TensorFlow: Movie Rating Prediction - Revision history

Onnowpurbo at 03:35, 30 July 2019

2019-07-30T03:35:29Z

Onnowpurbo at 03:24, 30 July 2019

2019-07-30T03:24:10Z

Onnowpurbo at 02:46, 30 July 2019

2019-07-30T02:46:55Z

Onnowpurbo: Created page with "Download https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_score_comparison.csv https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_sc..."

2019-07-30T02:42:19Z

Created page with "Download https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_score_comparison.csv https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_sc..."

New page

Download

https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_score_comparison.csv
https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_scrape.csv

==Pranala Menarik==

* [[TensorFlow]]

@@ Line 35: / Line 35: @@
   RT_lst = df['RT_norm'] >= 4.
   my_heatmap(df[RT_lst][rankings_lst].corr(method='pearson'))
-−
-−
   feature_cols = ['Fandango_Stars', 'RT_user_norm', 'RT_norm', 'Metacritic_user_norm', 'Metacritic_norm']
   X = df.loc[:, feature_cols]
   y = df['IMDB_norm']
   X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.50, random_state=43)
-−
   dim = len(feature_cols)
   dim += 1
   X_train = X_train.assign( independent = pd.Series([1] * len(y_train), index=X_train.index))
   X_test = X_test.assign( independent = pd.Series([1] * len(y_train), index=X_test.index))
-−
   P_train = X_train.as_matrix(columns=None)
   P_test = X_test.as_matrix(columns=None)
   q_train = np.array(y_train.values).reshape(-1,1)
   q_test = np.array(y_test.values).reshape(-1,1)
-−
   P = tf.placeholder(tf.float32,[None,dim])
   q = tf.placeholder(tf.float32,[None,1])
@@ Line 57: / Line 58: @@
   bias = tf.Variable(tf.constant(1.0, shape = [dim]))
   q_ = tf.add(tf.matmul(P, T),bias)
-−
   cost = tf.reduce_mean(tf.square(q_ - q))
   learning_rate = 0.0001
   training_op = tf.train.GradientDescentOptimizer(learning_rate=learning_rate).minimize(cost)
-−
   init_op = tf.global_variables_initializer()
   cost_history = np.empty(shape=[1],dtype=float)
-−
   training_epochs = 50000
   with tf.Session() as sess:
@@ Line 78: / Line 80: @@
       mse_temp = mse.eval()
       sess.close()
-−
   print(mse_temp)
   RMSE = math.sqrt(mse_temp)
   print(RMSE)

← Older revision		Revision as of 03:24, 30 July 2019
Line 5:		Line 5:

	Source		Source
		+
		+	import numpy as np
		+	import pandas as pd
		+	from scipy import stats
		+	import sklearn
		+	from sklearn.model_selection import train_test_split
		+	import tensorflow as tf
		+	import matplotlib
		+	import matplotlib.pyplot as plt
		+	import seaborn as sns
		+	import math
		+
		+	df = pd.read_csv('fandango_score_comparison.csv')
		+	print(df.head())
		+	df.rename(columns={'Metacritic_user_nom':'Metacritic_user_norm'},inplace=True)
		+	rankings_lst = ['Fandango_Stars',
		+	'RT_user_norm',
		+	'RT_norm',
		+	'IMDB_norm',
		+	'Metacritic_user_norm',
		+	'Metacritic_norm']
		+	def my_heatmap(df):
		+	import seaborn as sns
		+	fig, axes = plt.subplots()
		+	sns.heatmap(df, annot=True)
		+	plt.show()
		+	plt.close()
		+	my_heatmap(df[rankings_lst].corr(method='pearson'))
		+	RT_lst = df['RT_norm'] >= 4.
		+	my_heatmap(df[RT_lst][rankings_lst].corr(method='pearson'))
		+
		+
		+	feature_cols = ['Fandango_Stars', 'RT_user_norm', 'RT_norm', 'Metacritic_user_norm', 'Metacritic_norm']
		+	X = df.loc[:, feature_cols]
		+	y = df['IMDB_norm']
		+	X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.50, random_state=43)
		+
		+	dim = len(feature_cols)
		+	dim += 1
		+	X_train = X_train.assign( independent = pd.Series([1] * len(y_train), index=X_train.index))
		+	X_test = X_test.assign( independent = pd.Series([1] * len(y_train), index=X_test.index))
		+
		+	P_train = X_train.as_matrix(columns=None)
		+	P_test = X_test.as_matrix(columns=None)
		+	q_train = np.array(y_train.values).reshape(-1,1)
		+	q_test = np.array(y_test.values).reshape(-1,1)
		+
		+	P = tf.placeholder(tf.float32,[None,dim])
		+	q = tf.placeholder(tf.float32,[None,1])
		+	T = tf.Variable(tf.ones([dim,1]))
		+	bias = tf.Variable(tf.constant(1.0, shape = [dim]))
		+	q_ = tf.add(tf.matmul(P, T),bias)
		+
		+	cost = tf.reduce_mean(tf.square(q_ - q))
		+	learning_rate = 0.0001
		+	training_op = tf.train.GradientDescentOptimizer(learning_rate=learning_rate).minimize(cost)
		+
		+	init_op = tf.global_variables_initializer()
		+	cost_history = np.empty(shape=[1],dtype=float)
		+
		+	training_epochs = 50000
		+	with tf.Session() as sess:
		+	sess.run(init_op)
		+	cost_history = np.empty(shape=[1], dtype=float)
		+	t_history = np.empty(shape=[dim, 1], dtype=float)
		+	for epoch in range(training_epochs):
		+	sess.run(training_op, feed_dict={P: P_train, q: q_train})
		+	cost_history = np.append(cost_history, sess.run(cost,feed_dict={P: P_train, q: q_train}))
		+	t_history = np.append(t_history, sess.run(T, feed_dict={P: P_train, q: q_train}), axis=1)
		+	q_pred = sess.run(q_, feed_dict={P: P_test})[:, 0]
		+	mse = tf.reduce_mean(tf.square(q_pred - q_test))
		+	mse_temp = mse.eval()
		+	sess.close()
		+
		+	print(mse_temp)
		+	RMSE = math.sqrt(mse_temp)
		+	print(RMSE)
		+

← Older revision		Revision as of 02:46, 30 July 2019
Line 4:		Line 4:
	https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_scrape.csv		https://github.com/fivethirtyeight/data/blob/master/fandango/fandango_scrape.csv

		+	Source