{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Machine Learning Final Project\n\n*Student name: Cao Đình Hoàng Minh*\n\n*MSSV: 19020063*\n\n*Class ID: INT3405E_20*","metadata":{}},{"cell_type":"markdown","source":"# 1. Problem Descriptions\n- Quora is a question and answer website where people go to find information. Every piece of content on the site is generated by users, meaning it is created, edited, and organized by the same people that use the website.\n- Quora has some qualities that those other research tools don’t:\n  + It allows users to create social networks and follow topics that interest them. \n  + It focuses on high-quality questions and answers. \n  + It enable users to vote on answers to highlight the most accurate information possible. \n- As being in top 10 most visited social networks in the world, censoring the content of questions is also an extremely important factor to make the value of this social network\n- This problem focus on determining whether a question asked on Quora is insincere or not. The input is a text string and the output is 0 (sincere) or 1 (insincere)","metadata":{}},{"cell_type":"markdown","source":"# 2. Data analyzing, cleaning & pre-processing\n**a. Data analyzing**\n- First we need to import some libraries and modules","metadata":{}},{"cell_type":"code","source":"# Just for checking folders \nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))      \n        \n# Prevent all kinds of warning in all output \nimport warnings\nwarnings.filterwarnings('ignore')\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport matplotlib.pyplot as plt # plot some graphs\nfrom wordcloud import WordCloud # for words statistics\n\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom keras import layers\nfrom keras.preprocessing.text import Tokenizer\nfrom keras.preprocessing.sequence import pad_sequences","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2022-01-08T15:56:55.598150Z","iopub.execute_input":"2022-01-08T15:56:55.598560Z","iopub.status.idle":"2022-01-08T15:57:00.321490Z","shell.execute_reply.started":"2022-01-08T15:56:55.598462Z","shell.execute_reply":"2022-01-08T15:57:00.320732Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Then we read the necessary csv files","metadata":{}},{"cell_type":"code","source":"# Training data\ntrain_data = pd.read_csv(\"../input/quora-insincere-questions-classification/train.csv\")\n# Testing data\ntest_data = pd.read_csv(\"../input/quora-insincere-questions-classification/test.csv\")","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:00.323578Z","iopub.execute_input":"2022-01-08T15:57:00.323955Z","iopub.status.idle":"2022-01-08T15:57:05.693277Z","shell.execute_reply.started":"2022-01-08T15:57:00.323917Z","shell.execute_reply":"2022-01-08T15:57:05.692460Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- We can see that there are 1306122 questions in the training set and 375806 questions on the test set. All of them are not null","metadata":{}},{"cell_type":"code","source":"# Show some information \ntrain_data.info()\ntest_data.info()","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:05.696756Z","iopub.execute_input":"2022-01-08T15:57:05.696986Z","iopub.status.idle":"2022-01-08T15:57:06.059019Z","shell.execute_reply.started":"2022-01-08T15:57:05.696959Z","shell.execute_reply":"2022-01-08T15:57:06.058292Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Let's take a look at those files","metadata":{}},{"cell_type":"code","source":"train_data.head(20)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:06.060984Z","iopub.execute_input":"2022-01-08T15:57:06.061382Z","iopub.status.idle":"2022-01-08T15:57:06.077477Z","shell.execute_reply.started":"2022-01-08T15:57:06.061346Z","shell.execute_reply":"2022-01-08T15:57:06.076680Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- In the train file we have 3 columns:\n  + qid: The question id\n  + question_text: The question we need to predict\n  + target: Mark the question as sincere (0) or insincere (1)","metadata":{}},{"cell_type":"code","source":"test_data.head(20)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:06.078921Z","iopub.execute_input":"2022-01-08T15:57:06.079170Z","iopub.status.idle":"2022-01-08T15:57:06.088749Z","shell.execute_reply.started":"2022-01-08T15:57:06.079137Z","shell.execute_reply":"2022-01-08T15:57:06.087939Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- In the test file we have 