', '', dialogue)
@@ -339,11 +332,12 @@ def generate_pfp_cache(character):
return img
return None
-def load_character(character, name1, name2):
+def load_character(character, name1, name2, instruct=False):
shared.character = character
shared.history['internal'] = []
shared.history['visible'] = []
- greeting = ""
+ context = greeting = end_of_turn = ""
+ greeting_field = 'greeting'
picture = None
# Deleting the profile picture cache, if any
@@ -351,9 +345,10 @@ def load_character(character, name1, name2):
Path("cache/pfp_character.png").unlink()
if character != 'None':
+ folder = "characters" if not instruct else "characters/instruction-following"
picture = generate_pfp_cache(character)
for extension in ["yml", "yaml", "json"]:
- filepath = Path(f'characters/{character}.{extension}')
+ filepath = Path(f'{folder}/{character}.{extension}')
if filepath.exists():
break
file_contents = open(filepath, 'r', encoding='utf-8').read()
@@ -369,19 +364,21 @@ def load_character(character, name1, name2):
if 'context' in data:
context = f"{data['context'].strip()}\n\n"
- greeting_field = 'greeting'
- else:
+ elif "char_persona" in data:
context = build_pygmalion_style_context(data)
greeting_field = 'char_greeting'
- if 'example_dialogue' in data and data['example_dialogue'] != '':
+ if 'example_dialogue' in data:
context += f"{data['example_dialogue'].strip()}\n"
- if greeting_field in data and len(data[greeting_field].strip()) > 0:
+ if greeting_field in data:
greeting = data[greeting_field]
+ if 'end_of_turn' in data:
+ end_of_turn = data['end_of_turn']
else:
context = shared.settings['context']
name2 = shared.settings['name2']
greeting = shared.settings['greeting']
+ end_of_turn = shared.settings['end_of_turn']
if Path(f'logs/{shared.character}_persistent.json').exists():
load_history(open(Path(f'logs/{shared.character}_persistent.json'), 'rb').read(), name1, name2)
@@ -389,10 +386,7 @@ def load_character(character, name1, name2):
shared.history['internal'] += [['<|BEGIN-VISIBLE-CHAT|>', greeting]]
shared.history['visible'] += [['', apply_extensions(greeting, "output")]]
- if shared.args.cai_chat:
- return name1, name2, picture, greeting, context, generate_chat_html(shared.history['visible'], name1, name2, reset_cache=True)
- else:
- return name1, name2, picture, greeting, context, shared.history['visible']
+ return name1, name2, picture, greeting, context, end_of_turn, chat_html_wrapper(shared.history['visible'], name1, name2, reset_cache=True)
def load_default_history(name1, name2):
load_character("None", name1, name2)
@@ -423,7 +417,7 @@ def upload_tavern_character(img, name1, name2):
_json = {"char_name": _json['name'], "char_persona": _json['description'], "char_greeting": _json["first_mes"], "example_dialogue": _json['mes_example'], "world_scenario": _json['scenario']}
return upload_character(json.dumps(_json), img, tavern=True)
-def upload_your_profile_picture(img, name1, name2):
+def upload_your_profile_picture(img, name1, name2, mode):
cache_folder = Path("cache")
if not cache_folder.exists():
cache_folder.mkdir()
@@ -436,7 +430,4 @@ def upload_your_profile_picture(img, name1, name2):
img.save(Path('cache/pfp_me.png'))
print('Profile picture saved to "cache/pfp_me.png"')
- if shared.args.cai_chat:
- return generate_chat_html(shared.history['visible'], name1, name2, reset_cache=True)
- else:
- return shared.history['visible']
+ return chat_html_wrapper(shared.history['visible'], name1, name2, mode, reset_cache=True)
diff --git a/modules/html_generator.py b/modules/html_generator.py
index e1c085a6..6fb8457f 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -21,6 +21,8 @@ with open(Path(__file__).resolve().parent / '../css/html_4chan_style.css', 'r')
