markdown stringlengths 0 37k | code stringlengths 1 33.3k | path stringlengths 8 215 | repo_name stringlengths 6 77 | license stringclasses 15
values |
|---|---|---|---|---|
Test you model
Run your best model on the validation and test sets. You should achieve above 50% accuracy on the validation set. | y_test_pred = np.argmax(best_model.loss(data['X_test']), axis=1)
y_val_pred = np.argmax(best_model.loss(data['X_val']), axis=1)
print('Validation set accuracy: ', (y_val_pred == data['y_val']).mean())
print('Test set accuracy: ', (y_test_pred == data['y_test']).mean()) | MOOC/stanford_cnn_cs231n/assignment2/FullyConnectedNets.ipynb | ThyrixYang/LearningNotes | gpl-3.0 |
The following function is used to run the different benchmarks.
It takes a target function to test, a setup function to create the file and the number of iterations the function should be run to get a decent average: | import time
def benchmark(target, setup=None, teardown=None, iterations=10):
total_time = 0
setup_teardown_start = time.time()
for i in range(iterations):
data = tuple()
if setup is not None:
data = setup()
time.sleep(1) # allow changes to be flushed to disk
star... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
The following functions are used as wrappers to make it easy to run a benchmark of Exdir or h5py: | import pandas as pd
import numpy as np
all_results = []
def benchmark_both(function, iterations=10, name_validation=True):
if name_validation:
setup_exdir_ = setup_exdir
name = function.__name__
else:
setup_exdir_ = setup_exdir_no_validation
name = function.__name__ + " (minima... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
We are now ready to start running the different benchmarks.
Benchmark functions
The following benchmark creates a small number of attributes.
This should be very fast with both h5py and Exdir: | def add_few_attributes(obj):
for i in range(5):
obj.attrs["hello" + str(i)] = "world"
benchmark_both(add_few_attributes) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
The following benchmark adds a larger number of attributes one-by-one.
Because Exdir needs to read back and rewrite the entire file in case someone changed it between each write, this is significantly slower with Exdir than h5py: | def add_many_attributes(obj):
for i in range(200):
obj.attrs["hello" + str(i)] = "world"
benchmark_both(add_many_attributes, 10) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
However, Exdir is capable of writing all attributes in one operation.
This makes writing the same attributes about as fast (or even faster than h5py).
Writing a large number of attributes in a single operation is not possible with h5py.
We therefore need to run this only with Exdir: | def add_many_attributes_single_operation(obj):
attributes = {}
for i in range(200):
attributes["hello" + str(i)] = "world"
obj.attrs = attributes
benchmark_exdir(add_many_attributes_single_operation) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Exdir also supports adding nested attributes, such as Python dictionaries, which is not supported by h5py: | def add_attribute_tree(obj):
tree = {}
for i in range(100):
tree["hello" + str(i)] = "world"
tree["intermediate"] = {}
intermediate = tree["intermediate"]
for level in range(10):
level_str = "level" + str(level)
intermediate[level_str] = {}
intermediate = intermediate... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
The following benchmarks create a small, a medium, and a large dataset: | def add_small_dataset(obj):
data = np.zeros((100, 100, 100))
obj.create_dataset("foo", data=data)
obj.close()
benchmark_both(add_small_dataset)
def add_medium_dataset(obj):
data = np.zeros((1000, 100, 100))
obj.create_dataset("foo", data=data)
obj.close()
benchmark_both(add_medium_dat... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
There is some overhead in creating the objects themselves.
This is rather small in h5py, but can be high in Exdir with name validation enabled.
This is because the name of every created object must be checked against all the existing objects in the same group: | def create_many_objects(obj):
for i in range(5000):
group = obj.create_group("group{}".format(i))
benchmark_both(create_many_objects, 3) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Without minimal validation, this is almost as fast in Exdir as it is in h5py.
Minimal name validation only checks if file with the exact same name exist in the folder: | benchmark_both(create_many_objects, 3, name_validation=False) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Not only the number of created objects matter.