2 columns:\n  + qid: The question id\n  + question_text: The question we need to and predict","metadata":{}},{"cell_type":"markdown","source":"- Let's count the number of each kind of questions","metadata":{}},{"cell_type":"code","source":"sincere_questions = train_data[train_data['target']==0]\ninsincere_questions = train_data[train_data['target']==1]\nnum_of_sincere = sincere_questions.shape[0]\nnum_of_insincere = insincere_questions.shape[0]\nprint(\"Number Of Sincere Questions:\", num_of_sincere)\nprint(\"Number Of Insincere Questions:\", num_of_insincere)\nquantity = [num_of_sincere, num_of_insincere]\nlabels = ['Sincere Questions', 'Insincere Questions']\nplt.pie(quantity, labels=labels, autopct='%1.2f%%', shadow=False)\nplt.title('Target Distribution')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:06.090356Z","iopub.execute_input":"2022-01-08T15:57:06.090737Z","iopub.status.idle":"2022-01-08T15:57:06.289234Z","shell.execute_reply.started":"2022-01-08T15:57:06.090703Z","shell.execute_reply":"2022-01-08T15:57:06.288437Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Bad news here, we have too many sincere questions in the training set, which can badly affect the model's accuracy. To prevent this, we should use f1-score, which is the harmonic mean of precision and recall\n- Precision is the ability of a model to identify only the insincere questions\n- Recall is the ability of a model to find all the insincere questions\n\n$$Precision = \\frac{True Positive}{True Positive + False Positive}$$\n\n$$Recall = \\frac{True Positive}{True Positive + False Negative}$$\n\n$$F1 = \\frac{2}{Recall^{-1} + Precision^{-1}}$$\n\n\n- Before we are going to clean the data, we will see which words appear the most in each kinds of questions","metadata":{}},{"cell_type":"code","source":"sincere_wordcloud = WordCloud(width=800, height=450, background_color='white', min_font_size=10).generate(\" \".join(sincere_questions.question_text))\nplt.figure(figsize=(16,9), facecolor=None)\nplt.imshow(sincere_wordcloud)\nplt.axis(\"off\")\nplt.title(\"Common Words in Sincere Questions\", fontsize=30,color='k')\nplt.tight_layout(pad=0)\nplt.show();","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:06.290780Z","iopub.execute_input":"2022-01-08T15:57:06.291042Z","iopub.status.idle":"2022-01-08T15:57:58.965667Z","shell.execute_reply.started":"2022-01-08T15:57:06.291008Z","shell.execute_reply":"2022-01-08T15:57:58.964921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"insincere_wordcloud = WordCloud(width=800, height=450, background_color='white', min_font_size=10).generate(\" \".join(insincere_questions.question_text))\nplt.figure(figsize=(16,9), facecolor=None)\nplt.imshow(insincere_wordcloud)\nplt.axis(\"off\")\nplt.title(\"Common Words in Insincere Questions\", fontsize=30,color='k')\nplt.tight_layout(pad=0)\nplt.show();","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:57:58.966688Z","iopub.execute_input":"2022-01-08T15:57:58.966912Z","iopub.status.idle":"2022-01-08T15:58:05.385903Z","shell.execute_reply.started":"2022-01-08T15:57:58.966884Z","shell.execute_reply":"2022-01-08T15:58:05.381660Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- As we can see there are some offensive words (pornography, racist, political,..) appearing a lot in the insincere questions.","metadata":{}},{"cell_type":"markdown","source":"**b. Data cleaning**\n- First we need some arrays to store the main data","metadata":{}},{"cell_type":"code","source":"# 4/5 of the questions wil be used to train\n# the rest of them are used for validations\ntrain_ratio = 0.8 \nnum_of_train  = int(train_ratio * (num_of_sincere + num_of_insincere))\ntrain_sen = [] # array of training questions\nval_sen = [] # array of validating questions\ntest_sen = [] # array of testing questions\n\nfor i in range(0, len(train_data['question_text'])):\n    if i < num_of_train:\n        train_sen.append(train_data['question_text'].loc[i])\n    else:\n        val_sen.append(train_data['question_text'].loc[i])\n        \nfor i in range(0, len(test_data['question_text'])):\n    test_sen.append(test_data['question_text'].loc[i])\n\ntrain_label = [] # array of training questions' labels\nval_label = [] # array of validating questions' labels\nfor i in range(0, len(train_data['target'])):\n    if i < num_of_train:\n        train_label.append(float(train_data['target'].loc[i]))\n    else:\n        