{"id":2027,"date":"2019-06-25T09:05:35","date_gmt":"2019-06-25T09:05:35","guid":{"rendered":"https:\/\/www.codeastar.com\/?p=2027"},"modified":"2019-06-25T09:29:11","modified_gmt":"2019-06-25T09:29:11","slug":"save-and-load-your-rnn-model","status":"publish","type":"post","link":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/","title":{"rendered":"Save and Load your RNN model"},"content":{"rendered":"\n<p>In this blog, we tasted different kinds of machine learning projects so far. Our projects included <a href=\"https:\/\/www.codeastar.com\/get-rich-stock-trading-machine-learning\/\">prediction on stock price<\/a>, <a href=\"https:\/\/www.codeastar.com\/convolutional-neural-network-python\/\">image recognizer on hand writing<\/a>, <a href=\"https:\/\/www.codeastar.com\/easy-accuweather-forecast-in-python\/\">NLP on comment classification<\/a> and others. There was one thing in common &#8212; we used long time to train a model. It is okay to use several hours for model training in research projects. But what if we want to use it as a production service? It doesn&#8217;t make sense to tell our clients, &#8220;please wait a few more hours, we are training the model&#8221;. Don&#8217;t worry, we have a solution. Do you remember how do we handle <a href=\"https:\/\/www.codeastar.com\/word-embedding-in-nlp-and-python-part-1\/\">word embedding<\/a> with 600 billion tokens? Yes, we didn&#8217;t train it, we use the pre-trained model. So this is our answer, we train a model, save and load it then use it in production.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Train and Save RNN model<\/h3>\n\n\n\n<p>Before we use a pre-trained model, we need to train a mode. Let&#8217;s use the toxic comment classification project that we did last time as our material. Therefore, we use <a rel=\"noreferrer noopener\" aria-label=\"Recurrent Neural Network (RNN) (opens in a new tab)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Recurrent_neural_network\" target=\"_blank\">Recurrent Neural Network (RNN)<\/a> and <a rel=\"noreferrer noopener\" aria-label=\"word embedding (opens in a new tab)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Word_embedding\" target=\"_blank\">word embedding<\/a> to find out toxic comments. For technical details om RNN and word embedding, please read our posts: <a href=\"https:\/\/www.codeastar.com\/word-embedding-in-nlp-and-python-part-1\/\">NLP and Python Part 1<\/a> and <a href=\"https:\/\/www.codeastar.com\/recurrent-neural-network-rnn-in-nlp-and-python-part-2\/\">Part 2<\/a>. In this post, we focus on how to save and load the RNN model.<\/p>\n\n\n\n<p>In addition to start saving our model, remember to: <\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>enable GPU<\/li><li>get <a rel=\"noreferrer noopener\" aria-label=\"fastText (opens in a new tab)\" href=\"https:\/\/fasttext.cc\/docs\/en\/english-vectors.html\" target=\"_blank\">fastText<\/a> vector file<\/li><li>get <a rel=\"noreferrer noopener\" aria-label=\"GloVe (opens in a new tab)\" href=\"https:\/\/nlp.stanford.edu\/projects\/glove\/\" target=\"_blank\">GloVe<\/a> word file<\/li><\/ul>\n\n\n\n<p>When we are ready, we can load training dataset and pre-process the text.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">import pickle\nfrom nltk.tokenize.treebank import TreebankWordTokenizer\nfrom keras.models import Model, model_from_json\nfrom keras.layers import Input, Dense, Embedding, SpatialDropout1D, add, concatenate,  Dropout\nfrom keras.layers import CuDNNLSTM, Bidirectional, GlobalMaxPooling1D, GlobalAveragePooling1D\nfrom keras.preprocessing import text, sequence\nfrom keras.callbacks import LearningRateScheduler\n\ntrain_df = pd.read_csv('..\/input\/jigsaw-unintended-bias-in-toxicity-classification\/train.csv')\ntrain_df['comment_text'] = train_df['comment_text'].apply(lambda x:preprocess(x))\n<\/pre>\n\n\n\n<p>After that, we fit a tokenizer with training text and save it into a <a rel=\"noreferrer noopener\" aria-label=\"pickle (opens in a new tab)\" href=\"https:\/\/docs.python.org\/3\/library\/pickle.html\" target=\"_blank\">pickle<\/a>. Pickle is a Python model to store a Python object into a byte stream. So we can store the tokenizer to a file, i.e.<em> saved_tokenizer.pickle<\/em> from the code below.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">x_train = train_df[TEXT_COLUMN].astype(str)\ntokenizer = text.Tokenizer(filters=\"\")\ntokenizer.fit_on_texts(list(x_train))\n\nwith open('saved_tokenizer.pickle', 'wb') as handle:\n    pickle.dump(tokenizer, handle, protocol=pickle.HIGHEST_PROTOCOL)\n<\/pre>\n\n\n\n<p>Firstly, we start to run the word embedding process. Please note that we only embed the training text, as in save and load case, we never know what the testing text is.