Machine Translation (English to French) RNN_LSTM.ipynb

{"cells":[{"metadata":{},"cell_type":"markdown","source":"## What is Machine Translation?\nMachine translation (MT) refers to fully automated software that can translate source content into target languages. Humans may use MT to help them render text and speech into another language, or the MT software may operate without human intervention.\n\n![image.png](attachment:image.png)\n\nMT tools are often used to translate vast amounts of information involving millions of words that could not possibly be translated the traditional way. The quality of MT output can vary considerably; MT systems require “training” in the desired domain and language pair to increase quality.\n\nTranslation companies use MT to augment productivity of their translators, cut costs, and provide post-editing services to clients. MT use by language service providers is growing quickly. \n\nSource: https://www.gala-global.org/what-machine-translation","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"This notebook is an implementation of the article : https://blog.keras.io/a-ten-minute-introduction-to-sequence-to-sequence-learning-in-keras.html"},{"metadata":{},"cell_type":"markdown","source":"## Let's start the process of building a model which translates English to French"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 5GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":8,"outputs":[{"output_type":"stream","text":"/kaggle/input/fra.txt\n/kaggle/input/_about.txt\n","name":"stdout"}]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"from tensorflow.keras.models import Model\nfrom tensorflow.keras.layers import Input,LSTM,Dense\n\nbatch_size=64\nepochs=100\nlatent_dim=256\nnum_samples=10000\n\ndata_path='../input/fra.txt'","execution_count":15,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Vectorize the data.\ninput_texts = []\ntarget_texts = []\ninput_characters = set()\ntarget_characters = set()\nwith open(data_path, 'r', encoding='utf-8') as f:\n    lines = f.read().split('\\n')\nfor line in lines[: min(num_samples, len(lines) - 1)]:\n    input_text, target_text, _ = line.split('\\t')\n    # We use \"tab\" as the \"start sequence\" character\n    # for the targets, and \"\\n\" as \"end sequence\" character.\n    target_text = '\\t' + target_text + '\\n'\n    input_texts.append(input_text)\n    target_texts.append(target_text)\n    for char in input_text:\n        if char not in input_characters:\n            input_characters.add(char)\n    for char in target_text:\n        if char not in target_characters:\n            target_characters.add(char)","execution_count":16,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"input_characters=sorted(list(input_characters))\ntarget_characters=sorted(list(target_characters))\n\nnum_encoder_tokens=len(input_characters)\nnum_decoder_tokens=len(target_characters)\n\nmax_encoder_seq_length=max([len(txt) for txt in input_texts])\nmax_decoder_seq_length=max([len(txt) for txt in target_texts])","execution_count":17,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print('Number of samples:', len(input_texts))\nprint('Number of unique input tokens:', num_encoder_tokens)\nprint('Number of unique output tokens:', num_decoder_tokens)\nprint('Max sequence length for inputs:', max_encoder_seq_length)\nprint('Max sequence length for outputs:', max_decoder_seq_length)","execution_count":22,"outputs":[{"output_type":"stream","text":"Number of samples: 10000\nNumber of unique input tokens: 71\nNumber of unique output tokens: 92\nMax sequence length for inputs: 16\nMax sequence length for outputs: 59\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"input_token_index=dict(\n    [(char,i) for i, char in enumerate(input_characters)])\ntarget_token_index=dict(\n[(char,i) for i, char in enumerate(target_characters)])","execution_count":18,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"encoder_input_data = np.zeros(\n    (len(input_texts), max_encoder_seq_length, num_encoder_tokens),\n    dtype='float32')\ndecoder_input_data = np.zeros(\n    (len(input_texts), max_decoder_seq_length, num_decoder_tokens),\n    dtype='float32')\ndecoder_target_data = np.zeros(\n    (len(input_texts), max_decoder_seq_length, num_decoder_tokens),\n    dtype='float32')","execution_count":20,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for i, (input_text, target_text) in enumerate(zip(input_texts, target_texts)):\n    for t, char in enumerate(input_text):\n        encoder_input_data[i, t, input_token_index[char]] = 1.