Add model files and tokenizer configuration; include special tokens and initialization settings

main.py

@@ -1,21 +1,17 @@
 import os
-import shutil
-from typing import Optional, Dict, Any, Literal
+from typing import Optional, Dict, Any
 from enum import Enum
 from fastapi import FastAPI, HTTPException, status
 from pathlib import Path
 import logging
 import sys
-from pydantic import BaseModel, Field, validator
-from ctransformers import AutoModelForCausalLM
-from dataclasses import dataclass
+from pydantic import BaseModel, Field
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import json
 
-# Constants
-BASE_DIR = Path("/app")
-MODEL_DIR = BASE_DIR / "models"
-MODEL_NAME = "llama-2-7b-chat.q4_K_M.gguf"
-MODEL_PATH = MODEL_DIR / MODEL_NAME
-MODEL_URL = "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q4_K_M.gguf"
+# Define base model directory
+BASE_MODEL_DIR = "./models/"
 
 # Logging configuration
 logging.basicConfig(
@@ -23,292 +19,106 @@ logging.basicConfig(
     format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
     handlers=[
         logging.StreamHandler(sys.stdout),
-        logging.FileHandler('model_loading.log')
+        logging.FileHandler('poetry_generation.log')
     ]
 )
 logger = logging.getLogger(__name__)
 
-# Aligned with frontend enums
-class PoemStyle(str, Enum):
-    SONNET = "sonnet"
-    HAIKU = "haiku"
-    FREE_VERSE = "free-verse"
-    VILLANELLE = "villanelle"
-
-class EmotionalTone(str, Enum):
-    CONTEMPLATIVE = "contemplative"
-    JOYFUL = "joyful"
-    MELANCHOLIC = "melancholic"
-    ROMANTIC = "romantic"
-
-class Length(str, Enum):
-    SHORT = "short"
-    MEDIUM = "medium"
-    LONG = "long"
-
-@dataclass
-class StyleConfig:
-    """Maps style parameters to model parameters"""
-    temperature: float
-    top_p: float
-    top_k: int
-    repetition_penalty: float
-    max_tokens: int
-
-class StyleMapper:
-    """Maps style preferences to model parameters"""
-    
-    @staticmethod
-    def get_style_config(
-        style: PoemStyle,
-        emotional_tone: EmotionalTone,
-        creative_style: float,  # 0-100
-        language_variety: float,  # 0-1
-        length: Length,
-        word_repetition: float,  # 1-2
-    ) -> StyleConfig:
-        # Base configuration
-        config = {
-            "temperature": 0.7,
-            "top_p": 0.9,
-            "top_k": 40,
-            "repetition_penalty": 1.1,
-            "max_tokens": 512
-        }
-        
-        # Map creative_style (0-100) to temperature (0.5-1.0)
-        config["temperature"] = 0.5 + (creative_style / 100) * 0.5
-        
-        # Map length to tokens (assuming average word is 5 tokens)
-        length_token_map = {
-            Length.SHORT: 500,    # ~100 words
-            Length.MEDIUM: 1000,  # ~200 words
-            Length.LONG: 1500,    # ~300 words
-        }
-        config["max_tokens"] = length_token_map[length]
-        
-        # Map language_variety (0-1) to top_p
-        config["top_p"] = 0.7 + (language_variety * 0.3)
-        
-        # Map word_repetition (1-2) to repetition_penalty
-        config["repetition_penalty"] = word_repetition
-        
-        # Adjust based on emotional tone
-        tone_temp_adjustment = {
-            EmotionalTone.CONTEMPLATIVE: 0.0,
-            EmotionalTone.JOYFUL: 0.1,
-            EmotionalTone.MELANCHOLIC: -0.1,
-            EmotionalTone.ROMANTIC: 0.2
-        }
-        config["temperature"] += tone_temp_adjustment[emotional_tone]
-        
-        # Clamp temperature between 0.5 and 1.0
-        config["temperature"] = max(0.5, min(1.0, config["temperature"]))
-        
-        return StyleConfig(**config)
-
 class GenerateRequest(BaseModel):
-    prompt: str
-    style: PoemStyle
-    emotional_tone: EmotionalTone  # Use 'emotional_tone' directly
-    creative_style: float = Field(ge=0, le=100)  # 0-100 slider
-    language_variety: float = Field(ge=0, le=1)  # 0-1 slider
-    length: Length
-    word_repetition: float = Field(ge=1, le=2)  # 1-2 slider
-
-    @validator('creative_style')
-    def validate_creative_style(cls, v):
-        if not 0 <= v <= 100:
-            raise ValueError('creative_style must be between 0 and 100')
-        return v
-
-    @validator('language_variety')
-    def validate_language_variety(cls, v):
-        if not 0 <= v <= 1:
-            raise ValueError('language_variety must be between 0 and 1')
-        return v
-
-    @validator('word_repetition')
-    def validate_word_repetition(cls, v):
-        if not 1 <= v <= 2:
-            raise ValueError('word_repetition must be between 1 and 2')
-        return v
-
-
-    class Config:
-        allow_population_by_field_name = True
-
+    prompt: str = Field(..., min_length=1, max_length=500)
+    max_length: Optional[int] = Field(default=100, ge=10, le=500)
+    temperature: float = Field(default=0.9, ge=0.1, le=2.0)
+    top_k: int = Field(default=50, ge=1, le=100)
+    top_p: float = Field(default=0.95, ge=0.1, le=1.0)
+    repetition_penalty: float = Field(default=1.2, ge=1.0, le=2.0)
 
