allow custom stopping strings in all modes (#903)

This commit is contained in:
catalpaaa 2023-04-11 08:30:06 -07:00 committed by GitHub
parent 0f212093a3
commit 78bbc66fc4
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 37 additions and 23 deletions

View file

@ -232,10 +232,8 @@ def create_model_menus():
def create_settings_menus(default_preset):
generate_params = load_preset_values(default_preset if not shared.args.flexgen else 'Naive', {}, return_dict=True)
for k in ['max_new_tokens', 'seed', 'stop_at_newline', 'chat_prompt_size', 'chat_generation_attempts', 'add_bos_token']:
generate_params[k] = shared.settings[k]
shared.gradio['generate_state'] = gr.State(generate_params)
with gr.Row():
with gr.Column():
@ -273,7 +271,12 @@ def create_settings_menus(default_preset):
with gr.Column():
shared.gradio['length_penalty'] = gr.Slider(-5, 5, value=generate_params['length_penalty'], label='length_penalty')
shared.gradio['early_stopping'] = gr.Checkbox(value=generate_params['early_stopping'], label='early_stopping')
shared.gradio['add_bos_token'] = gr.Checkbox(value=shared.settings['add_bos_token'], label='Add the bos_token to the beginning of prompts', info='Disabling this can make the replies more creative.')
with gr.Row():
shared.gradio['add_bos_token'] = gr.Checkbox(value=shared.settings['add_bos_token'], label='Add the bos_token to the beginning of prompts', info='Disabling this can make the replies more creative.')
with gr.Row():
shared.gradio['custom_stopping_strings'] = gr.Textbox(lines=1, value=shared.settings["custom_stopping_strings"] or None, label='Custom stopping strings', info='In addition to the defaults. Written between "" and separated by commas. For instance: "\\nYour Assistant:", "\\nThe assistant:"')
with gr.Accordion('Soft prompt', open=False):
with gr.Row():
@ -284,7 +287,7 @@ def create_settings_menus(default_preset):
with gr.Row():
shared.gradio['upload_softprompt'] = gr.File(type='binary', file_types=['.zip'])
shared.gradio['preset_menu'].change(load_preset_values, [shared.gradio[k] for k in ['preset_menu', 'generate_state']], [shared.gradio[k] for k in ['generate_state', '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']])
shared.gradio['preset_menu'].change(load_preset_values, [shared.gradio[k] for k in ['preset_menu', 'interface_state']], [shared.gradio[k] for k in ['interface_state', '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']])
shared.gradio['softprompts_menu'].change(load_soft_prompt, shared.gradio['softprompts_menu'], shared.gradio['softprompts_menu'], show_progress=True)
shared.gradio['upload_softprompt'].upload(upload_soft_prompt, shared.gradio['upload_softprompt'], shared.gradio['softprompts_menu'])
@ -358,7 +361,7 @@ title = 'Text generation web UI'
def list_interface_input_elements(chat=False):
elements = ['max_new_tokens', 'seed', 'temperature', 'top_p', 'top_k', 'typical_p', 'repetition_penalty', 'encoder_repetition_penalty', 'no_repeat_ngram_size', 'min_length', 'do_sample', 'penalty_alpha', 'num_beams', 'length_penalty', 'early_stopping', 'add_bos_token']
elements = ['max_new_tokens', 'seed', 'temperature', 'top_p', 'top_k', 'typical_p', 'repetition_penalty', 'encoder_repetition_penalty', 'no_repeat_ngram_size', 'min_length', 'do_sample', 'penalty_alpha', 'num_beams', 'length_penalty', 'early_stopping', 'add_bos_token', 'custom_stopping_strings']
if chat:
elements += ['name1', 'name2', 'greeting', 'context', 'end_of_turn', 'chat_prompt_size', 'chat_generation_attempts', 'stop_at_newline', 'mode']
return elements
@ -368,6 +371,7 @@ def gather_interface_values(*args):
output = {}
for i, element in enumerate(shared.input_elements):
output[element] = args[i]
output['custom_stopping_strings'] = eval(f"[{output['custom_stopping_strings']}]")
return output
@ -453,7 +457,7 @@ def create_interface():
shared.gradio['chat_prompt_size'] = gr.Slider(minimum=shared.settings['chat_prompt_size_min'], maximum=shared.settings['chat_prompt_size_max'], step=1, label='Maximum prompt size in tokens', value=shared.settings['chat_prompt_size'])
with gr.Column():
shared.gradio['chat_generation_attempts'] = gr.Slider(minimum=shared.settings['chat_generation_attempts_min'], maximum=shared.settings['chat_generation_attempts_max'], value=shared.settings['chat_generation_attempts'], step=1, label='Generation attempts (for longer replies)')
shared.gradio['stop_at_newline'] = gr.Checkbox(value=shared.settings['stop_at_newline'], label='Stop generating at new line character?')
shared.gradio['stop_at_newline'] = gr.Checkbox(value=shared.settings['stop_at_newline'], label='Stop generating at new line character')
create_settings_menus(default_preset)
@ -563,7 +567,7 @@ def create_interface():
with gr.Tab("Parameters", elem_id="parameters"):
create_settings_menus(default_preset)
shared.input_params = [shared.gradio[k] for k in ['textbox', 'generate_state']]
shared.input_params = [shared.gradio[k] for k in ['textbox', 'interface_state']]
output_params = [shared.gradio[k] for k in ['textbox', 'markdown', 'html']]
gen_events.append(shared.gradio['Generate'].click(
@ -607,7 +611,7 @@ def create_interface():
with gr.Tab("Parameters", elem_id="parameters"):
create_settings_menus(default_preset)
shared.input_params = [shared.gradio[k] for k in ['textbox', 'generate_state']]
shared.input_params = [shared.gradio[k] for k in ['textbox', 'interface_state']]
output_params = [shared.gradio[k] for k in ['output_textbox', 'markdown', 'html']]
gen_events.append(shared.gradio['Generate'].click(