Merge pull request #474 from ebolam/Model_Plugins

TTS update and double click to load story

aiserver.py

@@ -1477,7 +1477,7 @@ def general_startup(override_args=None):
     parser.add_argument("--cacheonly", action='store_true', help="Does not save the model to the models folder when it has been downloaded in the cache")
     parser.add_argument("--customsettings", help="Preloads arguements from json file. You only need to provide the location of the json file. Use customsettings.json template file. It can be renamed if you wish so that you can store multiple configurations. Leave any settings you want as default as null. Any values you wish to set need to be in double quotation marks")
     parser.add_argument("--no_ui", action='store_true', default=False, help="Disables the GUI and Socket.IO server while leaving the API server running.")
-    parser.add_argument("--summarizer_model", action='store', default="philschmid/bart-large-cnn-samsum", help="Huggingface model to use for summarization. Defaults to sshleifer/distilbart-cnn-12-6")
+    parser.add_argument("--summarizer_model", action='store', default="pszemraj/led-large-book-summary", help="Huggingface model to use for summarization. Defaults to pszemraj/led-large-book-summary")
     parser.add_argument("--max_summary_length", action='store', default=75, help="Maximum size for summary to send to image generation")
     parser.add_argument("--multi_story", action='store_true', default=False, help="Allow multi-story mode (experimental)")
     parser.add_argument("--peft", type=str, help="Specify the path or HuggingFace ID of a Peft to load it. Not supported on TPU. (Experimental)") 
@@ -7586,8 +7586,11 @@ def text2img_api(prompt, art_guide="") -> Image.Image:
 @socketio.on("clear_generated_image")
 @logger.catch
 def UI2_clear_generated_image(data):
-    koboldai_vars.picture = ""
-    koboldai_vars.picture_prompt = ""
+    if 'action_id' in data and data['action_id'] is not None:
+        koboldai_vars.actions.clear_picture(data['action_id'])
+    else:
+        koboldai_vars.picture = ""
+        koboldai_vars.picture_prompt = ""
 
 #==================================================================#
 # Retrieve previous images
@@ -7600,7 +7603,9 @@ def UI_2_get_story_image(data):
     print(filename)
     if filename is not None:
         with open(filename, "rb") as image_file:
-            return base64.b64encode(image_file.read()).decode("utf-8") 
+            return {'img': base64.b64encode(image_file.read()).decode("utf-8"), 'action_id': action_id}
+    else:
+        return {'img': None, 'action_id': action_id}
 
 #@logger.catch
 def get_items_locations_from_text(text):
@@ -7651,16 +7656,19 @@ def get_items_locations_from_text(text):
 #==================================================================#
 def summarize(text, max_length=100, min_length=30, unload=True):
     from transformers import pipeline as summary_pipeline
+    from transformers import AutoConfig
     start_time = time.time()
     if koboldai_vars.summarizer is None:
         if os.path.exists("functional_models/{}".format(args.summarizer_model.replace('/', '_'))):
             koboldai_vars.summary_tokenizer = AutoTokenizer.from_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
             koboldai_vars.summarizer = AutoModelForSeq2SeqLM.from_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
+            koboldai_vars.summary_model_config = AutoConfig.from_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
         else:
             koboldai_vars.summary_tokenizer = AutoTokenizer.from_pretrained(args.summarizer_model, cache_dir="cache")
             koboldai_vars.summarizer = AutoModelForSeq2SeqLM.from_pretrained(args.summarizer_model, cache_dir="cache")
             koboldai_vars.summary_tokenizer.save_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), max_shard_size="500MiB")
             koboldai_vars.summarizer.save_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), max_shard_size="500MiB")
+            koboldai_vars.summary_model_config = AutoConfig.from_pretrained(args.summarizer_model, cache_dir="cache")
 