 class ModelManager:
     def __init__(self):
         self.model = None
+        self.tokenizer = None
         
-    def ensure_model_directory(self):
-        """Ensure the model directory exists and is writable"""
-        try:
-            MODEL_DIR.mkdir(parents=True, exist_ok=True)
-            
-            # Verify directory exists and is writable
-            if not MODEL_DIR.exists():
-                raise RuntimeError(f"Failed to create directory: {MODEL_DIR}")
-            if not os.access(MODEL_DIR, os.W_OK):
-                raise RuntimeError(f"Directory not writable: {MODEL_DIR}")
-                
-            logger.info(f"Model directory verified: {MODEL_DIR}")
-        except Exception as e:
-            logger.error(f"Error setting up model directory: {str(e)}")
-            raise
-    
-    async def initialize(self):
-        """Initialize the model with error handling"""
-        try:
-            # Ensure directory exists before attempting download
-            self.ensure_model_directory()
-            
-            if not MODEL_PATH.exists():
-                await self.download_model()
-            
-            self.model = self.initialize_model(MODEL_PATH)
-            return self.model is not None
-        except Exception as e:
-            logger.error(f"Initialization failed: {str(e)}")
-            return False
-    
-    @staticmethod
-    async def download_model():
-        """Download the model if it doesn't exist"""
-        import requests
-        from tqdm import tqdm
-        
-        if MODEL_PATH.exists():
-            logger.info(f"Model already exists at {MODEL_PATH}")
-            return
-        
-        # Create a temporary file for downloading
-        temp_path = MODEL_PATH.with_suffix('.temp')
-        
-        logger.info(f"Downloading model to temporary file: {temp_path}")
+    def initialize(self):
+        """Initialize the model and tokenizer"""
         try:
-            response = requests.get(MODEL_URL, stream=True)
-            response.raise_for_status()
-            total_size = int(response.headers.get('content-length', 0))
-            
-            # Ensure we have enough disk space
-            free_space = shutil.disk_usage(MODEL_DIR).free
-            if free_space < total_size * 1.1:  # 10% buffer
-                raise RuntimeError(
-                    f"Insufficient disk space. Need {total_size * 1.1 / (1024**3):.2f}GB, "
-                    f"have {free_space / (1024**3):.2f}GB"
-                )
+            logger.info("Loading tokenizer...")
             
-            # Download to temporary file first
-            with open(temp_path, 'wb') as file, tqdm(
-                desc="Downloading",
-                total=total_size,
-                unit='iB',
-                unit_scale=True,
-                unit_divisor=1024,
-            ) as pbar:
-                for data in response.iter_content(chunk_size=8192):
-                    size = file.write(data)
-                    pbar.update(size)
+            # First, let's try to load the base GPT-2 tokenizer
+            self.tokenizer = AutoTokenizer.from_pretrained("gpt2")
             
