diff --git a/README-ja.md b/README-ja.md
index 064464c..319efe9 100644
--- a/README-ja.md
+++ b/README-ja.md
@@ -38,6 +38,18 @@ PowerShellを使う場合、venvを使えるようにするためには以下の
 - 「Set-ExecutionPolicy Unrestricted」と入力し、Yと答えます。
 - 管理者のPowerShellを閉じます。
 
+## Ubuntu環境でのインストール
+
+```
+git clone https://github.com/kohya-ss/sd-scripts.git
+cd sd-scripts
+bash ubuntu_setup.sh
+```
+
+をコマンドプロンプトで実行し、tkをインストールし、accelerateの質問をWindowsと同じように答えます。
+
+`./gui.sh`でGUIを実行します。
+
 ## Windows環境でのインストール
 
 以下の例ではPyTorchは1.12.1／CUDA 11.6版をインストールします。CUDA 11.3版やPyTorch 1.13を使う場合は適宜書き換えください。
diff --git a/README.md b/README.md
index 7c5dc8f..2dd649b 100644
--- a/README.md
+++ b/README.md
@@ -41,6 +41,19 @@ If you run on Linux and would like to use the GUI, there is now a port of it as
 
 ## Installation
 
+### Ubuntu
+In the terminal, run
+
+```
+git clone https://github.com/bmaltais/kohya_ss.git
+cd kohya_ss
+bash ubuntu_setup.sh
+```
+
+then configure accelerate with the same answers as in the Windows instructions when prompted.
+
+### Windows
+
 Give unrestricted script access to powershell so venv can work:
 
 - Run PowerShell as an administrator
diff --git a/dreambooth_gui.py b/dreambooth_gui.py
index 84517e0..b97e61d 100644
--- a/dreambooth_gui.py
+++ b/dreambooth_gui.py
@@ -456,7 +456,7 @@ def train_model(
 
     print(run_cmd)
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     # check if output_dir/last is a folder... therefore it is a diffuser model
     last_dir = pathlib.Path(f'{output_dir}/{output_name}')
diff --git a/finetune_gui.py b/finetune_gui.py
index 251dee8..751ec21 100644
--- a/finetune_gui.py
+++ b/finetune_gui.py
@@ -33,6 +33,7 @@ refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
 document_symbol = '\U0001F4C4'   # 📄
 
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def save_configuration(
     save_as,
@@ -296,7 +297,7 @@ def train_model(
             os.mkdir(train_dir)
 
         run_cmd = (
-            f'./venv/Scripts/python.exe finetune/merge_captions_to_metadata.py'
+            f'{PYTHON} finetune/merge_captions_to_metadata.py'
         )
         if caption_extension == '':
             run_cmd += f' --caption_extension=".caption"'
@@ -310,12 +311,12 @@ def train_model(
         print(run_cmd)
 
         # Run the command
-        subprocess.run(run_cmd)
+        os.system(run_cmd)
 
     # create images buckets
     if generate_image_buckets:
         run_cmd = (
-            f'./venv/Scripts/python.exe finetune/prepare_buckets_latents.py'
+            f'{PYTHON} finetune/prepare_buckets_latents.py'
         )
         run_cmd += f' "{image_folder}"'
         run_cmd += f' "{train_dir}/{caption_metadata_filename}"'
@@ -334,7 +335,7 @@ def train_model(
         print(run_cmd)
 
         # Run the command
-        subprocess.run(run_cmd)
+        os.system(run_cmd)
 
     image_num = len(
         [
@@ -444,7 +445,7 @@ def train_model(
 
     print(run_cmd)
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     # check if output_dir/last is a folder... therefore it is a diffuser model
     last_dir = pathlib.Path(f'{output_dir}/{output_name}')
diff --git a/gui.sh b/gui.sh
new file mode 100755
index 0000000..90b26db
--- /dev/null
+++ b/gui.sh
@@ -0,0 +1,3 @@
+#!/bin/bash
+source venv/bin/activate
+python kohya_gui.py 
diff --git a/library/basic_caption_gui.py b/library/basic_caption_gui.py
index 2412dfb..23c94fd 100644
--- a/library/basic_caption_gui.py
+++ b/library/basic_caption_gui.py
@@ -2,7 +2,7 @@ import gradio as gr
 from easygui import msgbox
 import subprocess
 from .common_gui import get_folder_path, add_pre_postfix, find_replace
-
+import os
 
 def caption_images(
     caption_text_input,
@@ -38,7 +38,7 @@ def caption_images(
         print(run_cmd)
 