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">x_train = tokenizer.texts_to_sequences(x_train)\nx_train = sequence.pad_sequences(x_train, maxlen=MAX_LEN)\n\nembedding_matrix = np.concatenate(\n    [build_matrix(tokenizer.word_index, f) for f in EMBEDDING_FILES], axis=-1)\n<\/pre>\n\n\n\n<p>Secondly, we build the model in RNN (you can find the complete source at the end of this post) and save the <em>model&#8217; structure<\/em> to a JSON file. <\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">model = build_model(embedding_matrix, y_aux_train.shape[-1])\nmodel_json = model.to_json()\nwith open(\"saved_model.json\", \"w\") as json_file:\n    json_file.write(model_json)\n<\/pre>\n\n\n\n<p>Thirdly, what time is it? It&#8217;s training time! After around 2 hours training time, we save our <em>model weights<\/em> to a <a rel=\"noreferrer noopener\" aria-label=\"HDF5 (opens in a new tab)\" href=\"https:\/\/www.hdfgroup.org\/\" target=\"_blank\">HDF5<\/a> file. <\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">for global_epoch in range(EPOCHS):\n  model.fit(\n            x_train,\n            [y_train, y_aux_train],\n            batch_size=BATCH_SIZE,\n            verbose=2,\n            sample_weight=[sample_weights.values, np.ones_like(sample_weights)],\n            callbacks=[\n                LearningRateScheduler(lambda _: 1e-3 * (0.4 ** global_epoch))\n            ]\n        )\nmodel.save_weights(\"saved_model.h5\")\n<\/pre>\n\n\n\n<p>Actually we can save both model&#8217; structure and weights into a single file, but it is more flexible if we separate them into 2 files. <\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Load and Run RNN model<\/h3>\n\n\n\n<p>Since we have saved our RNN model, it is the time to load the pre-trained model. But before that, let&#8217;s get the test dataset and the tokenizer we saved previously.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">KERAS_PATH = \"keras-93487\"     #fill the path for your saved model\/tokenizer\nTEXT_COLUMN = 'comment_text'\nMAX_LEN = 220\n\ntest_df = pd.read_csv('..\/input\/jigsaw-unintended-bias-in-toxicity-classification\/test.csv')\nwith open(f\"..\/input\/{KERAS_PATH}\/saved_tokenizer.pickle\", 'rb') as handle:\n    tokenizer = pickle.load(handle)\n<\/pre>\n\n\n\n<p>Moreover, we do the same pre-processing and tokenizing using the saved tokenizer.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">test_df[TEXT_COLUMN] = test_df[TEXT_COLUMN].apply(lambda x:preprocess(x))\nx_test = test_df[TEXT_COLUMN].astype(str)\nx_test = tokenizer.texts_to_sequences(x_test)\nx_test = sequence.pad_sequences(x_test, maxlen=MAX_LEN)\n<\/pre>\n\n\n\n<p>Once we have prepared our testing data, we can use the saved model to predict the outcome. <\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">json_file = open(f\"..\/input\/{KERAS_PATH}\/saved_model.json\", 'r')\nmodel_json = json_file.read()\njson_file.close()\nmodel = model_from_json(model_json)\nmodel.load_weights(f\"..\/input\/{KERAS_PATH}\/saved_model.h5\")\nmodel.compile(loss='binary_crossentropy', optimizer='adam')\nprediction = model.predict(x_test, batch_size=2048)[0].flatten()\n<\/pre>\n\n\n\n<p>Let&#8217;s make a dataframe to store our outcome.<\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">validation = pd.DataFrame.from_dict({\n    'id': test_df.id,\n    'comment_text': test_df.comment_text,\n    'prediction': prediction\n})\n\ndef print_sample(df, column, value, sample_size):\n    if (value > 0): \n        df = df[df[column] > value].sample(sample_size)\n    else:\n        df = df[df[column] < abs(value)].sample(sample_size)\n    for index, row in df.iterrows():\n       print(f\"{row['id']} | {row['prediction']:6.4f} | {row['comment_text']} \\n\")\n<\/pre>\n\n\n\n<p>Then, let's take a look on those non-toxic comments first. <\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">print_sample(validation,'prediction', -0.5, 5)\n<\/pre>\n\n\n\n<pre class=\"wp-block-code\"><code>7024785 | 0.0049 | Agreed regarding government . But the BofC just controls overnight rates - short term . Market forces dictate the steepness of the yield curve . And the yield on the GoC 5 year has spiked 15 % in the past 2 days . That rate affects mortgage rates . Inflation is well above the stated 1 . 