Creating them in a tree structure can also incur a performance penalty.
The following test creates an object tree: | def create_large_tree(obj, level=0):
if level > 4:
return
for i in range(3):
group = obj.create_group("group_{}_{}".format(i, level))
data = np.zeros((10, 10, 10))
group.create_dataset("dataset_{}_{}".format(i, level), data=data)
create_large_tree(group, level + 1)
... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
The final benchmark tests writing a "slice" of a dataset, which means only a part of the entire dataset is modified.
This is typically fast in both h5py and in Exdir thanks to memory mapping. | def write_slice(dataset):
dataset[320:420, 0:300, 0:100] = np.ones((100, 300, 100))
def create_setup_dataset(setup_function):
def setup():
f, path = setup_function()
data = np.zeros((1000, 500, 100))
dataset = f.create_dataset("foo", data=data)
time.sleep(1) # allow changes to g... | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Benchmark summary
The results are summarized in the following table: | pd.concat(all_results) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Profiling the largest differences
While the performance of Exdir in many cases is close to h5py, there are a few cases that can be worth investigating further.
For instance, it might be interesting to know what takes most time in create_large_tree, which is about 2-3 times slower in Exdir than h5py: | import cProfile
f, path = setup_exdir()
cProfile.run('create_large_tree(f)', sort="cumtime")
teardown_exdir(f, path) | tests/benchmarks/benchmarks.ipynb | CINPLA/exdir | mit |
Unzipping files with Amazon Baby Products Reviews
The dataset consists of baby product reviews from Amazon.com. | # Put files in current direction into a list
files_list = [f for f in os.listdir('.') if os.path.isfile(f)]
# Filename of unzipped file
unzipped_file = 'amazon_baby.csv'
# If upzipped file not in files_list, unzip the file
if unzipped_file not in files_list:
zip_file = unzipped_file + '.zip'
unzipping = zipfi... | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Loading the products data
The dataset is loaded into a Pandas DataFrame called products. | products = pd.read_csv("amazon_baby.csv") | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, let us see a preview of what the dataset looks like. | products.head() | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Performing text cleaning
Let us explore a specific example of a baby product. | products.ix[1] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, we will perform 2 simple data transformations:
Remove punctuation using Python's built-in string functionality.
Transform the reviews into word-counts.
Aside. In this notebook, we remove all punctuations for the sake of simplicity. A smarter approach to punctuations would preserve phrases such as "I'd", "would'v... | products["review"] = products["review"].fillna("") | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Below, we are removing all the punctuation from the strings in the review column and saving the result into a new column in the dataframe. | products["review_clean"] = products["review"].str.translate(None, string.punctuation) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Extract sentiments
We will ignore all reviews with rating = 3, since they tend to have a neutral sentiment. | products = products[products['rating'] != 3]
len(products) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, we will assign reviews with a rating of 4 or higher to be positive reviews, while the ones with rating of 2 or lower are negative. For the sentiment column, we use +1 for the positive class label and -1 for the negative class label.
Below, we are create a function we will applyi to the "ratings" column of the data... | def sent_func(x):
# If rating is >=4, return a positive sentiment (+1)
if x>=4:
return 1
# Else, return a negative sentiment (-1)
else:
return -1 | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Creating a "sentiment" column by applying the sent_func to the "rating" column in the dataframe. | products['sentiment'] = products['rating'].apply(sent_func)
products.ix[20:22] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, we can see that the dataset contains an extra column called sentiment which is either positive (+1) or negative (-1).
Split data into training and test sets
Let's perform a train/test split with 80% of the data in the training set and 20% of the data in the test set.