-            # Verify file size
-            if temp_path.stat().st_size != total_size:
-                raise RuntimeError(
-                    f"Downloaded file size ({temp_path.stat().st_size}) "
-                    f"doesn't match expected size ({total_size})"
-                )
-            
-            # Move temporary file to final location
-            temp_path.rename(MODEL_PATH)
-            logger.info(f"Model downloaded successfully to {MODEL_PATH}")
+            # Now customize it with your vocabulary if needed
+            vocab_path = os.path.join(BASE_MODEL_DIR, "vocab.json")
+            if os.path.exists(vocab_path):
+                try:
+                    with open(vocab_path, 'r', encoding='utf-8') as f:
+                        custom_vocab = json.load(f)
+                    self.tokenizer.vocab = custom_vocab
+                    self.tokenizer.ids_to_tokens = {v: k for k, v in custom_vocab.items()}
+                except Exception as e:
+                    logger.warning(f"Could not load custom vocabulary: {str(e)}")
             
-        except Exception as e:
-            logger.error(f"Error downloading model: {str(e)}")
-            # Clean up temporary file if it exists
-            if temp_path.exists():
-                temp_path.unlink()
-            # Clean up partial download if it exists
-            if MODEL_PATH.exists():
-                MODEL_PATH.unlink()
-            raise RuntimeError(f"Model download failed: {str(e)}")
-
-    def initialize_model(self, model_path: Path):
-        """Initialize the model with the specified configuration"""
-        try:
-            if not model_path.exists():
-                raise FileNotFoundError(f"Model file not found: {model_path}")
-                
-            if not model_path.is_file():
-                raise RuntimeError(f"Model path is not a file: {model_path}")
+            logger.info("Loading model...")
+            model_path = os.path.join(BASE_MODEL_DIR, "poeticagpt.pth")
+            if not os.path.exists(model_path):
+                logger.error(f"Model file not found at {model_path}")
+                return False
                 
-            if not os.access(model_path, os.R_OK):
-                raise RuntimeError(f"Model file is not readable: {model_path}")
+            # Load the model weights
+            self.model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
             
-            logger.info(f"Initializing model from: {model_path}")
-            model = AutoModelForCausalLM.from_pretrained(
-                str(model_path.parent),
-                model_file=model_path.name,
-                model_type="llama",
-                max_new_tokens=1500,
-                context_length=2048,
-                gpu_layers=0
-            )
+            # Force model to CPU
+            self.model.to('cpu')
+            self.model.eval()
             
-            if model is None:
-                raise RuntimeError("Model initialization returned None")
-                
-            logger.info("Model initialized successfully")
-            return model
+            logger.info("Model and tokenizer loaded successfully")
+            return True
             
         except Exception as e:
             logger.error(f"Error initializing model: {str(e)}")
-            return None
+            logger.exception("Detailed traceback:")
+            return False
 
     def generate(self, request: GenerateRequest) -> Dict[str, Any]:
-        """Generate text based on the request and style parameters"""
-        if self.model is None:
+        """Generate poetry based on the request parameters"""
+        if self.model is None or self.tokenizer is None:
             raise HTTPException(
                 status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
-                detail="Model not loaded"
+                detail="Model or tokenizer not loaded"
             )
         
-        # Get style configuration
-        style_config = StyleMapper.get_style_config(
-            request.style,
-            request.emotional_tone,
-            request.creative_style,
-            request.language_variety,
-            request.length,
-            request.word_repetition
-        )
-        
         try:
-            # Prepare prompt based on style
-            style_prompts = {
-                PoemStyle.SONNET: "Write a sonnet about",
-                PoemStyle.HAIKU: "Write a haiku about",
-                PoemStyle.FREE_VERSE: "Write a free verse poem about",
-                PoemStyle.VILLANELLE: "Write a villanelle about"
-            }
+            # Encode input
+            inputs = self.tokenizer.encode(request.prompt, return_tensors='pt')
+            attention_mask = torch.ones(inputs.shape, dtype=torch.long)
             
-            styled_prompt = f"{style_prompts[request.style]} {request.prompt}"
+            # Generate
+            with torch.no_grad():
+                outputs = self.model.generate(
+                    inputs,
+                    attention_mask=attention_mask,
+                    max_length=request.max_length,
+                    num_return_sequences=1,
+                    temperature=request.temperature,
+                    top_k=request.top_k,
+                    top_p=request.top_p,
+                    repetition_penalty=request.repetition_penalty,
+                    do_sample=True,
+                    pad_token_id=self.tokenizer.eos_token_id,
+                )
             