         # Run the command
-        subprocess.run(run_cmd)
+        os.system(run_cmd)
 
     if overwrite_input:
         if not prefix == '' or not postfix == '':
diff --git a/library/blip_caption_gui.py b/library/blip_caption_gui.py
index 61acd75..4f7525e 100644
--- a/library/blip_caption_gui.py
+++ b/library/blip_caption_gui.py
@@ -3,7 +3,7 @@ from easygui import msgbox
 import subprocess
 import os
 from .common_gui import get_folder_path, add_pre_postfix
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def caption_images(
     train_data_dir,
@@ -32,7 +32,7 @@ def caption_images(
         return
 
     print(f'Captioning files in {train_data_dir}...')
-    run_cmd = f'.\\venv\\Scripts\\python.exe "finetune/make_captions.py"'
+    run_cmd = f'{PYTHON} "finetune/make_captions.py"'
     run_cmd += f' --batch_size="{int(batch_size)}"'
     run_cmd += f' --num_beams="{int(num_beams)}"'
     run_cmd += f' --top_p="{top_p}"'
@@ -48,7 +48,7 @@ def caption_images(
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     # Add prefix and postfix
     add_pre_postfix(
diff --git a/library/convert_model_gui.py b/library/convert_model_gui.py
index fca75fc..b99011a 100644
--- a/library/convert_model_gui.py
+++ b/library/convert_model_gui.py
@@ -9,7 +9,7 @@ folder_symbol = '\U0001f4c2'  # 📂
 refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
 document_symbol = '\U0001F4C4'   # 📄
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def convert_model(
     source_model_input,
@@ -40,7 +40,7 @@ def convert_model(
         msgbox('The provided target folder does not exist')
         return
 
-    run_cmd = f'.\\venv\Scripts\python.exe "tools/convert_diffusers20_original_sd.py"'
+    run_cmd = f'{PYTHON} "tools/convert_diffusers20_original_sd.py"'
 
     v1_models = [
         'runwayml/stable-diffusion-v1-5',
@@ -87,7 +87,7 @@ def convert_model(
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     if (
         not target_model_type == 'diffuser'
diff --git a/library/extract_lora_gui.py b/library/extract_lora_gui.py
index 074e3ed..d3864f1 100644
--- a/library/extract_lora_gui.py
+++ b/library/extract_lora_gui.py
@@ -12,7 +12,7 @@ folder_symbol = '\U0001f4c2'  # 📂
 refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
 document_symbol = '\U0001F4C4'   # 📄
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def extract_lora(
     model_tuned,
@@ -41,7 +41,7 @@ def extract_lora(
         return
 
     run_cmd = (
-        f'.\\venv\Scripts\python.exe "networks\extract_lora_from_models.py"'
+        f'{PYTHON} "{os.path.join("networks","extract_lora_from_models.py")}"'
     )
     run_cmd += f' --save_precision {save_precision}'
     run_cmd += f' --save_to "{save_to}"'
@@ -54,7 +54,7 @@ def extract_lora(
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
 
 ###
diff --git a/library/git_caption_gui.py b/library/git_caption_gui.py
index 3cc69d4..1e98e5f 100644
--- a/library/git_caption_gui.py
+++ b/library/git_caption_gui.py
@@ -3,7 +3,7 @@ from easygui import msgbox
 import subprocess
 import os
 from .common_gui import get_folder_path, add_pre_postfix
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def caption_images(
     train_data_dir,
@@ -25,7 +25,7 @@ def caption_images(
         return
 
     print(f'GIT captioning files in {train_data_dir}...')
-    run_cmd = f'.\\venv\\Scripts\\python.exe "finetune/make_captions_by_git.py"'
+    run_cmd = f'{PYTHON} "finetune/make_captions_by_git.py"'
     if not model_id == '':
         run_cmd += f' --model_id="{model_id}"'
     run_cmd += f' --batch_size="{int(batch_size)}"'
diff --git a/library/merge_lora_gui.py b/library/merge_lora_gui.py
index 03a1257..434d1a6 100644
--- a/library/merge_lora_gui.py
+++ b/library/merge_lora_gui.py
@@ -12,7 +12,7 @@ folder_symbol = '\U0001f4c2'  # 📂
 refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
 document_symbol = '\U0001F4C4'   # 📄
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 
 def merge_lora(
     lora_a_model,
@@ -43,7 +43,7 @@ def merge_lora(
     ratio_a = ratio
     ratio_b = 1 - ratio
 
-    run_cmd = f'.\\venv\Scripts\python.exe "networks\merge_lora.py"'
+    run_cmd = f'{PYTHON} "{os.path.join("networks","merge_lora.py")}"'
     run_cmd += f' --save_precision {save_precision}'
     run_cmd += f' --precision {precision}'
     run_cmd += f' --save_to "{save_to}"'
@@ -53,7 +53,7 @@ def merge_lora(
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
 
 ###
diff --git a/library/resize_lora_gui.py b/library/resize_lora_gui.py
index b9dec29..c7fff19 100644
--- a/library/resize_lora_gui.py
+++ b/library/resize_lora_gui.py
@@ -3,7 +3,7 @@ from easygui import msgbox
 import subprocess
 import os
 from .common_gui import get_saveasfilename_path, get_file_path
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 folder_symbol = '\U0001f4c2'  # 📂
 refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
@@ -30,7 +30,7 @@ def resize_lora(
     if device == '':
         device = 'cuda'
 