_4chan_css = css_f.read()
with open(Path(__file__).resolve().parent / '../css/html_cai_style.css', 'r') as f:
cai_css = f.read()
+with open(Path(__file__).resolve().parent / '../css/html_instruct_style.css', 'r') as f:
+ instruct_css = f.read()
def fix_newlines(string):
string = string.replace('\n', '\n\n')
@@ -117,7 +119,39 @@ def get_image_cache(path):
return image_cache[path][1]
-def generate_chat_html(history, name1, name2, reset_cache=False):
+def generate_instruct_html(history):
+ output = f''
+ for i,_row in enumerate(history[::-1]):
+ row = [convert_to_markdown(entry) for entry in _row]
+
+ output += f"""
+
+ """
+
+ if len(row[0]) == 0: # don't display empty user messages
+ continue
+
+ output += f"""
+
+ """
+
+ output += "
"
+
+ return output
+
+def generate_cai_chat_html(history, name1, name2, reset_cache=False):
output = f''
# The time.time() is to prevent the brower from caching the image
@@ -165,3 +199,17 @@ def generate_chat_html(history, name1, name2, reset_cache=False):
output += "
"
return output
+
+def generate_chat_html(history, name1, name2):
+ return generate_cai_chat_html(history, name1, name2)
+
+def chat_html_wrapper(history, name1, name2, mode="cai-chat", reset_cache=False):
+
+ if mode == "cai-chat":
+ return generate_cai_chat_html(history, name1, name2, reset_cache)
+ elif mode == "chat":
+ return generate_chat_html(history, name1, name2)
+ elif mode == "instruct":
+ return generate_instruct_html(history)
+ else:
+ return ''
diff --git a/modules/shared.py b/modules/shared.py
index 6c183a81..a6f58778 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -33,6 +33,7 @@ settings = {
'name2': 'Assistant',
'context': 'This is a conversation with your Assistant. The Assistant is very helpful and is eager to chat with you and answer your questions.',
'greeting': 'Hello there!',
+ 'end_of_turn': '',
'stop_at_newline': False,
'chat_prompt_size': 2048,
'chat_prompt_size_min': 0,
@@ -73,8 +74,8 @@ parser = argparse.ArgumentParser(formatter_class=lambda prog: argparse.HelpForma
# Basic settings
parser.add_argument('--notebook', action='store_true', help='Launch the web UI in notebook mode, where the output is written to the same text box as the input.')
-parser.add_argument('--chat', action='store_true', help='Launch the web UI in chat mode.')
-parser.add_argument('--cai-chat', action='store_true', help='Launch the web UI in chat mode with a style similar to the Character.AI website.')
+parser.add_argument('--chat', action='store_true', help='Launch the web UI in chat mode with a style similar to the Character.AI website.')
+parser.add_argument('--cai-chat', action='store_true', help='DEPRECATED: use --chat instead.')
parser.add_argument('--model', type=str, help='Name of the model to load by default.')
parser.add_argument('--lora', type=str, help='Name of the LoRA to apply to the model by default.')
parser.add_argument("--model-dir", type=str, default='models/', help="Path to directory with all the models")
@@ -131,12 +132,17 @@ parser.add_argument("--gradio-auth-path", type=str, help='Set the gradio authent
args = parser.parse_args()
-# Provisional, this will be deleted later
+# Deprecation warnings for parameters that have been renamed
deprecated_dict = {'gptq_bits': ['wbits', 0], 'gptq_model_type': ['model_type', None], 'gptq_pre_layer': ['prelayer', 0]}
for k in deprecated_dict:
if eval(f"args.{k}") != deprecated_dict[k][1]:
print(f"Warning: --{k} is deprecated and will be removed. Use --{deprecated_dict[k][0]} instead.")
exec(f"args.{deprecated_dict[k][0]} = args.{k}")
+# Deprecation warnings for parameters that have been removed
+if args.cai_chat:
+ print("Warning: --cai-chat is deprecated. Use --chat instead.")