-            response = self.model(
-                styled_prompt,
-                max_new_tokens=style_config.max_tokens,
-                temperature=style_config.temperature,
-                top_p=style_config.top_p,
-                top_k=style_config.top_k,
-                repetition_penalty=style_config.repetition_penalty
-            )
+            generated_text = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
             
             return {
-                "generated_text": response,
-                "prompt": styled_prompt,
-                "style_config": style_config.__dict__
+                "generated_text": generated_text,
+                "prompt": request.prompt,
+                "parameters": {
+                    "max_length": request.max_length,
+                    "temperature": request.temperature,
+                    "top_k": request.top_k,
+                    "top_p": request.top_p,
+                    "repetition_penalty": request.repetition_penalty
+                }
             }
+            
         except Exception as e:
             logger.error(f"Error generating text: {str(e)}")
             raise HTTPException(
@@ -323,23 +133,28 @@ model_manager = ModelManager()
 @app.on_event("startup")
 async def startup():
     """Initialize the model during startup"""
-    await model_manager.initialize()
+    if not model_manager.initialize():
+        logger.error("Failed to initialize model manager")
+        sys.exit(1)
 
 @app.get("/health")
 async def health_check():
     """Health check endpoint"""
     return {
         "status": "healthy",
-        "model_loaded": model_manager.model is not None
+        "model_loaded": model_manager.model is not None,
+        "tokenizer_loaded": model_manager.tokenizer is not None
     }
 
 @app.post("/generate")
 async def generate_text(request: GenerateRequest):
-    """Generate text with style parameters"""
+    """Generate poetry with parameters"""
     return model_manager.generate(request)
 
 @app.on_event("shutdown")
 async def shutdown():
     """Cleanup on shutdown"""
     if model_manager.model is not None:
-        del model_manager.model
+        del model_manager.model
+    if model_manager.tokenizer is not None:
+        del model_manager.tokenizer

models/poeticagpt.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f77da9534fcf01b36f4780cd24ebe46e4d7f8740a1b17b66d5173d8694d6a62e
+size 139310252

models/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/tokenizer_config.json

