hamletbatista

## atis_test_predictions.py
test_df = pd.read_csv("atis.test.csv", index_col=0)

model = LudwigModel.load("results/experiment_run_6/model")

predictions = model.predict(test_df)

test_df.reset_index().join(predictions)[["tokens", "intent_predictions", "slots_predictions"]]


## atis_model_definition.yaml
input_features:
    -
        name: tokens
        type: text
        level: word
        encoder: rnn
        cell_type: lstm
        bidirectional: true
        num_layers: 2
        reduce_output: null

## load_atis_dataset.py
#From https://www.kaggle.com/siddhadev/atis-dataset-clean/home
#unzip atis-dataset-clean.zip

df = pd.read_csv("atis.train.csv", index_col=0)

df.groupby("intent").count()

## spacy_hotel_word_vector.py
#hotel word vector representation
print(tokens[0]) #output -> hotel
print(tokens[0].vector)

#output
array([ 4.2714e-01,  1.2711e-01,  3.3711e-01, -1.0423e+00,  6.4144e-01,
       -1.2071e-01,  5.0131e-02, -4.7584e-01, -1.0312e-01,  2.9255e+00,
       -5.8213e-01, -3.2514e-01, -3.5034e-01,  2.1803e-01,  1.0048e-03,
       -8.2658e-01, -1.0869e-01,  1.4241e+00,  6.1642e-01, -4.4746e-02,
       -4.5506e-01,  4.3144e-01, -5.7901e-02, -1.7059e-02,  6.4482e-02,

## spacy_word_similarity.py
#python -m spacy download en_core_web_lg

import spacy

nlp = spacy.load('en_core_web_lg')

tokens = nlp(u'hotel resort car bike')

#comparing 3 words
for token1 in tokens:

## ludwig_get_questions_predictions.py
model = LudwigModel.load("results/experiment_run_3/model")

test_df = pd.read_csv("Question Report_Page 1_Table.csv")

#we rename Query to Questions to match what the model expects
predictions = model.predict(test_df.rename(columns={'Query': 'Questions'} ))

test_df.join(predictions)[["Query", "Category0_predictions"]]

## ludwig_parallel_cnn_questions.yaml
input_features:
    -
        name: Questions
        type: text
        level: word
        encoder: parallel_cnn

output_features:
    -
        name: Category0

## load_questions_dataset_colab.py
#https://colab.research.google.com/notebooks/io.ipynb
from google.colab import files

uploaded = files.upload()

for fn in uploaded.keys():
  print('User uploaded file "{name}" with length {length} bytes'.format(
      name=fn, length=len(uploaded[fn])))

df = pd.read_csv("Question_Classification_Dataset.csv", index_col=0)

## ludwig_get_predictions.py
from ludwig.api import LudwigModel

model = LudwigModel.load("results/experiment_run_0/model")

predictions = model.predict(test_df)

test_df.join(predictions)[["text", "category_predictions"]]


## ludwig_google_trends_dataset.py
gtrends_topics=["DAZN's Jamie Rice on UCL Final, Joshua's First US Fight and Original Global Content", "What date is London Pride 2019, what events are ha…ning in the capital and what's this year's route?", "2019 Memorial Tournament leaderboard: Live coverage, Tiger Woods score, golf scores on Saturday", "Live | AFGvAUS: Warner, Finch start 208-run chase against Afghanistan", "Exclusive: Martin Jol reveals Daniel Levy's amazing reaction beside him in stand to Ajax victory", "Ties with China beneficial, its envoy tells new El Salvador leader", "Knockout! Watch Frank Camacho finish Nick Hein on the feet at UFC Stockholm", "Police Ramp Up Security For Rock N' Roll Marathon", "Here's what Princess Diana Hated about Prince Charles", "Aleksandar Rakic vs Jimi Manuwa live streaming: Watch UFC online on ESPN+", "Lyndhurst, New Jersey Communities Rally To Support Kick-Off Of Pride Month", "Van Dijk and Wanyama to do battle in Champions League final", "The Latest: Djokovic into 4th round without dropping a set", "Totte