     #Try GPU accel
     if koboldai_vars.hascuda and torch.cuda.get_device_properties(0).total_memory - torch.cuda.memory_reserved(0) >= 1645778560:
@@ -7674,9 +7682,27 @@ def summarize(text, max_length=100, min_length=30, unload=True):
     #Actual sumarization
     start_time = time.time()
     #make sure text is less than 1024 tokens, otherwise we'll crash
-    if len(koboldai_vars.summary_tokenizer.encode(text)) > 1000:
-        text = koboldai_vars.summary_tokenizer.decode(koboldai_vars.summary_tokenizer.encode(text)[:1000])
-    output = tpool.execute(summarizer, text, max_length=max_length, min_length=min_length, do_sample=False)[0]['summary_text']
+    max_tokens = koboldai_vars.summary_model_config.max_encoder_position_embeddings if hasattr(koboldai_vars.summary_model_config, 'max_encoder_position_embeddings') else 1024
+    logger.info("Using max summary tokens of {}".format(max_tokens))
+    if len(koboldai_vars.summary_tokenizer.encode(text)) > max_tokens:
+        text_list = koboldai_vars.actions.sentence_re.findall(text)
+        i=0
+        while i <= len(text_list)-2:
+            if len(koboldai_vars.summary_tokenizer.encode(text_list[i] + text_list[i+1])) < max_tokens:
+                text_list[i] = text_list[i] + text_list[i+1]
+                del text_list[i+1]
+            else:
+                i+=1
+
+
+    else:
+        text_list = [text]
+
+    output = []
+    logger.info("Summarizing with {} chunks of length {}".format(len(text_list), [len(koboldai_vars.summary_tokenizer.encode(x)) for x in text_list]))
+    for text in text_list:
+        output.append(tpool.execute(summarizer, text, max_length=max_length, min_length=min_length, do_sample=False)[0]['summary_text'])
+    output = " ".join(output)
     logger.debug("Time to summarize: {}".format(time.time()-start_time))
     #move model back to CPU to save precious vram
     torch.cuda.empty_cache()
@@ -7696,40 +7722,33 @@ def summarize(text, max_length=100, min_length=30, unload=True):
 @socketio.on("refresh_auto_memory")
 @logger.catch
 def UI_2_refresh_auto_memory(data):
+    max_output_length=500
+    from transformers import AutoConfig
     koboldai_vars.auto_memory = "Generating..."
     if koboldai_vars.summary_tokenizer is None:
-        if os.path.exists("models/{}".format(args.summarizer_model.replace('/', '_'))):
-            koboldai_vars.summary_tokenizer = AutoTokenizer.from_pretrained("models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
+        if os.path.exists("functional_models/{}".format(args.summarizer_model.replace('/', '_'))):
+            koboldai_vars.summary_tokenizer = AutoTokenizer.from_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
+            koboldai_vars.summary_model_config = AutoConfig.from_pretrained("functional_models/{}".format(args.summarizer_model.replace('/', '_')), cache_dir="cache")
         else:
             koboldai_vars.summary_tokenizer = AutoTokenizer.from_pretrained(args.summarizer_model, cache_dir="cache")
-    #first, let's get all of our game text and split it into sentences
-    sentences = [x[0] for x in koboldai_vars.actions.to_sentences()]
-    sentences_lengths = [len(koboldai_vars.summary_tokenizer.encode(x)) for x in sentences]
+            koboldai_vars.summary_model_config = AutoConfig.from_pretrained(args.summarizer_model, cache_dir="cache")
+    max_tokens = koboldai_vars.summary_model_config.max_encoder_position_embeddings if hasattr(koboldai_vars.summary_model_config, 'max_encoder_position_embeddings') else 1024
+
+    #first, let's get all of our game text
+    sentences = "".join([x[0] for x in koboldai_vars.actions.to_sentences()])
 