@@ -0,0 +1,22 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

poetry_generation.log

@@ -0,0 +1,263 @@
+2024-11-16 23:03:13,077 - main - INFO - Loading tokenizer...
+2024-11-16 23:03:13,077 - main - ERROR - Error initializing model: Incorrect path_or_model_id: '\poetica'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:03:13,078 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:03:48,134 - main - INFO - Loading tokenizer...
+2024-11-16 23:03:48,135 - main - ERROR - Error initializing model: \ does not appear to have a file named config.json. Checkout 'https://huggingface.co/\/tree/None' for available files.
+2024-11-16 23:03:48,135 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:05:52,528 - main - INFO - Loading tokenizer...
+2024-11-16 23:05:52,530 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './models/tokenizer_config.json'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:05:52,530 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:06:20,012 - main - INFO - Loading tokenizer...
+2024-11-16 23:06:20,012 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './models/tokenizer_config.json'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:06:20,012 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:07:40,051 - main - INFO - Loading tokenizer...
+2024-11-16 23:07:40,135 - main - ERROR - Error initializing model: expected str, bytes or os.PathLike object, not NoneType
+2024-11-16 23:07:40,136 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 64, in initialize
+    self.tokenizer = AutoTokenizer.from_pretrained(
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\tokenization_auto.py", line 896, in from_pretrained
+    return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2291, in from_pretrained
+    return cls._from_pretrained(
+           ^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2329, in _from_pretrained
+    slow_tokenizer = (cls.slow_tokenizer_class)._from_pretrained(
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2525, in _from_pretrained
+    tokenizer = cls(*init_inputs, **init_kwargs)
+                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\gpt2\tokenization_gpt2.py", line 159, in __init__
+    with open(merges_file, encoding="utf-8") as merges_handle:
+         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+TypeError: expected str, bytes or os.PathLike object, not NoneType
+2024-11-16 23:07:40,163 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:07:56,229 - main - INFO - Loading tokenizer...
+2024-11-16 23:07:56,274 - main - ERROR - Error initializing model: expected str, bytes or os.PathLike object, not NoneType
+2024-11-16 23:07:56,275 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 64, in initialize
+    self.tokenizer = AutoTokenizer.from_pretrained(
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\tokenization_auto.py", line 896, in from_pretrained
+    return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
+           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2291, in from_pretrained
+    return cls._from_pretrained(
+           ^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2329, in _from_pretrained
+    slow_tokenizer = (cls.slow_tokenizer_class)._from_pretrained(
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_base.py", line 2525, in _from_pretrained
+    tokenizer = cls(*init_inputs, **init_kwargs)
+                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\gpt2\tokenization_gpt2.py", line 159, in __init__
+    with open(merges_file, encoding="utf-8") as merges_handle:
+         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+TypeError: expected str, bytes or os.PathLike object, not NoneType
+2024-11-16 23:07:56,280 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:09:15,013 - main - INFO - Loading tokenizer...
+2024-11-16 23:09:15,021 - main - ERROR - Error initializing model: expected `,` or `}` at line 2 column 18
+2024-11-16 23:09:15,021 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 80, in initialize
+    self.tokenizer = GPT2TokenizerFast(
+                     ^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\gpt2\tokenization_gpt2_fast.py", line 99, in __init__
+    super().__init__(
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_fast.py", line 115, in __init__
+    fast_tokenizer = TokenizerFast.from_file(fast_tokenizer_file)
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Exception: expected `,` or `}` at line 2 column 18
+2024-11-16 23:09:15,022 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:09:32,778 - main - INFO - Loading tokenizer...
+2024-11-16 23:09:32,778 - main - ERROR - Error initializing model: expected `,` or `}` at line 2 column 18
+2024-11-16 23:09:32,779 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 80, in initialize
+    self.tokenizer = GPT2TokenizerFast(
+                     ^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\gpt2\tokenization_gpt2_fast.py", line 99, in __init__
+    super().__init__(
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\tokenization_utils_fast.py", line 115, in __init__
+    fast_tokenizer = TokenizerFast.from_file(fast_tokenizer_file)
+                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+Exception: expected `,` or `}` at line 2 column 18
+2024-11-16 23:09:32,780 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:10:31,968 - main - INFO - Loading tokenizer...
+2024-11-16 23:10:31,978 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:10:32,575 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:10:32,579 - filelock - DEBUG - Attempting to acquire lock 3194343144000 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\be4d21d94f3b4687e5a54d84bf6ab46ed0f8defd.lock
+2024-11-16 23:10:32,580 - filelock - DEBUG - Lock 3194343144000 acquired on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\be4d21d94f3b4687e5a54d84bf6ab46ed0f8defd.lock
+2024-11-16 23:10:33,328 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "GET /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 26
+2024-11-16 23:10:33,362 - filelock - DEBUG - Attempting to release lock 3194343144000 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\be4d21d94f3b4687e5a54d84bf6ab46ed0f8defd.lock
+2024-11-16 23:10:33,363 - filelock - DEBUG - Lock 3194343144000 released on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\be4d21d94f3b4687e5a54d84bf6ab46ed0f8defd.lock