7 % , and historically the 5 Year yield has been above the rate of inflation . If the market dictates , I ca n't see the B of C containing mid to long term rates . \n\n7037094 | 0.0384 | Trump s nominee for Secretary of Health wants to get rid of Obamacare and privatize Medicare . My wife , who is in very poor health , may well lose her health insurance . Thank you , all who voted for Trump , especially utilitas . \n\n7018832 | 0.1073 | Standard and normal language for a politician engaged in Dog Whistle Politics . Many - though certainly not all - of his supporters seem to need to hear this sort of thing , it s why he talked in his inauguration speech about gangs and drugs and carnage and old factories dotting the landscape like tombstones . He s a real piece of work , unfortunately . \n\n7050923 | 0.0124 | Do they have English - immersion schools in Quebec . ? From what I have heard about the language laws in Quebec , English seems a very poor cousin . In the rest of Canada French immersion is very big and sought after . As one who taught Spanish in a high school in B . C . , I found French immersion students were my best students with the head start they had in dealing with the concept of other languages . \n\n7051534 | 0.2933 | You talking about the most destructive Mayors in the history of Hawaii . \n<\/code><\/pre>\n\n\n\n<p>I can say it is what we have expected. so what about the toxic comments? <\/p>\n\n\n\n<pre lang=\"python\" line=\"1\">print_sample(validation,'prediction', 0.5, 5)\n<\/pre>\n\n\n\n<pre class=\"wp-block-code\"><code>7051483 | 0.5751 | `` My grandmother was a typical white person . `` They cling to their religion and their guns `` He worshipped for years with racist friend and America hater Jeremiah Wright . He mocked special Olympics on Letterman . `` The cops acted stupidly `` `` ISIS is JV `` [ that alone recruited more terrorists for ISIS than Trump s comments ] \n\n7013218 | 0.5867 | Best of luck to you Emers ! You are doing god s work . One day , I firmly believe , the manifold beneficial properties of cannabis will be so revealed that we will only wonder at our stupidity in not fully embracing it sooner . Got me off years of pain medication that had increasingly deleterious effects . \n\n7092639 | 0.7496 | Stop making dumb comments . Please . \n\n7053226 | 0.8995 | What a bunch of idiots . You are being paid by CNN to say that ! \n\n7092559 | 0.5864 | This much we know : Trump is a bully who uses money as a club with which to beat his creditors into submission ; they ca n't afford to litigate their grievances , and he can litigate per omnia saecula saeculorum .  And we know he s lying about his money : He \u2019 s lying about how much money he has . He \u2019 s lying about where it comes from . He \u2019 s lying about where is goes - - especially about his miserly contributions to charity . He \u2019 s lying about , and covering up , his foreign entanglements - - whom he owes and who owes him . <\/code><\/pre>\n\n\n\n<p>In conclusion, we can see comments with negative words are scored 0.5 or above. Once again, the moral standard of machine learning is just a bit high to me :]] .  <\/p>\n\n\n\n<p>After that, by submitting the prediction to Kaggle, we got 0.934x accuracy, which is an improvement from our previous submission :]] . <\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Benefits of using Save and Load Model<\/h3>\n\n\n\n<p>You may ask, what do we get by using save and load model? <\/p>\n\n\n\n<p><em>Time and Portability<\/em>. <\/p>\n\n\n\n<p>We save the long training time, so we only need to handle input testing. From our case above, we only use 70 seconds for handling 97.3k records.<\/p>\n\n\n\n<p>Besides, since we have saved our model in files, we can use the model in other machine. It gives more room for us to think about our strategy in resource planning, knowledge sharing and regional support.<\/p>\n\n\n\n<div style=\"height:140px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">What have we learnt in Save and Load model?<\/h3>\n\n\n\n<ol class=\"wp-block-list\"><li>How to save a Keras RNN model <\/li><li>How to load a Keras RNN model<\/li><li>Benefits of using save and load model<\/li><\/ol>\n\n\n\n<p>(Complete sources can be found at <a rel=\"noreferrer noopener\" aria-label=\"https:\/\/www.kaggle.com\/codeastar\/load-keras-rnn-model (opens in a new tab)\" href=\"https:\/\/www.kaggle.com\/codeastar\/load-keras-rnn-model\" target=\"_blank\">https:\/\/www.kaggle.com\/codeastar\/save-keras-rnn-model<\/a> for model saving and <a href=\"https:\/\/www.kaggle.com\/codeastar\/load-keras-rnn-model\" target=\"_blank\" rel=\"noreferrer noopener\" aria-label=\"https:\/\/www.kaggle.com\/codeastar\/load-keras-rnn-model (opens in a new tab)\">https:\/\/www.kaggle.com\/codeastar\/load-keras-rnn-model<\/a> for model loading)<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>In this blog, we tasted different kinds of machine learning projects so far. Our projects included prediction on stock price, image recognizer on hand writing, NLP on comment classification and others. There was one