val_label.append(float(train_data['target'].loc[i]))","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:58:05.387322Z","iopub.execute_input":"2022-01-08T15:58:05.388136Z","iopub.status.idle":"2022-01-08T15:58:55.624374Z","shell.execute_reply.started":"2022-01-08T15:58:05.388093Z","shell.execute_reply":"2022-01-08T15:58:55.623661Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Stop words are the words which are generally filtered out before processing a natural language are called stop words. These are actually the most common words in any language (like articles, prepositions, pronouns, conjunctions, etc) and does not add much information to the text.","metadata":{}},{"cell_type":"code","source":"allStopWords = \"x y your yours yourself yourselves you yond yonder yon ye yet z zillion j u umpteen usually us username uponed upons uponing upon ups upping upped up unto until unless unlike unliker unlikest under underneath use used usedest r rath rather rathest rathe re relate related relatively regarding really res respecting respectively q quite que qua n neither neaths neath nethe nethermost necessary necessariest necessarier never nevertheless nigh nighest nigher nine noone nobody nobodies nowhere nowheres no noes nor nos no-one none not notwithstanding nothings nothing nathless natheless t ten tills till tilled tilling to towards toward towardest towarder together too thy thyself thus than that those thou though thous thouses thoroughest thorougher thorough thoroughly thru thruer thruest thro through throughout throughest througher thine this thises they thee the then thence thenest thener them themselves these therer there thereby therest thereafter therein thereupon therefore their theirs thing things three two o oh owt owning owned own owns others other otherwise otherwisest otherwiser of often oftener oftenest off offs offest one ought oughts our ours ourselves ourself out outest outed outwith outs outside over overallest overaller overalls overall overs or orer orest on oneself onest ons onto a atween at athwart atop afore afterward afterwards after afterest afterer ain an any anything anybody anyone anyhow anywhere anent anear and andor another around ares are aest aer against again accordingly abaft abafter abaftest abovest above abover abouter aboutest about aid amidst amid among amongst apartest aparter apart appeared appears appear appearing appropriating appropriate appropriatest appropriates appropriater appropriated already always also along alongside although almost all allest aller allyou alls albeit awfully as aside asides aslant ases astrider astride astridest astraddlest astraddler astraddle availablest availabler available aughts aught vs v variousest variouser various via vis-a-vis vis-a-viser vis-a-visest viz very veriest verier versus k g go gone good got gotta gotten get gets getting b by byandby by-and-by bist both but buts be beyond because became becomes become becoming becomings becominger becomingest behind behinds before beforehand beforehandest beforehander bettered betters better bettering betwixt between beneath been below besides beside m my myself mucher muchest much must musts musths musth main make mayest many mauger maugre me meanwhiles meanwhile mostly most moreover more might mights midst midsts h huh humph he hers herself her hereby herein hereafters hereafter hereupon hence hadst had having haves have has hast hardly hae hath him himself hither hitherest hitherer his how-do-you-do however how howbeit howdoyoudo hoos hoo w woulded woulding would woulds was wast we wert were with withal without within why what whatever whateverer whateverest whatsoeverer whatsoeverest whatsoever whence whencesoever whenever whensoever when whenas whether wheen whereto whereupon wherever whereon whereof where whereby wherewithal wherewith whereinto wherein whereafter whereas wheresoever wherefrom which whichever whichsoever whilst while whiles whithersoever whither whoever whosoever whoso whose whomever s syne syn shalling shall shalled shalls shoulding should shoulded shoulds she sayyid sayid said saider saidest same samest sames samer saved sans sanses sanserifs sanserif so soer soest sobeit someone somebody somehow some somewhere somewhat something sometimest sometimes sometimer sometime several severaler severalest serious seriousest seriouser senza send sent seem seems seemed seemingest seeminger seemings seven summat sups sup supping supped such since sine sines