+2024-11-16 23:10:33,624 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/config.json HTTP/11" 200 0
+2024-11-16 23:10:33,631 - filelock - DEBUG - Attempting to acquire lock 3194374134560 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\10c66461e4c109db5a2196bff4bb59be30396ed8.lock
+2024-11-16 23:10:33,632 - filelock - DEBUG - Lock 3194374134560 acquired on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\10c66461e4c109db5a2196bff4bb59be30396ed8.lock
+2024-11-16 23:10:33,922 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "GET /gpt2/resolve/main/config.json HTTP/11" 200 665
+2024-11-16 23:10:33,926 - filelock - DEBUG - Attempting to release lock 3194374134560 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\10c66461e4c109db5a2196bff4bb59be30396ed8.lock
+2024-11-16 23:10:33,926 - filelock - DEBUG - Lock 3194374134560 released on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\10c66461e4c109db5a2196bff4bb59be30396ed8.lock
+2024-11-16 23:10:34,218 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/vocab.json HTTP/11" 200 0
+2024-11-16 23:10:34,219 - filelock - DEBUG - Attempting to acquire lock 3194378990896 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\1f1d9aaca301414e7f6c9396df506798ff4eb9a6.lock
+2024-11-16 23:10:34,220 - filelock - DEBUG - Lock 3194378990896 acquired on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\1f1d9aaca301414e7f6c9396df506798ff4eb9a6.lock
+2024-11-16 23:10:34,475 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "GET /gpt2/resolve/main/vocab.json HTTP/11" 200 1042301
+2024-11-16 23:10:35,729 - filelock - DEBUG - Attempting to release lock 3194378990896 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\1f1d9aaca301414e7f6c9396df506798ff4eb9a6.lock
+2024-11-16 23:10:35,729 - filelock - DEBUG - Lock 3194378990896 released on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\1f1d9aaca301414e7f6c9396df506798ff4eb9a6.lock
+2024-11-16 23:10:36,400 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/merges.txt HTTP/11" 200 0
+2024-11-16 23:10:36,402 - filelock - DEBUG - Attempting to acquire lock 3194378990896 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\226b0752cac7789c48f0cb3ec53eda48b7be36cc.lock
+2024-11-16 23:10:36,403 - filelock - DEBUG - Lock 3194378990896 acquired on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\226b0752cac7789c48f0cb3ec53eda48b7be36cc.lock
+2024-11-16 23:10:36,670 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "GET /gpt2/resolve/main/merges.txt HTTP/11" 200 456318
+2024-11-16 23:10:36,918 - filelock - DEBUG - Attempting to release lock 3194378990896 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\226b0752cac7789c48f0cb3ec53eda48b7be36cc.lock
+2024-11-16 23:10:36,919 - filelock - DEBUG - Lock 3194378990896 released on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\226b0752cac7789c48f0cb3ec53eda48b7be36cc.lock
+2024-11-16 23:10:37,180 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer.json HTTP/11" 200 0
+2024-11-16 23:10:37,183 - filelock - DEBUG - Attempting to acquire lock 3194378979184 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\4b988bccc9dc5adacd403c00b4704976196548f8.lock
+2024-11-16 23:10:37,184 - filelock - DEBUG - Lock 3194378979184 acquired on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\4b988bccc9dc5adacd403c00b4704976196548f8.lock
+2024-11-16 23:10:37,879 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "GET /gpt2/resolve/main/tokenizer.json HTTP/11" 200 1355256
+2024-11-16 23:10:39,156 - filelock - DEBUG - Attempting to release lock 3194378979184 on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\4b988bccc9dc5adacd403c00b4704976196548f8.lock
+2024-11-16 23:10:39,157 - filelock - DEBUG - Lock 3194378979184 released on C:\Users\asus\.cache\huggingface\hub\.locks\models--gpt2\4b988bccc9dc5adacd403c00b4704976196548f8.lock
+2024-11-16 23:10:39,435 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/added_tokens.json HTTP/11" 404 0
+2024-11-16 23:10:39,694 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/special_tokens_map.json HTTP/11" 404 0
+2024-11-16 23:10:39,884 - main - WARNING - Could not load custom vocabulary: property 'vocab' of 'GPT2TokenizerFast' object has no setter
+2024-11-16 23:10:39,884 - main - INFO - Loading model...
+2024-11-16 23:10:39,885 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './models\poeticagpt-quantized-new.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:10:39,885 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 402, in cached_file
+    resolved_file = hf_hub_download(
+                    ^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
+    validate_repo_id(arg_value)
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 160, in validate_repo_id
+    raise HFValidationError(
+huggingface_hub.errors.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: './models\poeticagpt-quantized-new.pth'.
+
+The above exception was the direct cause of the following exception:
+
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 66, in initialize
+    self.model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
+                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\auto_factory.py", line 485, in from_pretrained
+    resolved_config_file = cached_file(
+                           ^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 466, in cached_file
+    raise EnvironmentError(
+OSError: Incorrect path_or_model_id: './models\poeticagpt-quantized-new.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:10:39,916 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:11:46,212 - main - INFO - Loading tokenizer...
+2024-11-16 23:11:46,216 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:11:47,226 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:11:47,382 - main - WARNING - Could not load custom vocabulary: property 'vocab' of 'GPT2TokenizerFast' object has no setter
+2024-11-16 23:11:47,382 - main - INFO - Loading model...
+2024-11-16 23:11:47,383 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './models/poeticagpt-quantized-new.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:11:47,383 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 402, in cached_file
+    resolved_file = hf_hub_download(
+                    ^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
