import os
import pickle
from contextlib import nullcontext
import torch
import tiktoken
from model import GPTConfig, GPT
import gradio as gr

def nanogpt(start:str , max_new_tokens = 500, num_samples =2):


  # -----------------------------------------------------------------------------
  init_from = 'resume' # either 'resume' (from an out_dir) or a gpt2 variant (e.g. 'gpt2-xl')
  
  temperature = 0.8 # 1.0 = no change, < 1.0 = less random, > 1.0 = more random, in predictions
  top_k = 200 # retain only the top_k most likely tokens, clamp others to have 0 probability
  seed = 1337
  device = 'cpu' # examples: 'cpu', 'cuda', 'cuda:0', 'cuda:1', etc.
  dtype = 'bfloat16' if torch.cuda.is_available() and torch.cuda.is_bf16_supported() else 'float16' # 'float32' or 'bfloat16' or 'float16'
  compile = False # use PyTorch 2.0 to compile the model to be faster
  #exec(open('configurator.py').read()) # overrides from command line or config file
  # -----------------------------------------------------------------------------

  torch.manual_seed(seed)
  torch.cuda.manual_seed(seed)
  torch.backends.cuda.matmul.allow_tf32 = True # allow tf32 on matmul
  torch.backends.cudnn.allow_tf32 = True # allow tf32 on cudnn
  device_type = 'cuda' if 'cuda' in device else 'cpu' # for later use in torch.autocast
  ptdtype = {'float32': torch.float32, 'bfloat16': torch.bfloat16, 'float16': torch.float16}[dtype]
  ctx = nullcontext() if device_type == 'cpu' else torch.amp.autocast(device_type=device_type, dtype=ptdtype)

  # model
  if init_from == 'resume':
      # init from a model saved in a specific directory
      ckpt_path = 'ckpt.pt'
      checkpoint = torch.load(ckpt_path, map_location=device)
      gptconf = GPTConfig(**checkpoint['model_args'])
      model = GPT(gptconf)
      state_dict = checkpoint['model']
      unwanted_prefix = '_orig_mod.'
      for k,v in list(state_dict.items()):
          if k.startswith(unwanted_prefix):
              state_dict[k[len(unwanted_prefix):]] = state_dict.pop(k)
      model.load_state_dict(state_dict)
  

  model.eval()
  model.to(device)
  if compile:
      model = torch.compile(model) # requires PyTorch 2.0 (optional)

  # look for the meta pickle in case it is available in the dataset folder
  load_meta = False
  if init_from == 'resume' and 'config' in checkpoint and 'dataset' in checkpoint['config']: # older checkpoints might not have these...
      meta_path = os.path.join('data', checkpoint['config']['dataset'], 'meta.pkl')
      load_meta = os.path.exists(meta_path)
  if load_meta:
      print(f"Loading meta from {meta_path}...")
      with open(meta_path, 'rb') as f:
          meta = pickle.load(f)
      # TODO want to make this more general to arbitrary encoder/decoder schemes
      stoi, itos = meta['stoi'], meta['itos']
      encode = lambda s: [stoi[c] for c in s]
      decode = lambda l: ''.join([itos[i] for i in l])
  else:
      # ok let's assume gpt-2 encodings by default
      print("No meta.pkl found, assuming GPT-2 encodings...")
      enc = tiktoken.get_encoding("gpt2")
      encode = lambda s: enc.encode(s, allowed_special={"<|endoftext|>"})
      decode = lambda l: enc.decode(l)

  
  start_ids = encode(start)
  x = (torch.tensor(start_ids, dtype=torch.long, device=device)[None, ...])

  # run generation
  with torch.no_grad():
      with ctx:
          
              y = model.generate(x, max_new_tokens, temperature=temperature, top_k=top_k)
              #print(decode(y[0].tolist()))
              output = decode(y[0].tolist())
  return output

INTERFACE = gr.Interface(fn=nanogpt, inputs=[gr.Textbox(label= "Prompt", value= 'All that glisters is not gold.'),
                    gr.Slider(minimum = 300, maximum = 500, value= 300,  label= "Maximum number of tokens to be generated")] ,
                    outputs=gr.Text(label= "Generated Text"), title="NanoGPT",
                 description="NanoGPT is a transformer-based language model with only 10.65 million parameters, trained on a small dataset of Shakespeare work (size: 1MB only). It is trained with character level tokenization with a simple objective: predict the next char, given all of the previous chars within a text.",
                 examples = [['We know what we are, but know not what we may be',300],]
                        ).launch(debug=True)