thing in common &#8212; we used long time to train a model. It is okay to use several hours for model [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":2045,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"site-sidebar-layout":"default","site-content-layout":"default","ast-site-content-layout":"","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"How to save a Keras RNN model and load the pre-trained model for prediction.","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","enabled":false},"version":2}},"categories":[18],"tags":[59,152,153,144,151],"class_list":["post-2027","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-machine-learning","tag-keras","tag-load-model","tag-pickle","tag-rnn","tag-save-model"],"jetpack_publicize_connections":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v24.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Save and Load your RNN model &#8902; Code A Star Learn Machine Learning<\/title>\n<meta name=\"description\" content=\"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Save and Load your RNN model &#8902; Code A Star Learn Machine Learning\" \/>\n<meta property=\"og:description\" content=\"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\" \/>\n<meta property=\"og:site_name\" content=\"Code A Star\" \/>\n<meta property=\"article:publisher\" content=\"codeastar\" \/>\n<meta property=\"article:author\" content=\"codeastar\" \/>\n<meta property=\"article:published_time\" content=\"2019-06-25T09:05:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2019-06-25T09:29:11+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1100\" \/>\n\t<meta property=\"og:image:height\" content=\"400\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Raven Hon\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@codeastar\" \/>\n<meta name=\"twitter:site\" content=\"@codeastar\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Raven Hon\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\"},\"author\":{\"name\":\"Raven Hon\",\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd\"},\"headline\":\"Save and Load your RNN model\",\"datePublished\":\"2019-06-25T09:05:35+00:00\",\"dateModified\":\"2019-06-25T09:29:11+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\"},\"wordCount\":691,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd\"},\"image\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1\",\"keywords\":[\"keras\",\"load model\",\"pickle\",\"RNN\",\"Save model\"],\"articleSection\":[\"Learn Machine Learning\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\",\"url\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\",\"name\":\"Save and Load your RNN model &#8902; Code A Star Learn Machine Learning\",\"isPartOf\":{\"@id\":\"https:\/\/www.codeastar.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1\",\"datePublished\":\"2019-06-25T09:05:35+00:00\",\"dateModified\":\"2019-06-25T09:29:11+00:00\",\"description\":\"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.\",\"breadcrumb\":{\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage\",\"url\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1\",\"contentUrl\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1\",\"width\":1100,\"height\":400,\"caption\":\"Save and Load Model\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.codeastar.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Save and Load your RNN model\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.codeastar.com\/#website\",\"url\":\"https:\/\/www.codeastar.com\/\",\"name\":\"Code A Star\",\"description\":\"We don&#039;t wish upon a star, we code a star\",\"publisher\":{\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.codeastar.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd\",\"name\":\"Raven Hon\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2018\/08\/logo70.png?fit=70%2C70&ssl=1\",\"contentUrl\":\"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2018\/08\/logo70.png?fit=70%2C70&ssl=1\",\"width\":70,\"height\":70,\"caption\":\"Raven Hon\"},\"logo\":{\"@id\":\"https:\/\/www.codeastar.com\/#\/schema\/person\/image\/\"},\"description\":\"Raven Hon is\u00a0a 20 years+ veteran in information technology industry who has worked on various projects from console, web, game, banking and mobile applications in different sized companies.