
|
87 |
|
88 |
with gr.Blocks() as demo:
|
89 |
with gr.Column():
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
90 |
with gr.Row():
|
91 |
dataset_id = gr.Textbox(label="Dataset ID", info="use one of your previously uploaded datasets on your HF profile", placeholder="diffusers/dog-example")
|
92 |
instance_prompt = gr.Textbox(label="Concept prompt", info="concept prompt - use a unique, made up word to avoid collisions")
|
@@ -95,8 +155,11 @@ with gr.Blocks() as demo:
|
|
95 |
model_output_folder = gr.Textbox(label="Output model folder name", placeholder="lora-trained-xl-folder")
|
96 |
max_train_steps = gr.Number(label="Max Training Steps", value=500)
|
97 |
checkpoint_steps = gr.Number(label="Checkpoints Steps", value=100)
|
|
|
98 |
train_button = gr.Button("Train !")
|
99 |
-
|
|
|
|
|
100 |
|
101 |
train_button.click(
|
102 |
fn = main,
|
@@ -105,7 +168,8 @@ with gr.Blocks() as demo:
|
|
105 |
model_output_folder,
|
106 |
instance_prompt,
|
107 |
max_train_steps,
|
108 |
-
checkpoint_steps
|
|
|
109 |
],
|
110 |
outputs = [status]
|
111 |
)
|
|
|
1 |
import gradio as gr
|
2 |
import os
|
3 |
import subprocess
|
4 |
+
from subprocess import getoutput
|
5 |
from huggingface_hub import snapshot_download
|
6 |
|
7 |
hf_token = os.environ.get("HF_TOKEN")
|
8 |
|
9 |
+
is_shared_ui = True if "fffiloni/train-dreambooth-lora-sdxl" in os.environ['SPACE_ID'] else False
|
10 |
+
|
11 |
+
|
12 |
+
is_gpu_associated = torch.cuda.is_available()
|
13 |
+
if is_gpu_associated:
|
14 |
+
gpu_info = getoutput('nvidia-smi')
|
15 |
+
if("A10G" in gpu_info):
|
16 |
+
which_gpu = "A10G"
|
17 |
+
elif("T4" in gpu_info):
|
18 |
+
which_gpu = "T4"
|
19 |
+
else:
|
20 |
+
which_gpu = "CPU"
|
21 |
|
22 |
def set_accelerate_default_config():
|
23 |
try:
|
|
|
26 |
except subprocess.CalledProcessError as e:
|
27 |
print(f"An error occurred: {e}")
|
28 |
|
29 |
+
def train_dreambooth_lora_sdxl(instance_data_dir, lora_trained_xl_folder, instance_prompt, max_train_steps, checkpoint_steps, remove_gpu):
|
30 |
|
31 |
script_filename = "train_dreambooth_lora_sdxl.py" # Assuming it's in the same folder
|
32 |
|
|
|
60 |
try:
|
61 |
subprocess.run(command, check=True)
|
62 |
print("Training is finished!")
|
63 |
+
if remove_gpu:
|
64 |
+
swap_hardware(hf_token, "cpu-basic")
|
65 |
except subprocess.CalledProcessError as e:
|
66 |
print(f"An error occurred: {e}")
|
67 |
+
|
68 |
+
title="There was an error on during your training"
|
69 |
+
description=f'''
|
70 |
+
Unfortunately there was an error during training your {model_name} model.
|
71 |
+
Please check it out below. Feel free to report this issue to [SD-XL Dreambooth LoRa Training](https://huggingface.co/spaces/fffiloni/train-dreambooth-lora-sdxl):
|
72 |
+
```
|
73 |
+
{str(e)}
|
74 |
+
```
|
75 |
+
'''
|
76 |
+
swap_hardware(hf_token, "cpu-basic")
|
77 |
+
write_to_community(title,description,hf_token)
|
78 |
|
79 |
def main(dataset_id,
|
80 |
lora_trained_xl_folder,
|
81 |
instance_prompt,
|
82 |
max_train_steps,
|
83 |
+
checkpoint_steps,
|
84 |
+
remove_gpu):
|
85 |
+
|
86 |
+
|
87 |
+
if is_shared_ui:
|
88 |
+
raise gr.Error("This Space only works in duplicated instances")
|
89 |
|
90 |
+
if not is_gpu_associated:
|
91 |
+
raise gr.Error("Please associate a T4 or A10G GPU for this Space")
|
92 |
+
|
93 |
+
gr.Warning("## Training is ongoing ⌛... You can close this tab if you like or just wait. If you did not check the `Remove GPU After training`, you can come back here to try your model and upload it after training. Don't forget to remove the GPU attribution after you are done. ")
|
94 |
+
|
95 |
dataset_repo = dataset_id
|
96 |
|
97 |
# Automatically set local_dir based on the last part of dataset_repo
|
|
|
117 |
gr.Info("Training begins ...")