Loading the indicies for the train and test dat... | with open('module-2-assignment-train-idx.txt', 'r') as train_file:
ind_list_train = map(int,train_file.read().split(','))
with open('module-2-assignment-test-idx.txt', 'r') as test_file:
ind_list_test = map(int,test_file.read().split(',')) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Using the indicies of the train and test data to create the train and test datasets. | train_data = products.iloc[ind_list_train,:]
test_data = products.iloc[ind_list_test,:]
print len(train_data)
print len(test_data) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Build the word count vector for each review
We will now compute the word count for each word that appears in the reviews. A vector consisting of word counts is often referred to as bag-of-word features. Since most words occur in only a few reviews, word count vectors are sparse. For this reason, scikit-learn and many o... | # Use this token pattern to keep single-letter words
vectorizer = CountVectorizer(token_pattern=r'\b\w+\b')
# First, learn vocabulary from the training data and assign columns to words
# Then convert the training data into a sparse matrix
train_matrix = vectorizer.fit_transform(train_data['review_clean'])
# Second, con... | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Train a sentiment classifier with logistic regression
We will now use logistic regression to create a sentiment classifier on the training data. This model will use the column word_count as a feature and the column sentiment as the target.
Note: This line may take 1-2 minutes.
Creating an instance of the LogisticRegres... | logreg = linear_model.LogisticRegression() | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Using the fit method to train the classifier. This model should use the sparse word count matrix (train_matrix) as features and the column sentiment of train_data as the target. Use the default values for other parameters. Call this model sentiment_model. | sentiment_model = logreg.fit(train_matrix, train_data["sentiment"]) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Putting all the weights from the model into a numpy array. | weights_list = list(sentiment_model.intercept_) + list(sentiment_model.coef_.flatten())
weights_sent_model = np.array(weights_list, dtype = np.double)
print len(weights_sent_model) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
There are a total of 121713 coefficients in the model. Recall from the lecture that positive weights $w_j$ correspond to weights that cause positive sentiment, while negative weights correspond to negative sentiment.
Quiz question: How many weights are >= 0? | num_positive_weights = len(weights_sent_model[weights_sent_model >= 0.0])
num_negative_weights = len(weights_sent_model[weights_sent_model < 0.0])
print "Number of positive weights: %i" % num_positive_weights
print "Number of negative weights: %i" % num_negative_weights | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Making predictions with logistic regression
Now that a model is trained, we can make predictions on the test data. In this section, we will explore this in the context of 3 examples in the test dataset. We refer to this set of 3 examples as the sample_test_data. | sample_test_data = test_data.ix[[59,71,91]]
print sample_test_data['rating']
sample_test_data | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Let's dig deeper into the first row of the sample_test_data. Here's the full review: | sample_test_data['review'].ix[59] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
That review seems pretty positive.
Now, let's see what the next row of the sample_test_data looks like. As we could guess from the sentiment (-1), the review is quite negative. | sample_test_data['review'].ix[71] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
We will now make a class prediction for the sample_test_data. The sentiment_model should predict +1 if the sentiment is positive and -1 if the sentiment is negative. Recall from the lecture that the score (sometimes called margin) for the logistic regression model is defined as:
$$
\mbox{score}_i = \mathbf{w}^T h(\mat... | sample_test_matrix = vectorizer.transform(sample_test_data['review_clean'])
scores = sentiment_model.decision_function(sample_test_matrix)
print scores | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Predicting sentiment
These scores can be used to make class predictions as follows:
$$
\hat{y} =
\left{
\begin{array}{ll}
+1 & \mathbf{w}^T h(\mathbf{x}_i) > 0 \
-1 & \mathbf{w}^T h(\mathbf{x}_i) \leq 0 \
\end{array}
\right.
$$
Using scores, write code to calculate $\hat{y}$, the class predictions: | pred_sent_test_data = []
for val in scores:
if val>0:
pred_sent_test_data.append(1)
else:
pred_sent_test_data.append(-1)
print pred_sent_test_data | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Checkpoint: Run the following code to verify that the class predictions obtained by your calculations are the same as that obtained from Scikit-Learn. | print "Class predictions according to Scikit-Learn:"
print sentiment_model.predict(sample_test_matrix) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Probability predictions
Recall from the lectures that we can also calculate the probability predictions from the scores using:
$$
P(y_i = +1 | \mathbf{x}_i,\mathbf{w}) = \frac{1}{1 + \exp(-\mathbf{w}^T h(\mathbf{x}_i))}.
$$
Using the variable scores calculated previously, write code to calculate the probability that a ... | prob_pos_score = 1.0/(1.0 + np.exp(-scores))
prob_pos_score | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Checkpoint: Make sure your probability predictions match the ones obtained from Scikit-Learn. | print "Class predictions according to Scikit-Learn:"
print sentiment_model.predict_proba(sample_test_matrix)[:,1] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Of the three data points in sample_test_data, which one (first, second, or third) has the lowest probability of being classified as a positive review?