\",\"sameAs\":[\"https:\/\/www.codeastar.com\",\"codeastar\",\"https:\/\/x.com\/codeastar\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Save and Load your RNN model &#8902; Code A Star Learn Machine Learning","description":"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/","og_locale":"en_US","og_type":"article","og_title":"Save and Load your RNN model &#8902; Code A Star Learn Machine Learning","og_description":"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.","og_url":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/","og_site_name":"Code A Star","article_publisher":"codeastar","article_author":"codeastar","article_published_time":"2019-06-25T09:05:35+00:00","article_modified_time":"2019-06-25T09:29:11+00:00","og_image":[{"width":1100,"height":400,"url":"https:\/\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png","type":"image\/png"}],"author":"Raven Hon","twitter_card":"summary_large_image","twitter_creator":"@codeastar","twitter_site":"@codeastar","twitter_misc":{"Written by":"Raven Hon","Est. reading time":"8 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#article","isPartOf":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/"},"author":{"name":"Raven Hon","@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd"},"headline":"Save and Load your RNN model","datePublished":"2019-06-25T09:05:35+00:00","dateModified":"2019-06-25T09:29:11+00:00","mainEntityOfPage":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/"},"wordCount":691,"commentCount":0,"publisher":{"@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd"},"image":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1","keywords":["keras","load model","pickle","RNN","Save model"],"articleSection":["Learn Machine Learning"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/","url":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/","name":"Save and Load your RNN model &#8902; Code A Star Learn Machine Learning","isPartOf":{"@id":"https:\/\/www.codeastar.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage"},"image":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage"},"thumbnailUrl":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1","datePublished":"2019-06-25T09:05:35+00:00","dateModified":"2019-06-25T09:29:11+00:00","description":"It take several hours or days for model training in machine learning. But we got a solution: save and load the pre-trained model and just use it.","breadcrumb":{"@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#primaryimage","url":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1","contentUrl":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1","width":1100,"height":400,"caption":"Save and Load Model"},{"@type":"BreadcrumbList","@id":"https:\/\/www.codeastar.com\/save-and-load-your-rnn-model\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.codeastar.com\/"},{"@type":"ListItem","position":2,"name":"Save and Load your RNN model"}]},{"@type":"WebSite","@id":"https:\/\/www.codeastar.com\/#website","url":"https:\/\/www.codeastar.com\/","name":"Code A Star","description":"We don&#039;t wish upon a star, we code a star","publisher":{"@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.codeastar.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":["Person","Organization"],"@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/832d202eb92a3d430097e88c6d0550bd","name":"Raven Hon","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/image\/","url":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2018\/08\/logo70.png?fit=70%2C70&ssl=1","contentUrl":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2018\/08\/logo70.png?fit=70%2C70&ssl=1","width":70,"height":70,"caption":"Raven Hon"},"logo":{"@id":"https:\/\/www.codeastar.com\/#\/schema\/person\/image\/"},"description":"Raven Hon is\u00a0a 20 years+ veteran in information technology industry who has worked on various projects from console, web, game, banking and mobile applications in different sized companies.","sameAs":["https:\/\/www.codeastar.com","codeastar","https:\/\/x.com\/codeastar"]}]}},"jetpack_featured_media_url":"https:\/\/i0.wp.com\/www.codeastar.com\/wp-content\/uploads\/2019\/06\/save_n_load.png?fit=1100%2C400&ssl=1","jetpack_sharing_enabled":true,"jetpack_shortlink":"https:\/\/wp.me\/p8PcRO-wH","jetpack-related-posts":[],"_links":{"self":[{"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/posts\/2027","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/comments?post=2027"}],"version-history":[{"count":27,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/posts\/2027\/revisions"}],"predecessor-version":[{"id":2057,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/posts\/2027\/revisions\/2057"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/media\/2045"}],"wp:attachment":[{"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/media?parent=2027"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/categories?post=2027"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.codeastar.com\/wp-json\/wp\/v2\/tags?post=2027"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}