👻

James Briggs jamescalam

👻

jamescalam / rotating_surface.py

Last active March 11, 2020 09:38

Matplotlib animation of rotating surface

	import numpy as np
	import matplotlib.pyplot as plt
	from matplotlib import animation, cm
	from mpl_toolkits.mplot3d import Axes3D


	# create a figure
	fig = plt.figure()
	# initialise 3D Axes
	ax = Axes3D(fig)

jamescalam / meditations_text.py

Last active April 4, 2020 17:19

Code snippet for part of data cleansing process for Meditations data import.

	import requests
	import re

	# import Meditations
	response = requests.get('http://classics.mit.edu/Antoninus/meditations.mb.txt')
	data = response.text

	# clean the text
	data = data.split("Translated by George Long")[1].replace("-", "").split("THE END")[0]
	data = re.sub("BOOK [A-Z]+\n", "", data)

jamescalam / meditations_char2idx.py

Last active April 1, 2020 16:28

Short code for building mapping of character to index (char2idx) and index to character (idx2char).

	char2idx = {c:i for i, c in enumerate(vocab)}
	idx2char = np.array(vocab)

jamescalam / meditations_gru.py

Last active April 1, 2020 18:58

Model definition for simple GRU network in TF Keras.

	model = tf.keras.Sequential([
	tf.keras.layers.Embedding(len(vocab), EMBED_DIM,
	batch_input_shape=[BATCH_SIZE, None]),
	tf.keras.layers.GRU(UNITS, return_sequences=True,
	stateful=True,
	dropout=0.1),
	tf.keras.layers.Dense(len(vocab))

jamescalam / meditations_save_model.py

Created April 1, 2020 16:40

Snippet showing how to setup model checkpoint saving using built-in call backs function

	checkpoint = tf.keras.callbacks.ModelCheckpoint(
	filepath='./training_checkpoints/ckpt_{epoch}',
	save_weights_only=True
	)

	history = model.fit(dataset, epochs=EPOCHS, callbacks=[checkpoint_callback])

jamescalam / meditations_load_model.py

Created April 1, 2020 16:44

Snippet showing how to load a model from saved weights. We load with batch size of 1 ready for text generation.

	model = build_model(len(vocab), EMBED_DIM, UNITS, 1)
	model.load_weights(tf.train.latest_checkpoint('./training_checkpoints')
	model.build(tf.TensorShape([1, None]))

jamescalam / meditations_generation.py

Created April 1, 2020 16:46

Example snippet on how to generate text using TF Keras model.

	meditations = "From " # initialize our meditations text
	# convert to indices
	input_eval = tf.expand_dims([char2idx[c] for x in meditations], 0)
	# initialize states
	model.reset_states()
	# loop through, generating 100K characters
	for i in range(100000):
	y_hat = model(input_eval) # make a prediction
	y_hat = tf.squeeze(y_hat, 0) # remove batch dimension
	predicted_idx = tf.random.categorical(y_hat, num_samples=1)[-1,0].numpy()

jamescalam / meditations_char_dataset.py

Created April 1, 2020 17:08

Snippet of how to convert indexed data into a TF Dataset object.

char_dataset = tf.data.Dataset.from_tensor_slices(data_idx)

jamescalam / meditations_sequences.py

Created April 1, 2020 17:09

Snippet showing how to split a dataset into sequences with .batch method in TF Dataset object.

sequences = char_dataset.batch(SEQ_LEN+1, drop_remainder=True)

jamescalam / meditations_split_input_output.py

Created April 1, 2020 17:11

Snippet showing how to split data for sequence prediction.

	def split_input_output(chunk):
	return chunk[:-1], chunk[1:]

	dataset = sequences.map(split_input_output)