sith six stop stopped p plaintiff plenty plenties please pleased pleases per perhaps particulars particularly particular particularest particularer pro providing provides provided provide probably l layabout layabouts latter latterest latterer latterly latters lots lotting lotted lot lest less ie ifs if i info information itself its it is idem idemer idemest immediate immediately immediatest immediater in inwards inwardest inwarder inward inasmuch into instead insofar indicates indicated indicate indicating indeed inc f fact facts fs figupon figupons figuponing figuponed few fewer fewest frae from failing failings five furthers furtherer furthered furtherest further furthering furthermore fourscore followthrough for forwhy fornenst formerly former formerer formerest formers forbye forby fore forever forer fores four d ddays dday do doing doings doe does doth downwarder downwardest downward downwards downs done doner dones donest dos dost did differentest differenter different describing describe describes described despiting despites despited despite during c cum circa chez cer certain certainest certainer cest canst cannot cant cants canting cantest canted co could couldst comeon comeons come-ons come-on concerning concerninger concerningest consequently considering e eg eight either even evens evenser evensest evened evenest ever everyone everything everybody everywhere every ere each et etc elsewhere else ex excepted excepts except excepting exes enough\"\nstopWords = allStopWords.split(\" \")","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:58:55.627365Z","iopub.execute_input":"2022-01-08T15:58:55.627762Z","iopub.status.idle":"2022-01-08T15:58:55.634647Z","shell.execute_reply.started":"2022-01-08T15:58:55.627723Z","shell.execute_reply":"2022-01-08T15:58:55.633997Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- The below function can help filtering out the stop words for better performances","metadata":{}},{"cell_type":"code","source":"# Paramater: String, for example \"I like machine learning\"\ndef removeStop(sentence): \n    words = sentence.split(\" \")\n    ans = \"\"\n    for s in words:\n        try: \n            index = stopWords.index(s)\n        except:\n            ans += (s + \" \")\n    return ans","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:58:55.635705Z","iopub.execute_input":"2022-01-08T15:58:55.635983Z","iopub.status.idle":"2022-01-08T15:58:55.649363Z","shell.execute_reply.started":"2022-01-08T15:58:55.635949Z","shell.execute_reply":"2022-01-08T15:58:55.648581Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Then we apply this method on all sentences to complete the cleaning phase","metadata":{}},{"cell_type":"code","source":"for i in range(0, len(train_sen)):\n    if not isinstance(train_sen[i], str):\n        train_sen[i] = str(train_sen[i])\n    train_sen[i] = removeStop(train_sen[i])\n\nfor i in range(0, len(val_sen)):\n    if not isinstance(val_sen[i], str):\n        val_sen[i] = str(val_sen[i])\n    val_sen[i] = removeStop(val_sen[i])\n\nfor i in range(0, len(test_sen)):\n    if not isinstance(test_sen[i], str):\n        test_sen[i] = str(test_sen[i])\n    test_sen[i] = removeStop(test_sen[i])","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:58:55.650638Z","iopub.execute_input":"2022-01-08T15:58:55.650888Z","iopub.status.idle":"2022-01-08T16:01:59.136775Z","shell.execute_reply.started":"2022-01-08T15:58:55.650856Z","shell.execute_reply":"2022-01-08T16:01:59.136013Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**c. Data pre-processing**\n- To help the model understand those questions, we need to convert those questions into some sequences of numbers by using Tokenizer","metadata":{}},{"cell_type":"code","source":"vocab_size = 30000 # the maximum number of words to keep, based on word frequency. Only the most common 30000 words will be kept\nembedding_dim = 64\nmax_length = 64 # Most questions have less than 64 words, so it can be set to the maximum length of all sequences\ntrunc_type = 'post' # Remove values from sequences larger than maxlen at the end of the sequences.