Path: blob/master/Sequence Models/Week 2/Emojify/__pycache__/emo_utils.cpython-36.pyc
17771 views
3
��lZf � @ s� d dl Z d dlZd dlZd dlZd dljZd dl m
Z
dd� Z dd� Z ddd �Z
d
d � Zd d
dddd�Zdd� Zdd� Zdejjfdd�Zdd� ZdS ) � N)�confusion_matrixc
C s� t | d���}t� }i }xF|D ]>}|j� j� }|d }|j|� tj|dd � tjd�||< q W d}i }i }x(t|�D ] } ||| < | ||< |d }qtW W d Q R X |||fS )N�rr � )�dtype) �open�set�strip�split�add�np�array�float64�sorted)
Z
glove_file�f�words�word_to_vec_map�lineZ curr_word�iZwords_to_indexZindex_to_words�w� r �-/home/jovyan/work/Week 2/Emojify/emo_utils.py�read_glove_vecs s
r c C s t j| t j| � �}||j� S )z4Compute softmax values for each sets of scores in x.)r �exp�max�sum)�xZe_xr r r �softmax s r �data/emojify_data.csvc C sp g }g }t | ��:}tj|�}x(|D ] }|j|d � |j|d � q"W W d Q R X tj|�}tj|td�}||fS )Nr r )r )r �csv�reader�appendr �asarray�int)�filename�phrase�emojiZ csvDataFileZ csvReader�row�X�Yr r r �read_csv! s
r) c C s t j|�| jd� } | S )Nr �����)r �eye�reshape)r( �Cr r r �convert_to_one_hot1 s r. u ❤️z
:baseball:z:smile:z:disappointed:z:fork_and_knife:)�0�1�2�3�4c C s t jtt| � dd�S )zi
Converts a label (int or string) into the corresponding emoji code (string) ready to be printed
T)Z use_aliases)r% Zemojize�emoji_dictionary�str)�labelr r r �label_to_emoji< s r7 c C s< t � x0t| jd �D ] }t | | tt|| ��� qW d S )Nr )�print�range�shaper7 r" )r' �predr r r r �print_predictionsC s r< zConfusion matrixc C s� t j| |j|jd �dgdgdd�}||jdd� }tj||d� tj� tj t
|j ��}tj ||j d d
� tj
||j� tj|jj� tj|j j� d S ) Nr ZActualZ PredictedT)ZrownamesZcolnames�marginsr )�axis)�cmap�- )Zrotation)�pdZcrosstabr, r: r �pltZmatshowZcolorbarr �arange�len�columnsZxticksZyticks�indexZylabel�nameZxlabel)Zy_actuZy_pred�titler? Z df_confusionZ df_conf_normZ
tick_marksr r r �plot_confusion_matrixI s $rI c
C s� | j d }tj|df�}xvt|�D ]j}| | j� j� }tjd�} x|D ]}
| ||
7 } qFW | t|� } tj|| �| } t| �} tj | �||< q"W t
dt tj |dd� |j
|j d d�dd� k�� � |S )a}
Given X (sentences) and Y (emoji indices), predict emojis and compute the accuracy of your model over the given set.
Arguments:
X -- input data containing sentences, numpy array of shape (m, None)
Y -- labels, containing index of the label emoji, numpy array of shape (m, 1)
Returns:
pred -- numpy array of shape (m, 1) with your predictions
r r �2 z
Accuracy: N)rJ )r: r �zerosr9 �lowerr rD �dotr �argmaxr8 r5 �meanr, )
r' r( �W�br �mr; �jr �avgr �Z�Ar r r �predictZ s
8rW )r )r �numpyr r% �pandasrA �matplotlib.pyplot�pyplotrB Zsklearn.metricsr r r r) r. r4 r7 r< �cmZgray_rrI rW r r r r �<module> s$