import sys, os
import torch as t
from data import get_data
from model import PoetryModel
from torch import nn
from torch.autograd import Variable
from utils import Visualizer
import tqdm
from torchnet import meter
import ipdb
 
class Config(object):
	data_path = 'data/'
	pickle_path = 'tang.npz'
	author = None
	constrain = None
	category = 'poet.tang' #or poet.song
	lr = 1e-3
	weight_decay = 1e-4
	use_gpu = True
	epoch = 20
	batch_size = 128
	maxlen = 125
	plot_every = 20
	#use_env = True #是否使用visodm
	env = 'poety' 
	#visdom env
	max_gen_len = 200
	debug_file = '/tmp/debugp'
	model_path = None
	prefix_words = '細(xì)雨魚兒出，微風(fēng)燕子斜。' 
	#不是詩歌組成部分，是意境
	start_words = '閑云潭影日悠悠' 
	#詩歌開始
	acrostic = False 
	#是否藏頭
	model_prefix = 'checkpoints/tang' 
	#模型保存路徑
opt = Config()
 
def generate(model, start_words, ix2word, word2ix, prefix_words=None):
	'''
	給定幾個(gè)詞，根據(jù)這幾個(gè)詞接著生成一首完整的詩歌
	'''
	results = list(start_words)
	start_word_len = len(start_words)
	# 手動設(shè)置第一個(gè)詞為<START>
	# 這個(gè)地方有問題，最后需要再看一下
	input = Variable(t.Tensor([word2ix['<START>']]).view(1,1).long())
	if opt.use_gpu:input=input.cuda()
	hidden = None
	
	if prefix_words:
		for word in prefix_words:
			output,hidden = model(input,hidden)
			# 下邊這句話是為了把input變成1*1？
			input = Variable(input.data.new([word2ix[word]])).view(1,1)
	for i in range(opt.max_gen_len):
		output,hidden = model(input,hidden)
		
		if i<start_word_len:
			w = results[i]
			input = Variable(input.data.new([word2ix[w]])).view(1,1)
		else:
			top_index = output.data[0].topk(1)[1][0]
			w = ix2word[top_index]
			results.append(w)
			input = Variable(input.data.new([top_index])).view(1,1)
		if w=='<EOP>':
			del results[-1] #-1的意思是倒數(shù)第一個(gè)
			break
	return results
 
def gen_acrostic(model,start_words,ix2word,word2ix, prefix_words = None):
 '''
 生成藏頭詩
 start_words : u'深度學(xué)習(xí)'
 生成：
 深木通中岳，青苔半日脂。
 度山分地險(xiǎn)，逆浪到南巴。
 學(xué)道兵猶毒，當(dāng)時(shí)燕不移。
 習(xí)根通古岸，開鏡出清羸。
 '''
 results = []
 start_word_len = len(start_words)
 input = Variable(t.Tensor([word2ix['<START>']]).view(1,1).long())
 if opt.use_gpu:input=input.cuda()
 hidden = None
 
 index=0 # 用來指示已經(jīng)生成了多少句藏頭詩
 # 上一個(gè)詞
 pre_word='<START>'
 
 if prefix_words:
  for word in prefix_words:
   output,hidden = model(input,hidden)
   input = Variable(input.data.new([word2ix[word]])).view(1,1)
 
 for i in range(opt.max_gen_len):
  output,hidden = model(input,hidden)
  top_index = output.data[0].topk(1)[1][0]
  w = ix2word[top_index]
 
  if (pre_word in {u'。',u'！','<START>'} ):
   # 如果遇到句號，藏頭的詞送進(jìn)去生成
 
   if index==start_word_len:
    # 如果生成的詩歌已經(jīng)包含全部藏頭的詞，則結(jié)束
    break
   else: 
    # 把藏頭的詞作為輸入送入模型
    w = start_words[index]
    index+=1
    input = Variable(input.data.new([word2ix[w]])).view(1,1) 
  else:
   # 否則的話，把上一次預(yù)測是詞作為下一個(gè)詞輸入
   input = Variable(input.data.new([word2ix[w]])).view(1,1)
  results.append(w)
  pre_word = w
 return results
 
def train(**kwargs):
	
	for k,v in kwargs.items():
		setattr(opt,k,v) #設(shè)置apt里屬性的值
	vis = Visualizer(env=opt.env)
	