\npad_type = 'post' # Pad after each sequence, if it's not long enough\noov_tok = \"<OOV>\" # If given, it will be added to word_index and used to replace out-of-vocabulary words during text_to_sequence calls","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:01:59.138114Z","iopub.execute_input":"2022-01-08T16:01:59.138489Z","iopub.status.idle":"2022-01-08T16:01:59.142802Z","shell.execute_reply.started":"2022-01-08T16:01:59.138408Z","shell.execute_reply":"2022-01-08T16:01:59.141996Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tokenizer = Tokenizer(num_words = vocab_size, oov_token=oov_tok)\ntokenizer.fit_on_texts(train_sen)\n\ntrain_seq = tokenizer.texts_to_sequences(train_sen)\ntrain_pad = pad_sequences(train_seq, maxlen=max_length, padding=pad_type, truncating=trunc_type)\n\nval_seq = tokenizer.texts_to_sequences(val_sen)\nval_pad = pad_sequences(val_seq, maxlen=max_length, padding=pad_type, truncating=trunc_type)\n\ntest_seq = tokenizer.texts_to_sequences(test_sen)\ntest_pad = pad_sequences(test_seq, maxlen=max_length, padding=pad_type, truncating=trunc_type)\n\ninput1 = keras.Input(len(train_pad[0]))\n\nembedded_vector = layers.Embedding(vocab_size, embedding_dim)(input1)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:01:59.143859Z","iopub.execute_input":"2022-01-08T16:01:59.144668Z","iopub.status.idle":"2022-01-08T16:02:49.639414Z","shell.execute_reply.started":"2022-01-08T16:01:59.144633Z","shell.execute_reply":"2022-01-08T16:02:49.638712Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 3. Model Description\n**a. Select Model**\n- The model selected here is Long Short-Term Memory (LTSM), which is an artificial Recurrent Neural Network (RNN) architecture used in the field of deep learning. Unlike standard feedforward neural networks, LSTM has feedback connections. It can process not only single data points (such as images), but also entire sequences of data (text)\n- LSTM is still an RNN in essence, but the cycle a in the figure above has been redesigned to solve the problem of memory time is not long enough. Other neural networks try to adjust parameters to make memory better. As a result, LSTM is born to remember, which is a blow to dimension reduction!\n- A in the ordinary RNN is shown in the figure below. The previous input and the current input are operated once. Tanh is used in the figure\n\n![image.png](attachment:bedd8f8e-6a3a-49ba-a98f-213c4baf5969.png)\n\n- In comparison, a in LSTM is much more complex. It is not a single neural network layer in the figure above, but has four layers, as shown in the figure below. =\n\n![image.png](attachment:a852aa5b-613a-4f6f-9d56-6ceb40ffe9ae.png)","metadata":{},"attachments":{"bedd8f8e-6a3a-49ba-a98f-213c4baf5969.png":{"image/png":"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"},"a852aa5b-613a-4f6f-9d56-6ceb40ffe9ae.png":{"image/png":"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"}}},{"cell_type":"code","source":"hidden_class_1 = layers.LSTM(64)(embedded_vector)\nhidden_class_1 = keras.Model(inputs=input1, outputs=hidden_class_1)\n\nhidden_class = hidden_class_1.output\noutput = layers.Dense(2, activation='softmax')(hidden_class) # Use softmax activation function\n\nmodel = keras.Model(inputs=[hidden_class_1.input], outputs=output)\n\nmodel.summary()\ntrain_label = np.array(train_label)\nval_label = np.array(val_label)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:02:49.640530Z","iopub.execute_input":"2022-01-08T16:02:49.641342Z","iopub.status.idle":"2022-01-08T16:02:50.059004Z","shell.execute_reply.started":"2022-01-08T16:02:49.641296Z","shell.execute_reply":"2022-01-08T16:02:50.058297Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**b. Train Model**\n- Since metrics have been removed from Keras core, only accuracy and loss are available, we need to calculate the precision, recall and f1-score manually","metadata":{}},{"cell_type":"code","source":"from keras import backend as K\n\ndef recall_m(y_true, y_pred):\n    true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\n    possible_positives = K.sum(K.round(K.clip(y_true, 0, 1)))\n    recall = true_positives / (possible_positives + K.epsilon()) # Prevent from dividing by 0\n    return recall\n\ndef precision_m(y_true, y_pred):\n    true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\n    predicted_positives = K.sum(K.round(K.clip(y_pred, 0, 1)))\n    precision = true_positives / (predicted_positives + K.epsilon()) # Prevent from dividing by 0\n    return precision\n\ndef f1_m(y_true, y_pred):\n    precision = precision_m(y_true, y_pred)\n    recall = recall_m(y_true, y_pred)\n    return 2*((precision*recall)/(precision+recall+K.epsilon())) # Prevent from dividing by 