+    validate_repo_id(arg_value)
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 154, in validate_repo_id
+    raise HFValidationError(
+huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': './models/poeticagpt-quantized-new.pth'. Use `repo_type` argument if needed.
+
+The above exception was the direct cause of the following exception:
+
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 66, in initialize
+    self.model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
+                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\auto_factory.py", line 485, in from_pretrained
+    resolved_config_file = cached_file(
+                           ^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 466, in cached_file
+    raise EnvironmentError(
+OSError: Incorrect path_or_model_id: './models/poeticagpt-quantized-new.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:11:47,386 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:12:20,475 - main - INFO - Loading tokenizer...
+2024-11-16 23:12:20,478 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:12:21,010 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:12:21,169 - main - WARNING - Could not load custom vocabulary: property 'vocab' of 'GPT2TokenizerFast' object has no setter
+2024-11-16 23:12:21,169 - main - INFO - Loading model...
+2024-11-16 23:12:21,170 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './models/poeticagpt.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:12:21,170 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 402, in cached_file
+    resolved_file = hf_hub_download(
+                    ^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
+    validate_repo_id(arg_value)
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 154, in validate_repo_id
+    raise HFValidationError(
+huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': './models/poeticagpt.pth'. Use `repo_type` argument if needed.
+
+The above exception was the direct cause of the following exception:
+
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 66, in initialize
+    self.model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
+                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\auto_factory.py", line 485, in from_pretrained
+    resolved_config_file = cached_file(
+                           ^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 466, in cached_file
+    raise EnvironmentError(
+OSError: Incorrect path_or_model_id: './models/poeticagpt.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:12:21,173 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:13:01,527 - main - INFO - Loading tokenizer...
+2024-11-16 23:13:01,531 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:13:02,097 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:13:02,216 - main - INFO - Loading model...
+2024-11-16 23:13:02,217 - main - ERROR - Model file not found at poetica\models\poeticagpt.pth\poeticagpt.pth
+2024-11-16 23:13:02,217 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:13:08,762 - main - INFO - Loading tokenizer...
+2024-11-16 23:13:08,765 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:13:09,732 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:13:09,845 - main - INFO - Loading model...
+2024-11-16 23:13:09,846 - main - ERROR - Model file not found at poetica\models\poeticagpt.pth
+2024-11-16 23:13:09,846 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:13:33,649 - main - INFO - Loading tokenizer...
+2024-11-16 23:13:33,652 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:13:34,215 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:13:34,334 - main - INFO - Loading model...
+2024-11-16 23:13:34,335 - main - ERROR - Model file not found at .\poeticagpt.pth
+2024-11-16 23:13:34,335 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:14:10,849 - main - INFO - Loading tokenizer...
+2024-11-16 23:14:10,852 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:14:11,883 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:14:12,000 - main - INFO - Loading model...
+2024-11-16 23:14:12,001 - main - ERROR - Error initializing model: Incorrect path_or_model_id: './poeticagpt.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:14:12,001 - main - ERROR - Detailed traceback:
+Traceback (most recent call last):
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 402, in cached_file
+    resolved_file = hf_hub_download(
+                    ^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
+    validate_repo_id(arg_value)
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\huggingface_hub\utils\_validators.py", line 160, in validate_repo_id
+    raise HFValidationError(
+huggingface_hub.errors.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: './poeticagpt.pth'.
+
+The above exception was the direct cause of the following exception:
+
+Traceback (most recent call last):
+  File "E:\Self Work\My Projects\Poetica HuggingFace Server\poetica\main.py", line 66, in initialize
+    self.model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
+                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\models\auto\auto_factory.py", line 485, in from_pretrained
+    resolved_config_file = cached_file(
+                           ^^^^^^^^^^^^
+  File "e:\Self Work\My Projects\Poetica HuggingFace Server\.venv\Lib\site-packages\transformers\utils\hub.py", line 466, in cached_file
+    raise EnvironmentError(
+OSError: Incorrect path_or_model_id: './poeticagpt.pth'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
+2024-11-16 23:14:12,003 - main - ERROR - Failed to initialize model manager
+2024-11-16 23:14:22,432 - main - INFO - Loading tokenizer...
+2024-11-16 23:14:22,435 - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): huggingface.co:443
+2024-11-16 23:14:22,975 - urllib3.connectionpool - DEBUG - https://huggingface.co:443 "HEAD /gpt2/resolve/main/tokenizer_config.json HTTP/11" 200 0
+2024-11-16 23:14:23,132 - main - WARNING - Could not load custom vocabulary: property 'vocab' of 'GPT2TokenizerFast' object has no setter
+2024-11-16 23:14:23,132 - main - INFO - Loading model...
+2024-11-16 23:14:23,132 - main - ERROR - Model file not found at ./models/poeticagpt.pth
+2024-11-16 23:14:23,134 - main - ERROR - Failed to initialize model manager