	#獲取數(shù)據(jù)
	data, word2ix, ix2word = get_data(opt) #get_data是data.py里的函數(shù)
	data = t.from_numpy(data)
	#這個(gè)地方出錯(cuò)了，是大寫的L
	dataloader = t.utils.data.DataLoader(data, 
					batch_size = opt.batch_size,
					shuffle = True,
					num_workers = 1) #在python里，這樣寫程序可以嗎？
 #模型定義
	model = PoetryModel(len(word2ix), 128, 256)
	optimizer = t.optim.Adam(model.parameters(), lr=opt.lr)
	criterion = nn.CrossEntropyLoss()
 
	if opt.model_path:
		model.load_state_dict(t.load(opt.model_path))
	if opt.use_gpu:
		model.cuda()
		criterion.cuda()
		
	#The tnt.AverageValueMeter measures and returns the average value 
	#and the standard deviation of any collection of numbers that are 
	#added to it. It is useful, for instance, to measure the average 
	#loss over a collection of examples.
 
 #The add() function expects as input a Lua number value, which 
 #is the value that needs to be added to the list of values to 
 #average. It also takes as input an optional parameter n that 
 #assigns a weight to value in the average, in order to facilitate 
 #computing weighted averages (default = 1).
 
 #The tnt.AverageValueMeter has no parameters to be set at initialization time. 
	loss_meter = meter.AverageValueMeter()
	
	for epoch in range(opt.epoch):
		loss_meter.reset()
		for ii,data_ in tqdm.tqdm(enumerate(dataloader)):
			#tqdm是python中的進(jìn)度條
			#訓(xùn)練
			data_ = data_.long().transpose(1,0).contiguous()
			#上邊一句話，把data_變成long類型，把1維和0維轉(zhuǎn)置，把內(nèi)存調(diào)成連續(xù)的
			if opt.use_gpu: data_ = data_.cuda()
			optimizer.zero_grad()
			input_, target = Variable(data_[:-1,:]), Variable(data_[1:,:])
			#上邊一句，將輸入的詩句錯(cuò)開一個(gè)字，形成訓(xùn)練和目標(biāo)
			output,_ = model(input_)
			loss = criterion(output, target.view(-1))
			loss.backward()
			optimizer.step()
			
			loss_meter.add(loss.data[0]) #為什么是data[0]?
			
			#可視化用到的是utlis.py里的函數(shù)
			if (1+ii)%opt.plot_every ==0:
				
				if os.path.exists(opt.debug_file):
					ipdb.set_trace()
				vis.plot('loss',loss_meter.value()[0])
				
				# 下面是對目前模型情況的測試，詩歌原文
				poetrys = [[ix2word[_word] for _word in data_[:,_iii]] 
									for _iii in range(data_.size(1))][:16]
				#上面句子嵌套了兩個(gè)循環(huán)，主要是將詩歌索引的前十六個(gè)字變成原文
				vis.text('</br>'.join([''.join(poetry) for poetry in 
				poetrys]),win = u'origin_poem')
				gen_poetries = []
				#分別以以下幾個(gè)字作為詩歌的第一個(gè)字，生成8首詩
				for word in list(u'春江花月夜涼如水'):
					gen_poetry = ''.join(generate(model,word,ix2word,word2ix))
					gen_poetries.append(gen_poetry)
				vis.text('</br>'.join([''.join(poetry) for poetry in 
				gen_poetries]), win = u'gen_poem')
		t.save(model.state_dict(), '%s_%s.pth' %(opt.model_prefix,epoch))
 
def gen(**kwargs):
	'''
	提供命令行接口，用以生成相應(yīng)的詩
	'''
	
	for k,v in kwargs.items():
		setattr(opt,k,v)
	data, word2ix, ix2word = get_data(opt)
	model = PoetryModel(len(word2ix), 128, 256)
	map_location = lambda s,l:s
	# 上邊句子里的map_location是在load里用的，用以加載到指定的CPU或GPU，
	# 上邊句子的意思是將模型加載到默認(rèn)的GPU上
	state_dict = t.load(opt.model_path, map_location = map_location)
	model.load_state_dict(state_dict)
	
	if opt.use_gpu:
		model.cuda()
	if sys.version_info.major == 3:
		if opt.start_words.insprintable():
			start_words = opt.start_words
			prefix_words = opt.prefix_words if opt.prefix_words else None
		else:
			start_words = opt.start_words.encode('ascii',\
			'surrogateescape').decode('utf8')
			prefix_words = opt.prefix_words.encode('ascii',\
			'surrogateescape').decode('utf8') if opt.prefix_words else None
		start_words = start_words.replace(',',u'，')\
											.replace('.',u'。')\
											.replace('?',u'？')
		gen_poetry = gen_acrostic if opt.acrostic else generate
		result = gen_poetry(model,start_words,ix2word,word2ix,prefix_words)
		print(''.join(result))
if __name__ == '__main__':
	import fire
	fire.Fire()