"""
Minimal character-level Vanilla RNN model. Written by Andrej Karpathy (@karpathy)
BSD License
"""
import numpy as np

# data I/O 导入英文纯文本文件，建立词到索引、索引到词的映射
data = open('input.txt', 'r').read() # should be simple plain text file
chars = list(set(data))
data_size, vocab_size = len(data), len(chars)
print 'data has %d characters, %d unique.' % (data_size, vocab_size)
char_to_ix = { ch:i for i,ch in enumerate(chars) }
ix_to_char = { i:ch for i,ch in enumerate(chars) }

# hyperparameters 定义隐藏层的神经元数量、每一步处理多长的序列以及学习速率
hidden_size = 100 # size of hidden layer of neurons
seq_length = 25 # number of steps to unroll the RNN for
learning_rate = 1e-1

# model parameters 随机生成神经网络参数矩阵，Wxh即U，Whh即W，Why即V，以及偏置单元
Wxh = np.random.randn(hidden_size, vocab_size)*0.01 # input to hidden
Whh = np.random.randn(hidden_size, hidden_size)*0.01 # hidden to hidden
Why = np.random.randn(vocab_size, hidden_size)*0.01 # hidden to output
bh = np.zeros((hidden_size, 1)) # hidden bias
by = np.zeros((vocab_size, 1)) # output bias

def lossFun(inputs, targets, hprev):
  """
  inputs,targets are both list of integers.
  hprev is Hx1 array of initial hidden state
  returns the loss, gradients on model parameters, and last hidden state
  """
  xs, hs, ys, ps = {}, {}, {}, {}
  hs[-1] = np.copy(hprev)
  loss = 0
  # forward pass 正向传播过程，输入采用1-k向量表示，隐藏层函数采用tanh
  for t in xrange(len(inputs)):
    xs[t] = np.zeros((vocab_size,1)) # encode in 1-of-k representation
    xs[t][inputs[t]] = 1
    hs[t] = np.tanh(np.dot(Wxh, xs[t]) + np.dot(Whh, hs[t-1]) + bh) # hidden state 参考文献Learning Recurrent Neural Networks with Hessian-Free Optimization
    ys[t] = np.dot(Why, hs[t]) + by # unnormalized log probabilities for next chars
    ps[t] = np.exp(ys[t]) / np.sum(np.exp(ys[t])) # probabilities for next chars
    loss += -np.log(ps[t][targets[t],0]) # softmax (cross-entropy loss)
  # backward pass: compute gradients going backwards
  dWxh, dWhh, dWhy = np.zeros_like(Wxh), np.zeros_like(Whh), np.zeros_like(Why)
  dbh, dby = np.zeros_like(bh), np.zeros_like(by)
  dhnext = np.zeros_like(hs[0])
  for t in reversed(xrange(len(inputs))):
    dy = np.copy(ps[t])
    dy[targets[t]] -= 1 # backprop into y. see http://cs231n.github.io/neural-networks-case-study/#grad if confused here
    dWhy += np.dot(dy, hs[t].T)
    dby += dy
    dh = np.dot(Why.T, dy) + dhnext # backprop into h
    dhraw = (1 - hs[t] * hs[t]) * dh # backprop through tanh nonlinearity
    dbh += dhraw
    dWxh += np.dot(dhraw, xs[t].T)
    dWhh += np.dot(dhraw, hs[t-1].T)
    dhnext = np.dot(Whh.T, dhraw)
  for dparam in [dWxh, dWhh, dWhy, dbh, dby]:
    np.clip(dparam, -5, 5, out=dparam) # clip to mitigate exploding gradients
  return loss, dWxh, dWhh, dWhy, dbh, dby, hs[len(inputs)-1]

def sample(h, seed_ix, n):
  """ 
  sample a sequence of integers from the model 
  h is memory state, seed_ix is seed letter for first time step
  """
  x = np.zeros((vocab_size, 1))
  x[seed_ix] = 1
  ixes = []
  for t in xrange(n):
    h = np.tanh(np.dot(Wxh, x) + np.dot(Whh, h) + bh)
    y = np.dot(Why, h) + by
    p = np.exp(y) / np.sum(np.exp(y))
    ix = np.random.choice(range(vocab_size), p=p.ravel())
    x = np.zeros((vocab_size, 1))
    x[ix] = 1
    ixes.append(ix)
  return ixes

n, p = 0, 0
mWxh, mWhh, mWhy = np.zeros_like(Wxh), np.zeros_like(Whh), np.zeros_like(Why)
mbh, mby = np.zeros_like(bh), np.zeros_like(by) # memory variables for Adagrad
smooth_loss = -np.log(1.0/vocab_size)*seq_length # loss at iteration 0
while True:
  # prepare inputs (we're sweeping from left to right in steps seq_length long)
  if p+seq_length+1 >= len(data) or n == 0: 
    hprev = np.zeros((hidden_size,1)) # reset RNN memory
    p = 0 # go from start of data
  inputs = [char_to_ix[ch] for ch in data[p:p+seq_length]]
  targets = [char_to_ix[ch] for ch in data[p+1:p+seq_length+1]]