The 3rd data point has the lowest probability of being positive
Find the most positive (and negative) review
We now turn to examining the full test data... | scores_test_data = sentiment_model.decision_function(test_matrix)
prob_test_data = 1.0/(1.0 + np.exp(-scores_test_data)) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
To find the 40 most positive and the 40 most negative values, we will create a list of tuples with the entries (probability, index). We will then sort the list and will be able to extract the indicies corresponding to each entry. | # List of indicies in the test data
ind_vals_test_data = test_data.index.values
# Empty list that will be filled with the tuples (probability, index)
score_label_lst_test = len(scores_test_data)*[-1] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Filling the list of tuples with the (probability, index) values | for i in range(len(scores_test_data)):
score_label_lst_test[i] = (prob_test_data[i],ind_vals_test_data[i]) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Sorting the list with the entries (probability, index) | score_label_lst_test.sort() | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Extracting the top 40 positive reviews and the top 40 negative reviews | top_40_pos_test_rev = score_label_lst_test[-40:]
top_40_neg_test_rev = score_label_lst_test[0:40] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Getting the indicies of the top 40 positive reviews. | ind_top_40_pos_test = 40*[-1]
for i,val in enumerate(top_40_pos_test_rev):
ind_top_40_pos_test[i] = val[1] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Getting the indicies of the top 40 negative reviews. | ind_top_40_neg_test = 40*[-1]
for i,val in enumerate(top_40_neg_test_rev):
ind_top_40_neg_test[i] = val[1] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Which of the following products are represented in the 40 most positive reviews? [multiple choice] | test_data.ix[ind_top_40_pos_test]["name"] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Which of the following products are represented in the 20 most negative reviews? [multiple choice] | test_data.ix[ind_top_40_neg_test]["name"] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Compute accuracy of the classifier
We will now evaluate the accuracy of the trained classifer. Recall that the accuracy is given by
$$
\mbox{accuracy} = \frac{\mbox{# correctly classified examples}}{\mbox{# total examples}}
$$
This can be computed as follows:
Step 1: Use the trained model to compute class predictions
... | def get_classification_accuracy(model, data, true_labels):
# Constructing the wordcount vector
data_matrix = vectorizer.transform(data['review_clean'])
# Getting the predictions
preds_data = model.predict(data_matrix)
# Computing the number of correctly classified examples and the tot... | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, let's compute the classification accuracy of the sentiment_model on the test_data. | acc_sent_mod_test = get_classification_accuracy(sentiment_model, test_data, test_data['sentiment'])
print acc_sent_mod_test | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: What is the accuracy of the sentiment_model on the test_data? Round your answer to 2 decimal places (e.g. 0.76). | print "Accuracy on Test Data: %.2f" %(acc_sent_mod_test) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Does a higher accuracy value on the training_data always imply that the classifier is better?
No, you may be overfitting.
Now, computing the accuracy of the sentiment model on the training data for a future quiz question. | acc_sent_mod_train = get_classification_accuracy(sentiment_model, train_data, train_data['sentiment'])
print acc_sent_mod_train | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Finding the weights of significant words for the sentiment_model.
In this section, we will find the weights of significant words for the sentiment_model.