\n    encoder_input_data[i, t + 1:, input_token_index[' ']] = 1.\n    for t, char in enumerate(target_text):\n        # decoder_target_data is ahead of decoder_input_data by one timestep\n        decoder_input_data[i, t, target_token_index[char]] = 1.\n        if t > 0:\n            # decoder_target_data will be ahead by one timestep\n            # and will not include the start character.\n            decoder_target_data[i, t - 1, target_token_index[char]] = 1.\n    decoder_input_data[i, t + 1:, target_token_index[' ']] = 1.\n    decoder_target_data[i, t:, target_token_index[' ']] = 1.","execution_count":21,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Defining the encoder and decoder"},{"metadata":{"trusted":true},"cell_type":"code","source":"# Define an input sequence and process it.\nencoder_inputs = Input(shape=(None, num_encoder_tokens))\nencoder = LSTM(latent_dim, return_state=True)\nencoder_outputs, state_h, state_c = encoder(encoder_inputs)\n# We discard `encoder_outputs` and only keep the states.\nencoder_states = [state_h, state_c]\n\n# Set up the decoder, using `encoder_states` as initial state.\ndecoder_inputs = Input(shape=(None, num_decoder_tokens))\n# We set up our decoder to return full output sequences,\n# and to return internal states as well. We don't use the\n# return states in the training model, but we will use them in inference.\ndecoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True)\ndecoder_outputs, _, _ = decoder_lstm(decoder_inputs,\n                                     initial_state=encoder_states)\ndecoder_dense = Dense(num_decoder_tokens, activation='softmax')\ndecoder_outputs = decoder_dense(decoder_outputs)\n","execution_count":27,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Define the model that will turn\n# `encoder_input_data` & `decoder_input_data` into `decoder_target_data`\nmodel = Model([encoder_inputs, decoder_inputs], decoder_outputs)\n\n# Run training\nmodel.compile(optimizer='rmsprop', loss='categorical_crossentropy',\n              metrics=['accuracy'])\nmodel.fit([encoder_input_data, decoder_input_data], decoder_target_data,\n          batch_size=batch_size,\n          epochs=epochs,\n          validation_split=0.2)","execution_count":28,"outputs":[{"output_type":"stream","text":"Epoch 1/100\n125/125 [==============================] - 3s 21ms/step - loss: 1.1876 - accuracy: 0.7237 - val_loss: 1.0682 - val_accuracy: 0.6998\nEpoch 2/100\n125/125 [==============================] - 2s 17ms/step - loss: 0.8543 - accuracy: 0.7684 - val_loss: 0.8437 - val_accuracy: 0.7628\nEpoch 3/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.6907 - accuracy: 0.8059 - val_loss: 0.7236 - val_accuracy: 0.7932\nEpoch 4/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.5982 - accuracy: 0.8270 - val_loss: 0.6592 - val_accuracy: 0.8069\nEpoch 5/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.5491 - accuracy: 0.8397 - val_loss: 0.6195 - val_accuracy: 0.8190\nEpoch 6/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.5130 - accuracy: 0.8492 - val_loss: 0.5839 - val_accuracy: 0.8274\nEpoch 7/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.4842 - accuracy: 0.8572 - val_loss: 0.5626 - val_accuracy: 0.8342\nEpoch 8/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.4631 - accuracy: 0.8626 - val_loss: 0.5395 - val_accuracy: 0.8408\nEpoch 9/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.4410 - accuracy: 0.8685 - val_loss: 0.5269 - val_accuracy: 0.8434\nEpoch 10/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.4222 - accuracy: 0.8738 - val_loss: 0.5132 - val_accuracy: 0.8468\nEpoch 11/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.4052 - accuracy: 0.8787 - val_loss: 0.4949 - val_accuracy: 0.8527\nEpoch 12/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3895 - accuracy: 0.8832 - val_loss: 0.4874 - val_accuracy: 0.8556\nEpoch 13/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.3744 - accuracy: 0.8880 - val_loss: 0.4759 - val_accuracy: 0.8585\nEpoch 14/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3610 - accuracy: 0.8918 - val_loss: 0.4707 - val_accuracy: 0.8597\nEpoch 15/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.3478 - accuracy: 0.8954 - val_loss: 0.4622 - val_accuracy: 0.8633\nEpoch 16/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3355 - accuracy: 0.8989 - val_loss: 0.4605 - val_accuracy: 0.8631\nEpoch 17/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3243 - accuracy: 0.9023 - val_loss: 0.4543 - val_accuracy: 0.8664\nEpoch 18/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3136 - accuracy: 0.9055 - val_loss: 0.4529 - val_accuracy: 0.8668\nEpoch 19/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.3028 - accuracy: 0.9087 - val_loss: 0.4452 - val_accuracy: 0.8687\nEpoch 20/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2929 - accuracy: 0.9117 - val_loss: 0.4431 - val_accuracy: 0.8705\nEpoch 21/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2838 - accuracy: 0.9142 - val_loss: 0.4464 - val_accuracy: 0.8700\nEpoch 22/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2744 - accuracy: 0.9171 - val_loss: 0.4428 - val_accuracy: 0.8727\nEpoch 23/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2658 - accuracy: 0.9196 - val_loss: 0.4432 - val_accuracy: 0.8720\nEpoch 24/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2576 - accuracy: 0.9221 - val_loss: 0.4437 - val_accuracy: 0.8719\nEpoch 25/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2494 - accuracy: 0.9244 - val_loss: 0.4456 - val_accuracy: 0.8731\nEpoch 26/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2418 - accuracy: 0.9265 - val_loss: 0.4433 - val_accuracy: 0.8742\nEpoch 27/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2344 - accuracy: 0.9287 - val_loss: 0.4508 - val_accuracy: 0.8732\nEpoch 28/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2274 - accuracy: 0.9308 - val_loss: 0.4438 - val_accuracy: 0.8758\nEpoch 29/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.2211 - accuracy: 0.9327 - val_loss: 0.4484 - val_accuracy: 0.8748\nEpoch 30/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2146 - accuracy: 0.9346 - val_loss: 0.4528 - val_accuracy: 0.8749\nEpoch 31/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2084 - accuracy: 0.9362 - val_loss: 0.4526 - val_accuracy: 0.8759\nEpoch 32/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.2024 - accuracy: 0.9380 - val_loss: 0.4618 - val_accuracy: 0.8754\nEpoch 33/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1965 - accuracy: 0.9402 - val_loss: 0.4560 - val_accuracy: 0.8762\nEpoch 34/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1912 - accuracy: 0.9412 - val_loss: 0.4616 - val_accuracy: 0.8743\nEpoch 35/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.1857 - accuracy: 0.9431 - val_loss: 0.4649 - val_accuracy: 0.8753\nEpoch 36/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.1809 - accuracy: 0.9448 - val_loss: 0.4707 - val_accuracy: 0.8759\nEpoch 37/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1758 - accuracy: 0.9460 - val_loss: 0.4747 - val_accuracy: 0.8756\nEpoch 38/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1715 - accuracy: 0.9474 - val_loss: 0.4812 - val_accuracy: 0.8746\nEpoch 39/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1665 - accuracy: 0.9487 - val_loss: 0.4833 - val_accuracy: 0.8750\nEpoch 40/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.1619 - accuracy: 0.9502 - val_loss: 0.4886 - val_accuracy: 0.8762\nEpoch 41/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.1579 - accuracy: 0.9514 - val_loss: 0.4905 - val_accuracy: 0.8750\nEpoch 42/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1541 - accuracy: 0.9524 - val_loss: 0.4928 - val_accuracy: 0.8750\nEpoch 43/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1495 - accuracy: 0.9541 - val_loss: 0.5019 - val_accuracy: 0.8748\nEpoch 44/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1461 - accuracy: 0.9549 - val_loss: 0.5074 - val_accuracy: 0.8756\nEpoch 45/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1424 - accuracy: 0.9563 - val_loss: 0.5086 - val_accuracy: 0.8747\nEpoch 46/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1391 - accuracy: 0.9574 - val_loss: 0.5083 - val_accuracy: 0.8755\nEpoch 47/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1352 - accuracy: 0.9584 - val_loss: 0.5192 - val_accuracy: 0.8742\nEpoch 48/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1324 - accuracy: 0.9590 - val_loss: 0.5235 - val_accuracy: 0.8749\nEpoch 49/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1288 - accuracy: 0.9602 - val_loss: 0.5282 - val_accuracy: 0.8744\nEpoch 50/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1259 - accuracy: 0.9611 - val_loss: 0.5322 - val_accuracy: 0.8755\nEpoch 51/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1232 - accuracy: 0.9617 - val_loss: 0.5281 - val_accuracy: 0.8753\nEpoch 52/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1203 - accuracy: 0.9626 - val_loss: 0.5397 - val_accuracy: 0.8750\nEpoch 53/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1175 - accuracy: 0.9632 - val_loss: 0.5448 - val_accuracy: 0.8742\nEpoch 54/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.1149 - accuracy: 0.9642 - val_loss: 0.5446 - val_accuracy: 0.8750\nEpoch 55/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1126 - accuracy: 0.9648 - val_loss: 0.5485 - val_accuracy: 0.8742\nEpoch 56/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1098 - accuracy: 0.9658 - val_loss: 0.5527 - val_accuracy: 0.8748\nEpoch 57/100\n","name":"stdout"},{"output_type":"stream","text":"125/125 [==============================] - 2s 14ms/step - loss: 0.1075 - accuracy: 0.9664 - val_loss: 0.5616 - val_accuracy: 0.8744\nEpoch 58/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1056 - accuracy: 0.9668 - val_loss: 0.5607 - val_accuracy: 0.8746\nEpoch 59/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.1029 - accuracy: 0.9677 - val_loss: 0.5694 - val_accuracy: 0.8738\nEpoch 60/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.1012 - accuracy: 0.9679 - val_loss: 0.5738 - val_accuracy: 0.8740\nEpoch 61/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0988 - accuracy: 0.9691 - val_loss: 0.5791 - val_accuracy: 0.8738\nEpoch 62/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0967 - accuracy: 0.9695 - val_loss: 0.5852 - val_accuracy: 0.8733\nEpoch 63/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0944 - accuracy: 0.9700 - val_loss: 0.5812 - val_accuracy: 0.8744\nEpoch 64/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0927 - accuracy: 0.9705 - val_loss: 0.5879 - val_accuracy: 0.8739\nEpoch 65/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0907 - accuracy: 0.9711 - val_loss: 0.5912 - val_accuracy: 0.8737\nEpoch 66/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0893 - accuracy: 0.9714 - val_loss: 0.5984 - val_accuracy: 0.8736\nEpoch 67/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0874 - accuracy: 0.9719 - val_loss: 0.6007 - val_accuracy: 0.8728\nEpoch 68/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.0855 - accuracy: 0.9723 - val_loss: 0.6105 - val_accuracy: 0.8736\nEpoch 69/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0836 - accuracy: 0.9731 - val_loss: 0.6059 - val_accuracy: 0.8740\nEpoch 70/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.0822 - accuracy: 0.9734 - val_loss: 0.6156 - val_accuracy: 0.8730\nEpoch 71/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.0804 - accuracy: 0.9741 - val_loss: 0.6201 - val_accuracy: 0.8730\nEpoch 72/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0791 - accuracy: 0.9742 - val_loss: 0.6253 - val_accuracy: 0.8735\nEpoch 73/100\n125/125 [==============================] - 2s 15ms/step - loss: 0.0774 - accuracy: 0.9749 - val_loss: 0.6275 - val_accuracy: 0.8733\nEpoch 74/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0761 - accuracy: 0.9753 - val_loss: 0.6303 - val_accuracy: 0.8729\nEpoch 75/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0747 - accuracy: 0.9756 - val_loss: 0.6323 - val_accuracy: 0.8738\nEpoch 76/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0731 - accuracy: 0.9762 - val_loss: 0.6386 - val_accuracy: 0.8734\nEpoch 77/100\n125/125 [==============================] - 2s 16ms/step - loss: 0.0720 - accuracy: 0.9764 - val_loss: 0.6466 - val_accuracy: 0.8723\nEpoch 78/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0708 - accuracy: 0.9768 - val_loss: 0.6460 - val_accuracy: 0.8724\nEpoch 79/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0690 - accuracy: 0.9774 - val_loss: 0.6504 - val_accuracy: 0.8732\nEpoch 80/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.0679 - accuracy: 0.9775 - val_loss: 0.6489 - val_accuracy: 0.8735\nEpoch 81/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0667 - accuracy: 0.9779 - val_loss: 0.6575 - val_accuracy: 0.8722\nEpoch 82/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.0655 - accuracy: 0.9782 - val_loss: 0.6570 - val_accuracy: 0.8737\nEpoch 83/100\n125/125 [==============================] - 2s 13ms/step - loss: 0.0642 - accuracy: 0.9786 - val_loss: 0.6674 - val_accuracy: 0.8725\nEpoch 84/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0631 - accuracy: 0.9790 - val_loss: 0.6661 - val_accuracy: 0.8726\nEpoch 85/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0619 - accuracy: 0.9792 - val_loss: 0.6771 - val_accuracy: 0.8717\nEpoch 86/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0609 - accuracy: 0.9797 - val_loss: 0.6719 - val_accuracy: 0.8731\nEpoch 87/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0600 - accuracy: 0.9799 - val_loss: 0.6772 - val_accuracy: 0.8732\nEpoch 88/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0590 - accuracy: 0.9802 - val_loss: 0.6829 - val_accuracy: 0.8729\nEpoch 89/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0580 - accuracy: 0.9806 - val_loss: 0.6863 - val_accuracy: 0.8726\nEpoch 90/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0570 - accuracy: 0.9809 - val_loss: 0.6920 - val_accuracy: 0.8727\nEpoch 91/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0563 - accuracy: 0.9809 - val_loss: 0.6926 - val_accuracy: 0.8734\nEpoch 92/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0552 - accuracy: 0.9814 - val_loss: 0.6963 - val_accuracy: 0.8725\nEpoch 93/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0545 - accuracy: 0.9814 - val_loss: 0.7032 - val_accuracy: 0.8727\nEpoch 94/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0535 - accuracy: 0.9816 - val_loss: 0.7019 - val_accuracy: 0.8728\nEpoch 95/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0528 - accuracy: 0.9820 - val_loss: 0.7092 - val_accuracy: 0.8726\nEpoch 96/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0517 - accuracy: 0.9823 - val_loss: 0.7148 - val_accuracy: 0.8718\nEpoch 97/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0509 - accuracy: 0.9824 - val_loss: 0.7080 - val_accuracy: 0.8728\nEpoch 98/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0506 - accuracy: 0.9825 - val_loss: 0.7185 - val_accuracy: 0.8727\nEpoch 99/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0500 - accuracy: 0.9826 - val_loss: 0.7179 - val_accuracy: 0.8722\nEpoch 100/100\n125/125 [==============================] - 2s 14ms/step - loss: 0.0488 - accuracy: 0.9831 - val_loss: 0.7236 - val_accuracy: 0.8718\n","name":"stdout"},{"output_type":"execute_result","execution_count":28,"data":{"text/plain":"<tensorflow.python.keras.callbacks.History