|
118 |
|
119 |
instance_data_dir = repo_parts[-1]
|
120 |
+
train_dreambooth_lora_sdxl(instance_data_dir, lora_trained_xl_folder, instance_prompt, max_train_steps, checkpoint_steps, remove_gpu)
|
121 |
|
122 |
return f"Done, your trained model has been stored in your models library: your_user_name/{lora-trained-xl-folder}"
|
123 |
|
124 |
with gr.Blocks() as demo:
|
125 |
with gr.Column():
|
126 |
+
if is_shared_ui:
|
127 |
+
top_description = gr.HTML(f'''
|
128 |
+
<div class="gr-prose" style="max-width: 80%">
|
129 |
+
<h2>Attention - This Space doesn't work in this shared UI</h2>
|
130 |
+
<p>For it to work, you can duplicate the Space and run it on your own profile using a (paid) private T4-small or A10G-small GPU for training. A T4 costs US$0.60/h, so it should cost < US$1 to train most models using default settings with it! <a class="duplicate-button" style="display:inline-block" target="_blank" href="https://huggingface.co/spaces/{os.environ['SPACE_ID']}?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></p>
|
131 |
+
<img class="instruction" src="file=duplicate.png">
|
132 |
+
<img class="arrow" src="file=arrow.png" />
|
133 |
+
</div>
|
134 |
+
''')
|
135 |
+
else:
|
136 |
+
if(is_gpu_associated):
|
137 |
+
top_description = gr.HTML(f'''
|
138 |
+
<div class="gr-prose" style="max-width: 80%">
|
139 |
+
<h2>You have successfully associated a {which_gpu} GPU to the SD-XL Dreambooth LoRa Training Space 🎉</h2>
|
140 |
+
<p>You can now train your model! You will be billed by the minute from when you activated the GPU until when it is turned it off.</p>
|
141 |
+
</div>
|
142 |
+
''')
|
143 |
+
else:
|
144 |
+
top_description = gr.HTML(f'''
|
145 |
+
<div class="gr-prose" style="max-width: 80%">
|
146 |
+
<h2>You have successfully duplicated the SD-XL Dreambooth LoRa Training Space 🎉</h2>
|
147 |
+
<p>There's only one step left before you can train your model: <a href="https://huggingface.co/spaces/{os.environ['SPACE_ID']}/settings" style="text-decoration: underline" target="_blank">attribute a <b>T4-small or A10G-small GPU</b> to it (via the Settings tab)</a> and run the training below. You will be billed by the minute from when you activate the GPU until when it is turned it off.</p>
|
148 |
+
</div>
|
149 |
+
''')
|
150 |
with gr.Row():
|
151 |
dataset_id = gr.Textbox(label="Dataset ID", info="use one of your previously uploaded datasets on your HF profile", placeholder="diffusers/dog-example")
|
152 |
instance_prompt = gr.Textbox(label="Concept prompt", info="concept prompt - use a unique, made up word to avoid collisions")
|
|
|
155 |
model_output_folder = gr.Textbox(label="Output model folder name", placeholder="lora-trained-xl-folder")
|
156 |
max_train_steps = gr.Number(label="Max Training Steps", value=500)
|
157 |
checkpoint_steps = gr.Number(label="Checkpoints Steps", value=100)
|
158 |
+
remove_gpu = gr.Checkbox(label="Remove GPU After Training", value=True)
|
159 |
train_button = gr.Button("Train !")
|
160 |
+
|
161 |
+
|
162 |
+
status = gr.Textbox(label="Training status")
|
163 |
|
164 |
train_button.click(
|
165 |
fn = main,
|
|
|
168 |
model_output_folder,
|
169 |
instance_prompt,
|
170 |
max_train_steps,
|
171 |
+
checkpoint_steps,
|
172 |
+
remove_gpu
|
173 |
],
|
174 |
outputs = [status]
|
175 |
)
|