     pass_number = 1
-    while len(koboldai_vars.summary_tokenizer.encode("".join(sentences))) > 1000:
-        #Now let's split them into 1000 token chunks
-        summary_chunks = [""]
-        summary_chunk_lengths = [0]
-        for i in range(len(sentences)):
-            if summary_chunk_lengths[-1] + sentences_lengths[i] <= 1000:
-                summary_chunks[-1] += sentences[i]
-                summary_chunk_lengths[-1] += sentences_lengths[i]
-            else:
-                summary_chunks.append(sentences[i])
-                summary_chunk_lengths.append(sentences_lengths[i])
-        new_sentences = []
-        i=0
-        for summary_chunk in summary_chunks:
-            logger.debug("summarizing chunk {}".format(i))
-            new_sentences.extend(re.split("(?<=[.!?])\s+", summarize(summary_chunk, unload=False)))
-            i+=1
+    while len(koboldai_vars.summary_tokenizer.encode(sentences)) > max_tokens:
+        new_sentences = summarize(sentences, unload=False, max_length=max_output_length)
         logger.debug("Pass {}:\nSummarized to {} sentencees from {}".format(pass_number, len(new_sentences), len(sentences)))
         sentences = new_sentences
-        koboldai_vars.auto_memory += "Pass {}:\n{}\n\n".format(pass_number, "\n".join(sentences))
+        koboldai_vars.auto_memory += "Pass {}:\n{}\n\n".format(pass_number, sentences)
         pass_number+=1
     logger.debug("OK, doing final summarization")
-    output = summarize(" ".join(sentences))
+    if len(koboldai_vars.summary_tokenizer.encode(sentences)) > max_output_length:
+        output = summarize(sentences, max_length=max_output_length)
+    else:
+        output = sentences
     koboldai_vars.auto_memory += "\n\n Final Result:\n" + output
 
 

koboldai_settings.py

@@ -687,8 +687,9 @@ def send_to_ui(self):
 