Creating a vocab list. The vocab list constains all the words used for the sentiment_model | vocab = vectorizer.get_feature_names()
print len(vocab) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Creating a list of the significant words in the utf-8 format | un_sig_words = [u'love', u'great', u'easy', u'old', u'little', u'perfect', u'loves',
u'well', u'able', u'car', u'broke', u'less', u'even', u'waste', u'disappointed',
u'work', u'product', u'money', u'would', u'return'] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Creating a list that will store all the indicies where the significant words appear in the vocab list. | ind_vocab_sig_words = [] | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Finding the index where each significant word appears. | for word in un_sig_words:
ind_vocab_sig_words.append(vocab.index(word)) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Creating an empty list that will store the weights of the significant words. Then, using the index to find the weight for each signigicant word. | ws_sent_mod_sig_words = []
for ind in ind_vocab_sig_words:
ws_sent_mod_sig_words.append(sentiment_model.coef_.flatten()[ind]) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Creating a series that will store the weights of the significant words and displaying this Series. | ws_sent_mod_ser = pd.Series(data=ws_sent_mod_sig_words, index=un_sig_words)
ws_sent_mod_ser | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Learn another classifier with fewer words
There were a lot of words in the model we trained above. We will now train a simpler logistic regression model using only a subet of words that occur in the reviews. For this assignment, we selected a 20 words to work with. These are: | significant_words = ['love', 'great', 'easy', 'old', 'little', 'perfect', 'loves',
'well', 'able', 'car', 'broke', 'less', 'even', 'waste', 'disappointed',
'work', 'product', 'money', 'would', 'return']
len(significant_words) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Compute a new set of word count vectors using only these words. The CountVectorizer class has a parameter that lets you limit the choice of words when building word count vectors: | vectorizer_word_subset = CountVectorizer(vocabulary=significant_words) # limit to 20 words
train_matrix_word_subset = vectorizer_word_subset.fit_transform(train_data['review_clean'])
test_matrix_word_subset = vectorizer_word_subset.transform(test_data['review_clean']) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Train a logistic regression model on a subset of data
We will now build a classifier with word_count_subset as the feature and sentiment as the target.
Creating an instance of the LogisticRegression class. Using the fit method to train the classifier. This model should use the sparse word count matrix (train_matrix) a... | log_reg = linear_model.LogisticRegression()
simple_model = logreg.fit(train_matrix_word_subset, train_data["sentiment"]) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Getting the weights for the 20 significant words from the simple_model | ws_simp_model = list(simple_model.coef_.flatten()) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Putting the weights in a Series with the words corresponding to the weights as the index. | ws_simp_mod_ser = pd.Series(data=ws_simp_model, index=significant_words)
ws_simp_mod_ser | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Consider the coefficients of simple_model. How many of the 20 coefficients (corresponding to the 20 significant_words and excluding the intercept term) are positive for the simple_model? | print len(simple_model.coef_[simple_model.coef_>0]) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Are the positive words in the simple_model (let us call them positive_significant_words) also positive words in the sentiment_model?
Yes, see weights below for the significant words for the sentiment model | ws_sent_mod_ser | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Comparing models
We will now compare the accuracy of the sentiment_model and the simple_model using the get_classification_accuracy method you implemented above.