+ args.chat = True
+
def is_chat():
- return any((args.chat, args.cai_chat))
+ return args.chat
diff --git a/server.py b/server.py
index a34c86f7..f367ca0d 100644
--- a/server.py
+++ b/server.py
@@ -12,7 +12,7 @@ from PIL import Image
import modules.extensions as extensions_module
from modules import chat, shared, training, ui
-from modules.html_generator import generate_chat_html
+from modules.html_generator import chat_html_wrapper
from modules.LoRA import add_lora_to_model
from modules.models import load_model, load_soft_prompt
from modules.text_generation import (clear_torch_cache, generate_reply,
@@ -48,6 +48,10 @@ def get_available_prompts():
def get_available_characters():
paths = (x for x in Path('characters').iterdir() if x.suffix in ('.json', '.yaml', '.yml'))
+ return ['None'] + sorted(set((k.stem for k in paths if k.stem != "instruction-following")), key=str.lower)
+
+def get_available_instruction_templates():
+ paths = (x for x in Path('characters/instruction-following').iterdir() if x.suffix in ('.json', '.yaml', '.yml'))
return ['None'] + sorted(set((k.stem for k in paths)), key=str.lower)
def get_available_extensions():
@@ -145,7 +149,7 @@ def load_prompt(fname):
if text[-1] == '\n':
text = text[:-1]
return text
-
+
def create_prompt_menus():
with gr.Row():
with gr.Column():
@@ -296,10 +300,7 @@ def create_interface():
if shared.is_chat():
shared.gradio['Chat input'] = gr.State()
with gr.Tab("Text generation", elem_id="main"):
- if shared.args.cai_chat:
- shared.gradio['display'] = gr.HTML(value=generate_chat_html(shared.history['visible'], shared.settings['name1'], shared.settings['name2']))
- else:
- shared.gradio['display'] = gr.Chatbot(value=shared.history['visible'], elem_id="gradio-chatbot")
+ shared.gradio['display'] = gr.HTML(value=chat_html_wrapper(shared.history['visible'], shared.settings['name1'], shared.settings['name2']))
shared.gradio['textbox'] = gr.Textbox(label='Input')
with gr.Row():
shared.gradio['Generate'] = gr.Button('Generate')
@@ -316,13 +317,17 @@ def create_interface():
shared.gradio['Clear history-confirm'] = gr.Button('Confirm', variant="stop", visible=False)
shared.gradio['Clear history-cancel'] = gr.Button('Cancel', visible=False)
+ shared.gradio["Chat mode"] = gr.Radio(choices=["cai-chat", "chat", "instruct"], value="cai-chat", label="Mode")
+ shared.gradio["Instruction templates"] = gr.Dropdown(choices=get_available_instruction_templates(), label="Instruction template", value="None", visible=False)
+
with gr.Tab("Character", elem_id="chat-settings"):
with gr.Row():
with gr.Column(scale=8):
shared.gradio['name1'] = gr.Textbox(value=shared.settings['name1'], lines=1, label='Your name')
shared.gradio['name2'] = gr.Textbox(value=shared.settings['name2'], lines=1, label='Character\'s name')
- shared.gradio['greeting'] = gr.Textbox(value=shared.settings['greeting'], lines=2, label='Greeting')
- shared.gradio['context'] = gr.Textbox(value=shared.settings['context'], lines=8, label='Context')
+ shared.gradio['greeting'] = gr.Textbox(value=shared.settings['greeting'], lines=4, label='Greeting')
+ shared.gradio['context'] = gr.Textbox(value=shared.settings['context'], lines=4, label='Context')
+ shared.gradio['end_of_turn'] = gr.Textbox(value=shared.settings["end_of_turn"], lines=1, label='End of turn string')
with gr.Column(scale=1):
shared.gradio['character_picture'] = gr.Image(label='Character picture', type="pil")
shared.gradio['your_picture'] = gr.Image(label='Your picture', type="pil", value=Image.open(Path("cache/pfp_me.png")) if Path("cache/pfp_me.png").exists() else None)
@@ -367,31 +372,31 @@ def create_interface():
create_settings_menus(default_preset)
- function_call = 'chat.cai_chatbot_wrapper' if shared.args.cai_chat else 'chat.chatbot_wrapper'