 class model_settings(settings):
     local_only_variables = ['apikey', 'default_preset']
-    no_save_variables = ['modelconfig', 'custmodpth', 'generated_tkns', 
-                         'loaded_layers', 'total_layers', 'total_download_chunks', 'downloaded_chunks', 'presets', 'default_preset', 
+    no_save_variables = ['modelconfig', 'custmodpth', 'generated_tkns',
+                         'loaded_layers', 'total_layers', 'loaded_checkpoints', 'total_checkpoints',
+                         'total_download_chunks', 'downloaded_chunks', 'presets', 'default_preset',
                          'welcome', 'welcome_default', 'simple_randomness', 'simple_creativity', 'simple_repitition',
                          'badwordsids', 'uid_presets', 'model', 'model_type', 'lazy_load', 'fp32_model', 'modeldim', 'horde_wait_time', 'horde_queue_position', 'horde_queue_size', 'newlinemode', 'tqdm_progress', 'tqdm_rem_time', '_tqdm']
     settings_name = "model"
@@ -705,6 +706,8 @@ def __init__(self, socketio, koboldai_vars):
         self.generated_tkns = 0    # If using a backend that supports Lua generation modifiers, how many tokens have already been generated, otherwise 0
         self.loaded_layers = 0     # Used in UI 2 to show model loading progress
         self.total_layers = 0      # Same as above
+        self.loaded_checkpoints = 0
+        self.total_checkpoints = 1
         self.total_download_chunks = 0 # tracks how much of the model has downloaded for the UI 2
         self.downloaded_chunks = 0 #as above
         self._tqdm        = tqdm.tqdm(total=self.genamt, file=self.ignore_tqdm())    # tqdm agent for generating tokens. This will allow us to calculate the remaining time
@@ -829,13 +832,22 @@ def __setattr__(self, name, value):
         #Setup TQDP for model loading
         elif name == "loaded_layers" and '_tqdm' in self.__dict__:
             if value == 0:
-                self._tqdm.reset(total=self.total_layers)
+                self._tqdm.reset(total=self.total_layers if self.total_checkpoints == 1 else 1000)
                 self.tqdm_progress = 0
             else:
-                self._tqdm.update(1)
-                self.tqdm_progress = int(float(self.loaded_layers)/float(self.total_layers)*100)
+                if self.total_checkpoints == 1:
+                    self._tqdm.update(1)
+                elif self.total_layers != 0 and self.total_checkpoints != 0:
+                    proper_progress = (self.loaded_checkpoints + value/self.total_layers)/self.total_checkpoints*1000
+                    self._tqdm.update(proper_progress - self._tqdm.n)
+
+                self.tqdm_progress = int(float(self._tqdm.n)/float(self._tqdm.total)*100)
+
                 if self._tqdm.format_dict['rate'] is not None:
-                    self.tqdm_rem_time = str(datetime.timedelta(seconds=int(float(self.total_layers-self.loaded_layers)/self._tqdm.format_dict['rate'])))  
+                    elapsed = self._tqdm.format_dict["elapsed"]
+                    rate = self._tqdm.format_dict["rate"]
+                    remaining = (self._tqdm.total - self._tqdm.n) / rate if rate and self._tqdm.total else 0
+                    self.tqdm_rem_time = str(datetime.timedelta(seconds=remaining))
         #Setup TQDP for model downloading
         elif name == "total_download_chunks" and '_tqdm' in self.__dict__:
             self._tqdm.reset(total=value)
@@ -1247,11 +1259,11 @@ def __setattr__(self, name, value):
 class system_settings(settings):
     local_only_variables = ['lua_state', 'lua_logname', 'lua_koboldbridge', 'lua_kobold', 
                             'lua_koboldcore', 'regex_sl', 'acregex_ai', 'acregex_ui', 'comregex_ai', 'comregex_ui',
-                            'sp', '_horde_pid', 'inference_config', 'image_pipeline', 
+                            'sp', '_horde_pid', 'inference_config', 'image_pipeline', 'summary_model_config',
                             'summarizer', 'summary_tokenizer', 'tts_model', 'rng_states', 'comregex_ai', 'comregex_ui', 'colab_arg']
-    no_save_variables = ['lua_state', 'lua_logname', 'lua_koboldbridge', 'lua_kobold', 
+    no_save_variables = ['lua_state', 'lua_logname', 'lua_koboldbridge', 'lua_kobold', 'summary_model_config',
                          'lua_koboldcore', 'sp', 'sp_length', '_horde_pid', 'horde_share', 'aibusy', 
-                         'serverstarted', 'inference_config', 'image_pipeline', 'summarizer', 'on_colab'
+                         'serverstarted', 'inference_config', 'image_pipeline', 'summarizer', 'on_colab', 'quiet',
                          'summary_tokenizer', 'use_colab_tpu', 'noai', 'disable_set_aibusy', 'cloudflare_link', 'tts_model',
                          'generating_image', 'bit_8_available', 'host', 'hascuda', 'usegpu', 'rng_states', 'comregex_ai', 'comregex_ui', 'git_repository', 'git_branch', 'colab_arg',
                          'disable_model_load']
@@ -1335,6 +1347,7 @@ def __init__(self, socketio, koboldai_var):
         self.image_pipeline = None
         self.summarizer = None
         self.summary_tokenizer = None
+        self.summary_model_config = {}
         self.keep_img_gen_in_memory = False
         self.cookies = {} #cookies for colab since colab's URL changes, cookies are lost
         self.experimental_features = False
@@ -2233,6 +2246,12 @@ def set_picture(self, action_id, filename, prompt):
             self.actions[action_id]['picture_filename'] = filename
             self.actions[action_id]['picture_prompt'] = prompt
 
+    def clear_picture(self, action_id):
+        action_id = int(action_id)
+        if action_id in self.actions:
+            del self.actions[action_id]['picture_filename']
+            del self.actions[action_id]['picture_prompt']
+
     def get_picture(self, action_id):
         if action_id == -1:
             if self.story_settings.prompt_picture_filename == "":

modeling/inference_models/exllama/class.py

@@ -340,7 +340,8 @@ def _raw_generate(
 
             self._post_token_gen(self.generator.sequence)
 