First, compute the classification accuracy of the sentiment_model on the train_data: | acc_sent_mod_train | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, compute the classification accuracy of the simple_model on the train_data: | preds_simp_mod_train = simple_model.predict(train_matrix_word_subset)
n_cor_preds_simp_mod_train = float(np.sum(preds_simp_mod_train == train_data['sentiment'].values))
n_tol_preds_simp_mod_train = float(len(preds_simp_mod_train))
acc_simp_mod_train = n_cor_preds_simp_mod_train/n_tol_preds_simp_mod_train
print acc_simp... | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Which model (sentiment_model or simple_model) has higher accuracy on the TRAINING set? | if acc_sent_mod_train>acc_simp_mod_train:
print "sentiment_model"
else:
print "simple_model" | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now, we will repeat this excercise on the test_data. Start by computing the classification accuracy of the sentiment_model on the test_data: | acc_sent_mod_test | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Next, we will compute the classification accuracy of the simple_model on the test_data: | preds_simp_mod_test = simple_model.predict(test_matrix_word_subset)
n_cor_preds_simp_mod_test = float(np.sum(preds_simp_mod_test == test_data['sentiment'].values))
n_tol_preds_simp_mod_test = float(len(preds_simp_mod_test))
acc_simp_mod_test = n_cor_preds_simp_mod_test/n_tol_preds_simp_mod_test
print acc_simp_mod_test | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Which model (sentiment_model or simple_model) has higher accuracy on the TEST set? | if acc_sent_mod_test>acc_simp_mod_test:
print "sentiment_model"
else:
print "simple_model" | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Baseline: Majority class prediction
It is quite common to use the majority class classifier as the a baseline (or reference) model for comparison with your classifier model. The majority classifier model predicts the majority class for all data points. At the very least, you should healthily beat the majority class cla... | num_positive = (train_data['sentiment'] == +1).sum()
num_negative = (train_data['sentiment'] == -1).sum()
acc_pos_train = float(num_positive)/float(len(train_data['sentiment']))
acc_neg_train = float(num_negative)/float(len(train_data['sentiment']))
if acc_pos_train>acc_neg_train:
print "Positive Sentiment is Majo... | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Now compute the accuracy of the majority class classifier on test_data.
Quiz Question: Enter the accuracy of the majority class classifier model on the test_data. Round your answer to two decimal places (e.g. 0.76). | num_pos_test = (test_data['sentiment'] == +1).sum()
acc_pos_test = float(num_pos_test)/float(len(test_data['sentiment']))
print "Accuracy of Majority Class Classifier on Test Data: %.2f" %(acc_pos_test) | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Quiz Question: Is the sentiment_model definitely better than the majority class classifier (the baseline)? | if acc_sent_mod_test>acc_pos_test:
print "Yes, the sentiment_model is better than majority class classifier"
else:
print "No, the majority class classifier is better than sentiment_model" | Week_1_Predicting_Sentiment_from_Reviews/week_1_lin_classifier_assign.ipynb | Santana9937/Classification_ML_Specialization | mit |
Using magic functions of Jupyter and timeit
https://docs.python.org/3.5/library/timeit.html
https://ipython.org/ipython-doc/3/interactive/magics.html#magic-time | %%timeit
fun()
%%time
fun() | content/handouts/concurrency-exercise.ipynb | oroszgy/oroszgy.github.io | mit |
Exercises
What is the fastest way to download 100 pages from index.hu?
How to calculate the factors of 1000 random integers effectively using factorize_naive function below? | import requests
def get_page(url):
response = requests.request(url=url, method="GET")
return response
get_page("http://index.hu")
def factorize_naive(n):
""" A naive factorization method. Take integer 'n', return list of
factors.
"""
if n < 2:
return []
factors = []
p = 2
... | content/handouts/concurrency-exercise.ipynb | oroszgy/oroszgy.github.io | mit |
Install dependencies | !apt-get install libicu-dev libpango1.0-dev libcairo2-dev libleptonica-dev
| Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Clone, compile and set up Tesseract | !git clone https://github.com/tesseract-ocr/tesseract
import os
os.chdir('tesseract')
!sh autogen.sh
!./configure --disable-graphics
!make -j 8
!make install
!ldconfig
!make training
!make training-install | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Grab some things to scrape the RIA corpus | import os
os.chdir('/content')
!git clone https://github.com/jimregan/tesseract-gle-uncial/
!apt-get install lynx | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Scrape the RIA corpus | ! for i in A B C D E F G H I J K L M N O P Q R S T U V W X Y Z;do lynx -dump "http://corpas.ria.ie/index.php?fsg_function=1&fsg_page=$i" |grep http://corpas.ria.ie|awk '{print $NF}' >> list;done
!grep 'function=3' list |sort|uniq|grep corpas.ria|sed -e 's/function=3/function=5/' > input
!wget -x -c -i input
!mkdir t... | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Get the raw corpus in a single text file | !cat text/*.txt|grep -v '^$' > ria-raw.txt
| Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Compress the raw text; this can be downloaded through the file browser on the left, so the scraping steps can be skipped in future | !gzip ria-raw.txt | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
...and can be re-added using the upload feature in the file browser | !gzip -d ria-raw.txt.gz | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
This next part is so I can update the langdata files | import os
os.chdir('/content')
!git clone https://github.com/tesseract-ocr/langdata
!cat ria-raw.txt | perl /content/tesseract-gle-uncial/scripts/toponc.pl > ria-ponc.txt
!mkdir genwlout
!perl /content/tesseract-gle-uncial/scripts/genlangdata.pl -i ria-ponc.txt -d genwlout -p gle_uncial
import os
os.chdir('/content... | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
Generate | os.chdir('/content')
!mkdir unpack
!combine_tessdata -u /content/gle_uncial.traineddata unpack/gle_uncial.