-    run_cmd = f'.\\venv\Scripts\python.exe "networks\\resize_lora.py"'
+    run_cmd = f'{PYTHON} "{os.path.join("networks","resize_lora.py")}"'
     run_cmd += f' --save_precision {save_precision}'
     run_cmd += f' --save_to {save_to}'
     run_cmd += f' --model {model}'
@@ -40,7 +40,7 @@ def resize_lora(
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
 
 ###
diff --git a/library/tensorboard_gui.py b/library/tensorboard_gui.py
index 1182773..8852e72 100644
--- a/library/tensorboard_gui.py
+++ b/library/tensorboard_gui.py
@@ -5,7 +5,7 @@ import subprocess
 import time
 
 tensorboard_proc = None   # I know... bad but heh
-
+TENSORBOARD = "tensorboard" if os.name == 'posix' else 'tensorboard.exe'
 
 def start_tensorboard(logging_dir):
     global tensorboard_proc
@@ -15,7 +15,7 @@ def start_tensorboard(logging_dir):
         msgbox(msg='Error: log folder is empty')
         return
 
-    run_cmd = f'tensorboard.exe --logdir "{logging_dir}"'
+    run_cmd = [f'{TENSORBOARD}', '--logdir', f'{logging_dir}']
 
     print(run_cmd)
     if tensorboard_proc is not None:
diff --git a/library/verify_lora_gui.py b/library/verify_lora_gui.py
index 51fa510..6ba9c33 100644
--- a/library/verify_lora_gui.py
+++ b/library/verify_lora_gui.py
@@ -7,7 +7,7 @@ from .common_gui import (
     get_any_file_path,
     get_file_path,
 )
-
+PYTHON = "python3" if os.name == 'posix' else './venv/Scripts/python.exe'
 folder_symbol = '\U0001f4c2'  # 📂
 refresh_symbol = '\U0001f504'  # 🔄
 save_style_symbol = '\U0001f4be'  # 💾
@@ -27,13 +27,15 @@ def verify_lora(
         msgbox('The provided model A is not a file')
         return
 
-    run_cmd = f'.\\venv\Scripts\python.exe "networks\check_lora_weights.py"'
-    run_cmd += f' {lora_model}'
+    run_cmd = [
+        PYTHON,
+        os.path.join("networks","check_lora_weights.py"),
+        f'{lora_model}'
+    ]
 
-    print(run_cmd)
+    print(" ".join(run_cmd))
 
     # Run the command
-    subprocess.run(run_cmd)
     process = subprocess.Popen(
         run_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE
     )
diff --git a/library/wd14_caption_gui.py b/library/wd14_caption_gui.py
index 3d97ebf..afbbfef 100644
--- a/library/wd14_caption_gui.py
+++ b/library/wd14_caption_gui.py
@@ -2,6 +2,7 @@ import gradio as gr
 from easygui import msgbox
 import subprocess
 from .common_gui import get_folder_path
+import os
 
 
 def caption_images(train_data_dir, caption_extension, batch_size, thresh):
@@ -30,7 +31,7 @@ def caption_images(train_data_dir, caption_extension, batch_size, thresh):
     print(run_cmd)
 
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     print('...captioning done')
 
diff --git a/lora_gui.py b/lora_gui.py
index 5026618..df39c2d 100644
--- a/lora_gui.py
+++ b/lora_gui.py
@@ -544,7 +544,7 @@ def train_model(
 
     print(run_cmd)
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     # check if output_dir/last is a folder... therefore it is a diffuser model
     last_dir = pathlib.Path(f'{output_dir}/{output_name}')
diff --git a/textual_inversion_gui.py b/textual_inversion_gui.py
index 48d2793..1f3d35e 100644
--- a/textual_inversion_gui.py
+++ b/textual_inversion_gui.py
@@ -498,7 +498,7 @@ def train_model(
 
     print(run_cmd)
     # Run the command
-    subprocess.run(run_cmd)
+    os.system(run_cmd)
 
     # check if output_dir/last is a folder... therefore it is a diffuser model
     last_dir = pathlib.Path(f'{output_dir}/{output_name}')
diff --git a/ubuntu_setup.sh b/ubuntu_setup.sh
new file mode 100755
index 0000000..1431155
--- /dev/null
+++ b/ubuntu_setup.sh
@@ -0,0 +1,12 @@
+#!/bin/bash
+echo installing tk
+sudo apt install python3-tk
+python3 -m venv venv
+source venv/bin/activate
+pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 --extra-index-url https://download.pytorch.org/whl/cu116
+pip install --use-pep517 --upgrade -r requirements.txt
+pip install -U -I --no-deps https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/linux/xformers-0.0.14.dev0-cp310-cp310-linux_x86_64.whl
+
+accelerate config
+
+echo -e "setup finished! run \e[0;92m./gui.sh\e[0m to start"