Minimal character-level language model with a Vanilla Recurrent Neural Network, in Python/numpy

n-byte-wise_seq2seq.py

"""
Minimal character-level Vanilla RNN model. Written by Andrej Karpathy (@karpathy). Edited by Geoffrey Churchill
BSD License
"""
import numpy as np
from sys import argv

try:
  num_steps=int(argv[3])
except IndexError:
  num_steps=100
  
# data I/O
try:
  chunk_size=int(argv[2]) # number of bytes per chunk
except IndexError:
  chunk_size=1


print('reading',argv[1])
with open(argv[1],'rb') as f:
  data=f.read()

print('chunking data')
data=[data[i:i+chunk_size] for i in range(0, len(data), chunk_size)]

print('making set of chunks')
chunks=set(data)

data_size, vocab_size = len(data), len(chunks)
print('data has {0} characters, {1} unique.'.format(data_size, vocab_size))

print('compiling lookup tables')
char_to_ix = { ch:i for i,ch in enumerate(chunks) }
ix_to_char = { i:ch for i,ch in enumerate(chunks) }

# hyperparameters
hidden_size =128 # size of hidden layer of neurons
seq_length = 128 # number of steps to unroll the RNN for
learning_rate = lambda i:0.1#10/(i+1)

# model parameters
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
  for t in range(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
    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(range(len(inputs))):
    dy = np.copy(ps[t])
    dy[targets[t]] -= 1 # backprop into y
    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
  for t in range(n):
    if not t&255:
      print('{0}%'.format(100*t/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
    yield ix

p=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
for n in range(num_steps):
  # 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(str(ix_to_char[ix]) for ix in sample_ix)
    print('----\n {0} \n----'.format(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
  print('iter {0}, loss: {1}'.format(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(n) * dparam / np.sqrt(mem + 1e-8) # adagrad update

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

num_out_chunks=int(argv[4])
print('writing {0} chunks of {1} bytes (total={2} bytes) to {3}'.format(num_out_chunks,chunk_size,num_out_chunks*chunk_size,argv[5]))
with open(argv[5],'wb') as of:
  of.write(b''.join(ix_to_char[ix] for ix in sample(hprev, inputs[0],num_out_chunks)))