at 0x7fd51824cb10>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.save('eng2french.h5')","execution_count":29,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Define sampling models\nencoder_model = Model(encoder_inputs, encoder_states)\n\ndecoder_state_input_h = Input(shape=(latent_dim,))\ndecoder_state_input_c = Input(shape=(latent_dim,))\ndecoder_states_inputs = [decoder_state_input_h, decoder_state_input_c]\ndecoder_outputs, state_h, state_c = decoder_lstm(\n    decoder_inputs, initial_state=decoder_states_inputs)\ndecoder_states = [state_h, state_c]\ndecoder_outputs = decoder_dense(decoder_outputs)\ndecoder_model = Model(\n    [decoder_inputs] + decoder_states_inputs,\n    [decoder_outputs] + decoder_states)\n\n# Reverse-lookup token index to decode sequences back to\n# something readable.\nreverse_input_char_index = dict(\n    (i, char) for char, i in input_token_index.items())\nreverse_target_char_index = dict(\n    (i, char) for char, i in target_token_index.items())\n","execution_count":32,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def decode_sequence(input_seq):\n    # Encode the input as state vectors.\n    states_value = encoder_model.predict(input_seq)\n\n    # Generate empty target sequence of length 1.\n    target_seq = np.zeros((1, 1, num_decoder_tokens))\n    # Populate the first character of target sequence with the start character.\n    target_seq[0, 0, target_token_index['\\t']] = 1.\n\n    # Sampling loop for a batch of sequences\n    # (to simplify, here we assume a batch of size 1).\n    stop_condition = False\n    decoded_sentence = ''\n    while not stop_condition:\n        output_tokens, h, c = decoder_model.predict(\n            [target_seq] + states_value)\n\n        # Sample a token\n        sampled_token_index = np.argmax(output_tokens[0, -1, :])\n        sampled_char = reverse_target_char_index[sampled_token_index]\n        decoded_sentence += sampled_char\n\n        # Exit condition: either hit max length\n        # or find stop character.\n        if (sampled_char == '\\n' or\n           len(decoded_sentence) > max_decoder_seq_length):\n            stop_condition = True\n\n        # Update the target sequence (of length 1).\n        target_seq = np.zeros((1, 1, num_decoder_tokens))\n        target_seq[0, 0, sampled_token_index] = 1.\n\n        # Update states\n        states_value = [h, c]\n\n    return decoded_sentence","execution_count":38,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for seq_index in range(100):\n    # Take one sequence (part of the training set)\n    # for trying out decoding.\n    input_seq = encoder_input_data[seq_index: seq_index + 1]\n    decoded_sentence = decode_sequence(input_seq)\n    print('-')\n    print('Input sentence:', input_texts[seq_index])\n    print('Decoded sentence:', decoded_sentence)","execution_count":39,"outputs":[{"output_type":"stream","text":"-\nInput sentence: Go.\nDecoded sentence: Va !\n\n-\nInput sentence: Hi.\nDecoded sentence: Salut !\n\n-\nInput sentence: Hi.\nDecoded sentence: Salut !\n\n-\nInput sentence: Run!\nDecoded sentence: Cours !\n\n-\nInput sentence: Run!\nDecoded sentence: Cours !\n\n-\nInput sentence: Who?\nDecoded sentence: Qui ?\n\n-\nInput sentence: Wow!\nDecoded sentence: Ça altera-t-il !\n\n-\nInput sentence: Fire!\nDecoded sentence: Au feu !\n\n-\nInput sentence: Help!\nDecoded sentence: À l'aide !\n\n-\nInput sentence: Jump.\nDecoded sentence: Saute.\n\n-\nInput sentence: Stop!\nDecoded sentence: Stop !\n\n-\nInput sentence: Stop!\nDecoded sentence: Stop !\n\n-\nInput sentence: Stop!\nDecoded sentence: Stop !\n\n-\nInput sentence: Wait!\nDecoded sentence: Attendez !\n\n-\nInput sentence: Wait!\nDecoded sentence: Attendez !\n\n-\nInput sentence: Go on.\nDecoded sentence: Poursuivez.\n\n-\nInput sentence: Go on.\nDecoded sentence: Poursuivez.\n\n-\nInput sentence: Go on.\nDecoded sentence: Poursuivez.\n\n-\nInput sentence: Hello!\nDecoded sentence: Salut !\n\n-\nInput sentence: Hello!