os.chdir('unpack')
!for i in gle_uncial.word.bigrams gle_uncial.wordlist gle_uncial.numbers gle_uncial.punc; do cp /content/genwlout/$i .;done
!wordlist2dawg gle_uncial.numbers gle_uncial.lstm-number-dawg gle_u... | Update_gle_uncial_traineddata_for_Tesseract_4.ipynb | jimregan/tesseract-gle-uncial | apache-2.0 |
SGDClassifier
SGD stands for Stochastic Gradient Descent, a very popular numerical procedure
to find the local minimum of a function (in this case, the loss function, which
measures how far every instance is from our boundary). The algorithm will learn the
coefficients of the hyperplane by minimizing the loss fun... | # instantiate
sgd = SGDClassifier()
# fitting
sgd.fit(X_train, y_train)
# coefficient
print("coefficient", sgd.coef_)
# intercept
print("intercept: ", sgd.intercept_)
# predicting for one
y_pred = sgd.predict(scaler.transform([[4.9,3.1,1.5,0.1]]))
print(y_pred)
# predicting for X_test
y_pred = sgd.predict(X_test)
... | Sklearn_MLPython/CH01.ipynb | atulsingh0/MachineLearning | gpl-3.0 |
Classification Report
Accuracy = (TP+TN)/m
Precision = TP/(TP+FP)
Recall = TP/(TP+FN)
F1-score = 2 * Precision * Recall / (Precision + Recall) | # predicting
print(classification_report(y_pred=y_pred, y_true=y_test))
confusion_matrix(y_pred=y_pred, y_true=y_test) | Sklearn_MLPython/CH01.ipynb | atulsingh0/MachineLearning | gpl-3.0 |
Using a pipeline mechanism to build and test our model | # create a composite estimator made by a pipeline of the standarization and the linear model
clf = pipeline.Pipeline([
('scaler', preprocessing.StandardScaler()),
('linear_model', SGDClassifier())
])
# create a k-fold cross validation iterator of k=5 folds
cv = KFold(X.shape[0], 5, shuffle=True, random... | Sklearn_MLPython/CH01.ipynb | atulsingh0/MachineLearning | gpl-3.0 |
You can use the single line call "analyze," which does all the available analysis simultaneously | # NOTE: This will take several minutes depending on the performance of your machine
audio_features = audioAnalyzer.analyze(audio_filename)
# plot the features
plt.rcParams['figure.figsize'] = [20, 8]
audioAnalyzer.plot(audio_features)
plt.show()
| demos/audio_analysis_demo.ipynb | sertansenturk/tomato | agpl-3.0 |
... or call all the methods individually | # audio metadata extraction
metadata = audioAnalyzer.crawl_musicbrainz_metadata(audio_filename)
# predominant melody extraction
pitch = audioAnalyzer.extract_pitch(audio_filename)
# pitch post filtering
pitch_filtered = audioAnalyzer.filter_pitch(pitch)
# histogram computation
pitch_distribution = audioAnalyzer.comp... | demos/audio_analysis_demo.ipynb | sertansenturk/tomato | agpl-3.0 |
Visualize the MetaLearning pipeline built on top NitroML.