0","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:02:50.060211Z","iopub.execute_input":"2022-01-08T16:02:50.060624Z","iopub.status.idle":"2022-01-08T16:02:50.070381Z","shell.execute_reply.started":"2022-01-08T16:02:50.060586Z","shell.execute_reply":"2022-01-08T16:02:50.069685Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- We are ready to train now","metadata":{}},{"cell_type":"code","source":"loss_function = keras.losses.SparseCategoricalCrossentropy()\nmodel.compile(optimizer='Adam', loss=loss_function, metrics=['accuracy',recall_m,precision_m,f1_m])\nhistory = model.fit(x=[train_pad], y=train_label, batch_size=32, epochs=2, validation_data=(val_pad, val_label))","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:23:39.111164Z","iopub.execute_input":"2022-01-08T16:23:39.111703Z","iopub.status.idle":"2022-01-08T16:43:18.865599Z","shell.execute_reply.started":"2022-01-08T16:23:39.111662Z","shell.execute_reply":"2022-01-08T16:43:18.864812Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**c. Validation**","metadata":{}},{"cell_type":"markdown","source":"- Now we will test the model on validation set","metadata":{}},{"cell_type":"code","source":"validation_check = model.evaluate(x=[val_pad], y=val_label, verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:22:17.044432Z","iopub.execute_input":"2022-01-08T16:22:17.045458Z","iopub.status.idle":"2022-01-08T16:22:46.720629Z","shell.execute_reply.started":"2022-01-08T16:22:17.045382Z","shell.execute_reply":"2022-01-08T16:22:46.719923Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**d. Predict the output**\n- Now we will use this trained model to predict the questions in the test file","metadata":{}},{"cell_type":"code","source":"pred_y = model.predict([test_pad], batch_size=1024, verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:44:33.099513Z","iopub.execute_input":"2022-01-08T16:44:33.100033Z","iopub.status.idle":"2022-01-08T16:44:34.738107Z","shell.execute_reply.started":"2022-01-08T16:44:33.099996Z","shell.execute_reply":"2022-01-08T16:44:34.737360Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- The pred_y is representing the probability of each label (first number: label 0, second number: label 1), not 0 or 1\n- Therefore we need one more step to convert it to an array of 0&1 to complete the prediction and submit the csv file","metadata":{}},{"cell_type":"code","source":"ans = []\nfor prediction in pred_y:\n    ans.append((prediction[1]*2).astype(int)) # if the value of label 1 is not less than 0.5, the model will predict it's 1\ndata_submit = pd.DataFrame({\"qid\":test_data[\"qid\"].values})\ndata_submit['prediction'] = ans\ndata_submit.to_csv(\"submission.csv\", index=False)\ndata_submit","metadata":{"execution":{"iopub.status.busy":"2022-01-08T16:44:34.739637Z","iopub.execute_input":"2022-01-08T16:44:34.739872Z","iopub.status.idle":"2022-01-08T16:44:38.217891Z","shell.execute_reply.started":"2022-01-08T16:44:34.739840Z","shell.execute_reply":"2022-01-08T16:44:38.217192Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**e. Summary by graph**","metadata":{}},{"cell_type":"code","source":"# Graph of model loss\nimport matplotlib.pyplot as plt # plot some graphs\nplt.plot(history.history['loss'])\nplt.plot(history.history['val_loss'])\nplt.title('model loss')\nplt.ylabel('loss')\nplt.xlabel('epoch')\nplt.legend(['train', 'val'])\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-01-08T17:16:39.282238Z","iopub.execute_input":"2022-01-08T17:16:39.282541Z","iopub.status.idle":"2022-01-08T17:16:39.504336Z","shell.execute_reply.started":"2022-01-08T17:16:39.282509Z","shell.execute_reply":"2022-01-08T17:16:39.503616Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Graph of training accuracy\nplt.plot(history.history['accuracy'])\nplt.plot(history.history['val_accuracy'])\nplt.title('model accuracy')\nplt.ylabel('accuracy')\nplt.xlabel('epoch')\nplt.legend(['train', 'val'])\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-01-08T17:51:51.270886Z","iopub.execute_input":"2022-01-08T17:51:51.271709Z","iopub.status.idle":"2022-01-08T17:51:51.369031Z","shell.execute_reply.started":"2022-01-08T17:51:51.271578Z","shell.execute_reply":"2022-01-08T17:51:51.367456Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Graph of f1\nplt.plot(history.history['f1_m'])\nplt.plot(history.history['val_f1_m'])\nplt.title('model f1')\nplt.ylabel('f1_m')\nplt.xlabel('epoch')\nplt.legend(['train', 'val'])\nplt.show()","metadata":{},"execution_count":null,"outputs":[]}]}