- shared.input_params = [shared.gradio[k] for k in ['Chat input', 'max_new_tokens', 'do_sample', 'temperature', 'top_p', 'typical_p', 'repetition_penalty', 'encoder_repetition_penalty', 'top_k', 'min_length', 'no_repeat_ngram_size', 'num_beams', 'penalty_alpha', 'length_penalty', 'early_stopping', 'seed', 'name1', 'name2', 'context', 'check', 'chat_prompt_size_slider', 'chat_generation_attempts']]
+ shared.input_params = [shared.gradio[k] for k in ['Chat input', 'max_new_tokens', 'do_sample', 'temperature', 'top_p', 'typical_p', 'repetition_penalty', 'encoder_repetition_penalty', 'top_k', 'min_length', 'no_repeat_ngram_size', 'num_beams', 'penalty_alpha', 'length_penalty', 'early_stopping', 'seed', 'name1', 'name2', 'context', 'check', 'chat_prompt_size_slider', 'chat_generation_attempts', 'Chat mode', 'end_of_turn']]
def set_chat_input(textbox):
return textbox, ""
gen_events.append(shared.gradio['Generate'].click(set_chat_input, shared.gradio['textbox'], [shared.gradio['Chat input'], shared.gradio['textbox']], show_progress=False))
- gen_events.append(shared.gradio['Generate'].click(eval(function_call), shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream))
+ gen_events.append(shared.gradio['Generate'].click(chat.cai_chatbot_wrapper, shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream))
gen_events.append(shared.gradio['textbox'].submit(set_chat_input, shared.gradio['textbox'], [shared.gradio['Chat input'], shared.gradio['textbox']], show_progress=False))
- gen_events.append(shared.gradio['textbox'].submit(eval(function_call), shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream))
+ gen_events.append(shared.gradio['textbox'].submit(chat.cai_chatbot_wrapper, shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream))
gen_events.append(shared.gradio['Regenerate'].click(chat.regenerate_wrapper, shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream))
gen_events.append(shared.gradio['Impersonate'].click(chat.impersonate_wrapper, shared.input_params, shared.gradio['textbox'], show_progress=shared.args.no_stream))
shared.gradio['Stop'].click(stop_everything_event, [], [], queue=False, cancels=gen_events if shared.args.no_stream else None)
shared.gradio['Copy last reply'].click(chat.send_last_reply_to_input, [], shared.gradio['textbox'], show_progress=shared.args.no_stream)
- shared.gradio['Replace last reply'].click(chat.replace_last_reply, [shared.gradio['textbox'], shared.gradio['name1'], shared.gradio['name2']], shared.gradio['display'], show_progress=shared.args.no_stream)
+ shared.gradio['Replace last reply'].click(chat.replace_last_reply, [shared.gradio[k] for k in ['textbox', 'name1', 'name2', 'Chat mode']], shared.gradio['display'], show_progress=shared.args.no_stream)
# Clear history with confirmation
clear_arr = [shared.gradio[k] for k in ['Clear history-confirm', 'Clear history', 'Clear history-cancel']]
shared.gradio['Clear history'].click(lambda :[gr.update(visible=True), gr.update(visible=False), gr.update(visible=True)], None, clear_arr)
shared.gradio['Clear history-confirm'].click(lambda :[gr.update(visible=False), gr.update(visible=True), gr.update(visible=False)], None, clear_arr)
- shared.gradio['Clear history-confirm'].click(chat.clear_chat_log, [shared.gradio['name1'], shared.gradio['name2'], shared.gradio['greeting']], shared.gradio['display'])
+ shared.gradio['Clear history-confirm'].click(chat.clear_chat_log, [shared.gradio[k] for k in ['name1', 'name2', 'greeting', 'Chat mode']], shared.gradio['display'])
shared.gradio['Clear history-cancel'].click(lambda :[gr.update(visible=False), gr.update(visible=True), gr.update(visible=False)], None, clear_arr)