We are using NitroML on Kubeflow:
This notebook allows users to analyze NitroML metalearning pipelines results. | # Step 1: Configure your cluster with gcloud
# `gcloud container clusters get-credentials <cluster_name> --zone <cluster-zone> --project <project-id>
# Step 2: Get the port where the gRPC service is running on the cluster
# `kubectl get configmap metadata-grpc-configmap -o jsonpath={.data}`
# Use `METADATA_GRPC_SERVIC... | examples/visualize_tuner_plots.ipynb | google/nitroml | apache-2.0 |
Connect to the ML Metadata (MLMD) database
First we need to connect to our MLMD database which stores the results of our
benchmark runs. | connection_config = metadata_store_pb2.MetadataStoreClientConfig()
connection_config.host = 'localhost'
connection_config.port = 9898
store = metadata_store.MetadataStore(connection_config) | examples/visualize_tuner_plots.ipynb | google/nitroml | apache-2.0 |
Get trial summary data (used to plot Area under Learning Curve) stored as AugmentedTuner artifacts. | # Name of the dataset/subbenchmark
# This is used to filter out the component path.
testdata = 'ilpd'
def get_metalearning_data(meta_algorithm: str = '', test_dataset: str = '', multiple_runs: bool = True):
d_list = []
execs = store.get_executions_by_type('nitroml.automl.metalearning.tuner.component.Augm... | examples/visualize_tuner_plots.ipynb | google/nitroml | apache-2.0 |
Majority Voting | algorithm = 'majority_voting'
d_list = get_metalearning_data(algorithm, testdata)
d_list
# Select the runs from `d_list` to visualize.
data_list = []
for d in d_list:
with tf.io.gfile.GFile(os.path.join(d, 'tuner_plot_data.txt'), mode='r') as fin:
data_list.append(json.load(fin))
tuner_utils.display_... | examples/visualize_tuner_plots.ipynb | google/nitroml | apache-2.0 |
Nearest Neighbor | algorithm = 'nearest_neighbor'
d_list = get_metalearning_data(algorithm, testdata)
d_list
# Select the runs from `d_list` to visualize.
data_list = []
for d in d_list:
with tf.io.gfile.GFile(os.path.join(d, 'tuner_plot_data.txt'), mode='r') as fin:
data_list.append(json.load(fin))
tuner_utils.display... | examples/visualize_tuner_plots.ipynb | google/nitroml | apache-2.0 |
Representamos ambos diámetros en la misma gráfica | datos.ix[:, "Diametro X":"Diametro Y"].plot(figsize=(16,3),ylim=(1.4,2)).hlines([1.85,1.65],0,3500,colors='r')
datos.ix[:, "Diametro X":"Diametro Y"].boxplot(return_type='axes') | medidas/04082015/estudio.datos.ipynb | darkomen/TFG | cc0-1.0 |
Mostramos la representación gráfica de la media de las muestras | pd.rolling_mean(datos[columns], 50).plot(subplots=True, figsize=(12,12)) | medidas/04082015/estudio.datos.ipynb | darkomen/TFG | cc0-1.0 |
Comparativa de Diametro X frente a Diametro Y para ver el ratio del filamento | plt.scatter(x=datos['Diametro X'], y=datos['Diametro Y'], marker='.') | medidas/04082015/estudio.datos.ipynb | darkomen/TFG | cc0-1.0 |
Filtrado de datos
Las muestras tomadas $d_x >= 0.9$ or $d_y >= 0.9$ las asumimos como error del sensor, por ello las filtramos de las muestras tomadas. | datos_filtrados = datos[(datos['Diametro X'] >= 0.9) & (datos['Diametro Y'] >= 0.9)] | medidas/04082015/estudio.datos.ipynb | darkomen/TFG | cc0-1.0 |
T = 0.05 | L_t0=plotter(T[0]) | 2015_Fall/MATH-578B/Homework2/Homework2.ipynb | saketkc/hatex | mit |
T=10 | L_t1= plotter(T[1]) | 2015_Fall/MATH-578B/Homework2/Homework2.ipynb | saketkc/hatex | mit |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.