-            utils.koboldai_vars.generated_tkns += 1
+            #This is taken care of in the core stopper class that's called below. If you're not using core stoppers then it should remain here
+            #utils.koboldai_vars.generated_tkns += 1
 
             # Apply stoppers
             do_stop = False
@@ -374,6 +375,12 @@ def _get_tokenizer(self, location: str):
         return tokenizer
 
     def get_requested_parameters(self, model_name, model_path, menu_path, parameters = {}):
+        saved_data = {'layers': [], 'max_ctx': 2048, 'compress_emb': 1, 'ntk_alpha': 1}
+        if os.path.exists("settings/{}.exllama.model_backend.settings".format(model_name.replace("/", "_"))) and 'base_url' not in vars(self):
+            with open("settings/{}.exllama.model_backend.settings".format(model_name.replace("/", "_")), "r") as f:
+                temp = json.load(f)
+                for key in temp:
+                    saved_data[key] = temp[key]
         requested_parameters = []
         gpu_count = torch.cuda.device_count()
         layer_count = self.model_config["n_layer"] if isinstance(self.model_config, dict) else self.model_config.num_layers if hasattr(self.model_config, "num_layers") else self.model_config.n_layer if hasattr(self.model_config, "n_layer") else self.model_config.num_hidden_layers if hasattr(self.model_config, 'num_hidden_layers') else None
@@ -400,7 +407,7 @@ def get_requested_parameters(self, model_name, model_path, menu_path, parameters
                                             "step": 1,
                                             "check": {"sum": ["{}_Layers".format(i) for i in range(gpu_count)], "value": layer_count, 'check': "="},
                                             "check_message": "The sum of assigned layers must equal {}".format(layer_count),
-                                            "default": [layer_count if i == 0 else 0],
+                                            "default": saved_data['layers'][i] if len(saved_data['layers']) > i else layer_count if i==0 else 0,
                                             "tooltip": "The number of layers to put on {}.".format(torch.cuda.get_device_name(i)),
                                             "menu_path": "Layers",
                                             "extra_classes": "",
@@ -415,7 +422,7 @@ def get_requested_parameters(self, model_name, model_path, menu_path, parameters
             "min": 2048,
             "max": 16384,
             "step": 512,
-            "default": 2048,
+            "default": saved_data['max_ctx'],
             "tooltip": "The maximum context size the model supports",
             "menu_path": "Configuration",
             "extra_classes": "",
@@ -430,7 +437,7 @@ def get_requested_parameters(self, model_name, model_path, menu_path, parameters
             "min": 1,
             "max": 8,
             "step": 0.25,
-            "default": 1,
+            "default": saved_data['compress_emb'],
             "tooltip": "If the model requires compressed embeddings, set them here",
             "menu_path": "Configuration",
             "extra_classes": "",
@@ -445,7 +452,7 @@ def get_requested_parameters(self, model_name, model_path, menu_path, parameters
             "min": 1,
             "max": 32,
             "step": 0.25,
-            "default": 1,
+            "default": saved_data['ntk_alpha'],
             "tooltip": "NTK alpha value",
             "menu_path": "Configuration",
             "extra_classes": "",
@@ -490,3 +497,21 @@ def set_input_parameters(self, parameters):
 
         self.model_name = parameters['custom_model_name'] if 'custom_model_name' in parameters else parameters['id']
         self.path = parameters['path'] if 'path' in parameters else None
+
+    def _save_settings(self):
+        with open(
+            "settings/{}.exllama.model_backend.settings".format(
+                self.model_name.replace("/", "_")
+            ),
+            "w",
+        ) as f:
+            json.dump(
+                {
+                    "layers": self.layers if "layers" in vars(self) else [],
+                    "max_ctx": self.model_config.max_seq_len,
+                    "compress_emb": self.model_config.compress_pos_emb,
+                    "ntk_alpha": self.model_config.alpha_value
+                },
+                f,
+                indent="",
+            )