+ shared.gradio['Chat mode'].change(lambda x : gr.update(visible= x=='instruct'), shared.gradio['Chat mode'], shared.gradio['Instruction templates'])
- shared.gradio['Remove last'].click(chat.remove_last_message, [shared.gradio['name1'], shared.gradio['name2']], [shared.gradio['display'], shared.gradio['textbox']], show_progress=False)
+ shared.gradio['Remove last'].click(chat.remove_last_message, [shared.gradio[k] for k in ['name1', 'name2', 'Chat mode']], [shared.gradio['display'], shared.gradio['textbox']], show_progress=False)
shared.gradio['download_button'].click(chat.save_history, inputs=[], outputs=[shared.gradio['download']])
shared.gradio['Upload character'].click(chat.upload_character, [shared.gradio['upload_json'], shared.gradio['upload_img_bot']], [shared.gradio['character_menu']])
@@ -404,18 +409,20 @@ def create_interface():
shared.gradio['textbox'].submit(lambda : chat.save_history(timestamp=False), [], [], show_progress=False)
shared.gradio['character_menu'].change(chat.load_character, [shared.gradio[k] for k in ['character_menu', 'name1', 'name2']], [shared.gradio[k] for k in ['name1', 'name2', 'character_picture', 'greeting', 'context', 'display']])
- shared.gradio['upload_chat_history'].upload(chat.load_history, [shared.gradio['upload_chat_history'], shared.gradio['name1'], shared.gradio['name2']], [])
+ shared.gradio['Instruction templates'].change(lambda character, name1, name2: chat.load_character(character, name1, name2, instruct=True), [shared.gradio[k] for k in ['Instruction templates', 'name1', 'name2']], [shared.gradio[k] for k in ['name1', 'name2', 'character_picture', 'greeting', 'context', 'end_of_turn', 'display']])
+ shared.gradio['upload_chat_history'].upload(chat.load_history, [shared.gradio[k] for k in ['upload_chat_history', 'name1', 'name2']], [])
shared.gradio['upload_img_tavern'].upload(chat.upload_tavern_character, [shared.gradio['upload_img_tavern'], shared.gradio['name1'], shared.gradio['name2']], [shared.gradio['character_menu']])
- shared.gradio['your_picture'].change(chat.upload_your_profile_picture, [shared.gradio[k] for k in ['your_picture', 'name1', 'name2']], shared.gradio['display'])
+ shared.gradio['your_picture'].change(chat.upload_your_profile_picture, [shared.gradio[k] for k in ['your_picture', 'name1', 'name2', 'Chat mode']], shared.gradio['display'])
- reload_func = chat.redraw_html if shared.args.cai_chat else lambda : shared.history['visible']
- reload_inputs = [shared.gradio['name1'], shared.gradio['name2']] if shared.args.cai_chat else []
- shared.gradio['upload_chat_history'].upload(reload_func, reload_inputs, [shared.gradio['display']])
- shared.gradio['Stop'].click(reload_func, reload_inputs, [shared.gradio['display']])
+ reload_inputs = [shared.gradio[k] for k in ['name1', 'name2', 'Chat mode']]
+ shared.gradio['upload_chat_history'].upload(chat.redraw_html, reload_inputs, [shared.gradio['display']])
+ shared.gradio['Stop'].click(chat.redraw_html, reload_inputs, [shared.gradio['display']])
+ shared.gradio['Instruction templates'].change(chat.redraw_html, reload_inputs, [shared.gradio['display']])
+ shared.gradio['Chat mode'].change(chat.redraw_html, reload_inputs, [shared.gradio['display']])
shared.gradio['interface'].load(None, None, None, _js=f"() => {{{ui.main_js+ui.chat_js}}}")
shared.gradio['interface'].load(lambda : chat.load_default_history(shared.settings['name1'], shared.settings['name2']), None, None)
- shared.gradio['interface'].load(reload_func, reload_inputs, [shared.gradio['display']], show_progress=True)
+ shared.gradio['interface'].load(chat.redraw_html, reload_inputs, [shared.gradio['display']], show_progress=True)
elif shared.args.notebook:
with gr.Tab("Text generation", elem_id="main"):