  # sample from the model now and then
  if n % 100 == 0:
    sample_ix = sample(hprev, inputs[0], 200)
    txt = ''.join(ix_to_char[ix] for ix in sample_ix)
    print '----\n %s \n----' % (txt, )
    
  # forward seq_length characters through the net and fetch gradient
  loss, dWxh, dWhh, dWhy, dbh, dby, hprev = lossFun(inputs, targets, hprev)
  smooth_loss = smooth_loss * 0.999 + loss * 0.001
  if n % 100 == 0: print 'iter %d, loss: %f' % (n, smooth_loss) # print progress

  # perform parameter update with Adagrad
  for param, dparam, mem in zip([Wxh, Whh, Why, bh, by], 
                                [dWxh, dWhh, dWhy, dbh, dby], 
                                [mWxh, mWhh, mWhy, mbh, mby]):
    mem += dparam * dparam
    param += -learning_rate * dparam / np.sqrt(mem + 1e-8) # adagrad update

  p += seq_length # move data pointer
  n += 1 # iteration counter

# -*- coding: utf-8 -*-
"""
Created on Sat May 21 14:34:08 2016

@author: yangsicong
"""

import numpy as np

from keras.models import Sequential
from keras.layers.core import Dense, Activation, Dropout, TimeDistributedDense
from keras.layers.recurrent import LSTM

text = open('./input.txt', 'r').read()
char_to_idx = { ch: i for (i, ch) in enumerate(sorted(list(set(text)))) }
idx_to_char = { i: ch for (ch, i) in char_to_idx.items() }
vocab_size = len(char_to_idx)

print('Working on %d characters (%d unique)' % (len(text), vocab_size))

SEQ_LENGTH = 64
BATCH_SIZE = 16
BATCH_CHARS = len(text) / BATCH_SIZE
LSTM_SIZE = 512
LAYERS = 3

def read_batches(text):
    T = np.asarray([char_to_idx[c] for c in text], dtype=np.int32)
    X = np.zeros((BATCH_SIZE, SEQ_LENGTH, vocab_size))
    Y = np.zeros((BATCH_SIZE, SEQ_LENGTH, vocab_size))

    for i in range(0, BATCH_CHARS - SEQ_LENGTH - 1, SEQ_LENGTH):
        X[:] = 0
        Y[:] = 0
        for batch_idx in range(BATCH_SIZE):
            start = batch_idx * BATCH_CHARS + i
            for j in range(SEQ_LENGTH):
                X[batch_idx, j, T[start+j]] = 1
                Y[batch_idx, j, T[start+j+1]] = 1

        yield X, Y


def build_model(batch_size, seq_len):
    model = Sequential()
    model.add(LSTM(LSTM_SIZE, return_sequences=True, batch_input_shape=(batch_size, seq_len, vocab_size), stateful=True))
    model.add(Dropout(0.2))
    for l in range(LAYERS - 1):
        model.add(LSTM(LSTM_SIZE, return_sequences=True, stateful=True))
        model.add(Dropout(0.2))

    model.add(TimeDistributedDense(vocab_size))
    model.add(Activation('softmax'))
    model.compile(loss='categorical_crossentropy', optimizer='adagrad')
    return model


print 'Building model.'
test_model = build_model(1, 1)
training_model = build_model(BATCH_SIZE, SEQ_LENGTH)
print '... done'

def sample(epoch, sample_chars=256):
    test_model.reset_states()
    test_model.load_weights('./tmp/keras_char_rnn.%d.h5' % epoch)
    header = 'LSTM based '
    sampled = [char_to_idx[c] for c in header]

    for c in header:
        batch = np.zeros((1, 1, vocab_size))
        batch[0, 0, char_to_idx[c]] = 1
        test_model.predict_on_batch(batch)

    for i in range(sample_chars):
        batch = np.zeros((1, 1, vocab_size))
        batch[0, 0, sampled[-1]] = 1
        softmax = test_model.predict_on_batch(batch)[0].ravel()
        sample = np.random.choice(range(vocab_size), p=softmax)
        sampled.append(sample)

    print ''.join([idx_to_char[c] for c in sampled])

for epoch in range(100):
    for i, (x, y) in enumerate(read_batches(text)):
        loss = training_model.train_on_batch(x, y)
        print epoch, i, loss

        if i % 1000 == 0:
            training_model.save_weights('./tmp/keras_char_rnn.%d.h5' % epoch, overwrite=True)
            sample(epoch)