File size: 6,608 Bytes
60be607
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import copy
from pathlib import Path

import gradio as gr
import torch
import yaml
from transformers import is_torch_xpu_available

from modules import shared

with open(Path(__file__).resolve().parent / '../css/NotoSans/stylesheet.css', 'r') as f:
    css = f.read()
with open(Path(__file__).resolve().parent / '../css/main.css', 'r') as f:
    css += f.read()
with open(Path(__file__).resolve().parent / '../js/main.js', 'r') as f:
    js = f.read()
with open(Path(__file__).resolve().parent / '../js/save_files.js', 'r') as f:
    save_files_js = f.read()
with open(Path(__file__).resolve().parent / '../js/switch_tabs.js', 'r') as f:
    switch_tabs_js = f.read()
with open(Path(__file__).resolve().parent / '../js/show_controls.js', 'r') as f:
    show_controls_js = f.read()
with open(Path(__file__).resolve().parent / '../js/update_big_picture.js', 'r') as f:
    update_big_picture_js = f.read()

refresh_symbol = '🔄'
delete_symbol = '🗑️'
save_symbol = '💾'

theme = gr.themes.Default(
    font=['Noto Sans', 'Helvetica', 'ui-sans-serif', 'system-ui', 'sans-serif'],
    font_mono=['IBM Plex Mono', 'ui-monospace', 'Consolas', 'monospace'],
).set(
    border_color_primary='#c5c5d2',
    button_large_padding='6px 12px',
    body_text_color_subdued='#484848',
    background_fill_secondary='#eaeaea'
)

if Path("notification.mp3").exists():
    audio_notification_js = "document.querySelector('#audio_notification audio')?.play();"
else:
    audio_notification_js = ""


def list_model_elements():
    elements = [
        'loader',
        'filter_by_loader',
        'cpu_memory',
        'auto_devices',
        'disk',
        'cpu',
        'bf16',
        'load_in_8bit',
        'trust_remote_code',
        'no_use_fast',
        'use_flash_attention_2',
        'load_in_4bit',
        'compute_dtype',
        'quant_type',
        'use_double_quant',
        'wbits',
        'groupsize',
        'model_type',
        'pre_layer',
        'triton',
        'desc_act',
        'no_inject_fused_attention',
        'no_inject_fused_mlp',
        'no_use_cuda_fp16',
        'disable_exllama',
        'disable_exllamav2',
        'cfg_cache',
        'no_flash_attn',
        'num_experts_per_token',
        'cache_8bit',
        'threads',
        'threads_batch',
        'n_batch',
        'no_mmap',
        'mlock',
        'no_mul_mat_q',
        'n_gpu_layers',
        'tensor_split',
        'n_ctx',
        'gpu_split',
        'max_seq_len',
        'compress_pos_emb',
        'alpha_value',
        'rope_freq_base',
        'numa',
        'logits_all',
        'no_offload_kqv',
        'tensorcores',
        'hqq_backend',
    ]
    if is_torch_xpu_available():
        for i in range(torch.xpu.device_count()):
            elements.append(f'gpu_memory_{i}')
    else:
        for i in range(torch.cuda.device_count()):
            elements.append(f'gpu_memory_{i}')

    return elements


def list_interface_input_elements():
    elements = [
        'max_new_tokens',
        'auto_max_new_tokens',
        'max_tokens_second',
        'seed',
        'temperature',
        'temperature_last',
        'top_p',
        'min_p',
        'top_k',
        'typical_p',
        'epsilon_cutoff',
        'eta_cutoff',
        'repetition_penalty',
        'presence_penalty',
        'frequency_penalty',
        'repetition_penalty_range',
        'encoder_repetition_penalty',
        'no_repeat_ngram_size',
        'min_length',
        'do_sample',
        'penalty_alpha',
        'num_beams',
        'length_penalty',
        'early_stopping',
        'mirostat_mode',
        'mirostat_tau',
        'mirostat_eta',
        'grammar_string',
        'negative_prompt',
        'guidance_scale',
        'add_bos_token',
        'ban_eos_token',
        'custom_token_bans',
        'truncation_length',
        'custom_stopping_strings',
        'skip_special_tokens',
        'stream',
        'tfs',
        'top_a',
    ]

    # Chat elements
    elements += [
        'textbox',
        'start_with',
        'character_menu',
        'history',
        'name1',
        'name2',
        'greeting',
        'context',
        'mode',
        'custom_system_message',
        'instruction_template_str',
        'chat_template_str',
        'chat_style',
        'chat-instruct_command',
    ]

    # Notebook/default elements
    elements += [
        'textbox-notebook',
        'textbox-default',
        'output_textbox',
        'prompt_menu-default',
        'prompt_menu-notebook',
    ]

    # Model elements
    elements += list_model_elements()

    return elements


def gather_interface_values(*args):
    output = {}
    for i, element in enumerate(list_interface_input_elements()):
        output[element] = args[i]

    if not shared.args.multi_user:
        shared.persistent_interface_state = output

    return output


def apply_interface_values(state, use_persistent=False):
    if use_persistent:
        state = shared.persistent_interface_state

    elements = list_interface_input_elements()
    if len(state) == 0:
        return [gr.update() for k in elements]  # Dummy, do nothing
    else:
        return [state[k] if k in state else gr.update() for k in elements]


def save_settings(state, preset, extensions, show_controls):
    output = copy.deepcopy(shared.settings)
    exclude = ['name2', 'greeting', 'context', 'turn_template']
    for k in state:
        if k in shared.settings and k not in exclude:
            output[k] = state[k]

    output['preset'] = preset
    output['prompt-default'] = state['prompt_menu-default']
    output['prompt-notebook'] = state['prompt_menu-notebook']
    output['character'] = state['character_menu']
    output['default_extensions'] = extensions
    output['seed'] = int(output['seed'])
    output['show_controls'] = show_controls

    return yaml.dump(output, sort_keys=False, width=float("inf"))


def create_refresh_button(refresh_component, refresh_method, refreshed_args, elem_class, interactive=True):
    """
    Copied from https://github.com/AUTOMATIC1111/stable-diffusion-webui
    """
    def refresh():
        refresh_method()
        args = refreshed_args() if callable(refreshed_args) else refreshed_args

        for k, v in args.items():
            setattr(refresh_component, k, v)

        return gr.update(**(args or {}))

    refresh_button = gr.Button(refresh_symbol, elem_classes=elem_class, interactive=interactive)
    refresh_button.click(
        fn=refresh,
        inputs=[],
        outputs=[refresh_component]
    )

    return refresh_button