\nDecoded sentence: Salut !\n\n-\nInput sentence: I see.\nDecoded sentence: Je comprends.\n\n-\nInput sentence: I try.\nDecoded sentence: J'ai mangé.\n\n-\nInput sentence: I won!\nDecoded sentence: J'ai gagné !\n\n-\nInput sentence: I won!\nDecoded sentence: J'ai gagné !\n\n-\nInput sentence: I won.\nDecoded sentence: J’ai gagné.\n\n-\nInput sentence: Oh no!\nDecoded sentence: Oh non !\n\n-\nInput sentence: Attack!\nDecoded sentence: Attaque !\n\n-\nInput sentence: Attack!\nDecoded sentence: Attaque !\n\n-\nInput sentence: Cheers!\nDecoded sentence: Santé !\n\n-\nInput sentence: Cheers!\nDecoded sentence: Santé !\n\n-\nInput sentence: Cheers!\nDecoded sentence: Santé !\n\n-\nInput sentence: Cheers!\nDecoded sentence: Santé !\n\n-\nInput sentence: Get up.\nDecoded sentence: Lève-toi.\n\n-\nInput sentence: Go now.\nDecoded sentence: Va, maintenant.\n\n-\nInput sentence: Go now.\nDecoded sentence: Va, maintenant.\n\n-\nInput sentence: Go now.\nDecoded sentence: Va, maintenant.\n\n-\nInput sentence: Got it!\nDecoded sentence: Pars !\n\n-\nInput sentence: Got it!\nDecoded sentence: Pars !\n\n-\nInput sentence: Got it?\nDecoded sentence: Pigé ?\n\n-\nInput sentence: Got it?\nDecoded sentence: Pigé ?\n\n-\nInput sentence: Got it?\nDecoded sentence: Pigé ?\n\n-\nInput sentence: Hop in.\nDecoded sentence: Monte.\n\n-\nInput sentence: Hop in.\nDecoded sentence: Monte.\n\n-\nInput sentence: Hug me.\nDecoded sentence: Serrez-moi dans vos bras !\n\n-\nInput sentence: Hug me.\nDecoded sentence: Serrez-moi dans vos bras !\n\n-\nInput sentence: I fell.\nDecoded sentence: Je suis tombé.\n\n-\nInput sentence: I fell.\nDecoded sentence: Je suis tombé.\n\n-\nInput sentence: I know.\nDecoded sentence: Je sais.\n\n-\nInput sentence: I left.\nDecoded sentence: Je suis parti.\n\n-\nInput sentence: I left.\nDecoded sentence: Je suis parti.\n\n-\nInput sentence: I lied.\nDecoded sentence: J'ai menti.\n\n-\nInput sentence: I lost.\nDecoded sentence: J'ai perdu.\n\n-\nInput sentence: I paid.\nDecoded sentence: J’ai payé.\n\n-\nInput sentence: I'm 19.\nDecoded sentence: J'ai de le pourtir.\n\n-\nInput sentence: I'm OK.\nDecoded sentence: Je suis un enfant.\n\n-\nInput sentence: I'm OK.\nDecoded sentence: Je suis un enfant.\n\n-\nInput sentence: Listen.\nDecoded sentence: Écoutez !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: No way!\nDecoded sentence: C'est pas possible !\n\n-\nInput sentence: Really?\nDecoded sentence: Vrai ?\n\n-\nInput sentence: Really?\nDecoded sentence: Vrai ?\n\n-\nInput sentence: Really?\nDecoded sentence: Vrai ?\n\n-\nInput sentence: Thanks.\nDecoded sentence: Merci !\n\n-\nInput sentence: We try.\nDecoded sentence: Nous avons grosse.\n\n-\nInput sentence: We won.\nDecoded sentence: Nous gagnâmes.\n\n-\nInput sentence: We won.\nDecoded sentence: Nous gagnâmes.\n\n-\nInput sentence: We won.\nDecoded sentence: Nous gagnâmes.\n\n-\nInput sentence: We won.\nDecoded sentence: Nous gagnâmes.\n\n-\nInput sentence: Ask Tom.\nDecoded sentence: Demande à Tom.\n\n-\nInput sentence: Awesome!\nDecoded sentence: Fantastique !\n\n-\nInput sentence: Be calm.\nDecoded sentence: Soyez calme !\n\n-\nInput sentence: Be calm.\nDecoded sentence: Soyez calme !\n\n-\nInput sentence: Be calm.\nDecoded sentence: Soyez calme !\n\n-\nInput sentence: Be cool.\nDecoded sentence: Sois détendu !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be fair.\nDecoded sentence: Soyez équitable !\n\n-\nInput sentence: Be kind.\nDecoded sentence: Sois gentil.\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Be nice.\nDecoded sentence: Soyez gentille !\n\n-\nInput sentence: Beat it.\nDecoded sentence: Dégage !\n\n-\nInput sentence: Call me.\nDecoded sentence: Appelez-moi !\n\n-\nInput sentence: Call me.\nDecoded sentence: Appelez-moi !\n\n-\nInput sentence: Call us.\nDecoded sentence: Appelle-nous !\n\n-\nInput sentence: Call us.\nDecoded sentence: Appelle-nous !\n\n-\nInput sentence: Come in.\nDecoded sentence: Entre.\n\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":4}