KUSAL BERA kusal1990
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # https://www.kaggle.com/suicaokhoailang/lstm-attention-baseline-0-652-lb | |
| class Attention(Layer): | |
| def __init__(self, step_dim, | |
| W_regularizer=None, b_regularizer=None, | |
| W_constraint=None, b_constraint=None, | |
| bias=True, **kwargs): | |
| self.supports_masking = True | |
| self.init = initializers.get('glorot_uniform') |
kusal1990
/ Matthews_Correlation_Coefficient.py
Created
June 2, 2022 18:35
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Matthews correlation coefficient calculation used inside Keras model | |
| def matthews_correlation(y_true, y_pred): | |
| """ | |
| Calculate Matthews Correlation Coefficient. | |
| References | |
| ---------- | |
| .. [1] https://en.wikipedia.org/wiki/Matthews_correlation_coefficient | |
| .. [2] https://www.kaggle.com/tarunpaparaju/vsb-competition-attention-bilstm-with-features/notebook?scriptVersionId=10690570 | |
| """ |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def model_lstm(input_shape): | |
| # The shape was explained above, must have this order | |
| inp = Input(shape=(input_shape[1], input_shape[2],)) | |
| # This is the LSTM layer | |
| # Bidirecional implies that the 160 chunks are calculated in both ways, 0 to 159 and 159 to zero | |
| # although it appear that just 0 to 159 way matter, I have tested with and without, and tha later worked best | |
| # 128 and 64 are the number of cells used, too many can overfit and too few can underfit | |
| x = Bidirectional(CuDNNLSTM(128, return_sequences=True))(inp) | |
| # x = Activation('relu')(x) | |
| # x = Dropout(0.25)(x) |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def threshold_search(y_true, y_proba): | |
| best_threshold = 0 | |
| best_score = 0 | |
| for threshold in tqdm([i * 0.01 for i in range(100)]): | |
| score = K.eval(matthews_correlation(y_true.astype(np.float64), (y_proba > threshold).astype(np.float64))) | |
| if score > best_score: | |
| best_threshold = threshold | |
| best_score = score | |
| search_result = {'threshold': best_threshold, 'matthews_correlation': best_score} | |
| return search_result |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| preds_test = [] | |
| for i in range(N_SPLITS): | |
| model.load_weights('weights_{}.h5'.format(i)) | |
| pred = model.predict(X_test_input, batch_size=300, verbose=1) | |
| pred_3 = [] | |
| for pred_scalar in pred: | |
| for i in range(3): | |
| pred_3.append(pred_scalar) | |
| preds_test.append(pred_3) | |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import flask | |
| app = Flask(__name__) | |
| @app.route('/') | |
| def hello_world(): | |
| return 'Hello World!' | |
| @app.route('/index8050') | |
| def index8050(): |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # to open/create a new html file in the write mode | |
| f = open('index8050.html', 'w') | |
| # the html code which will go in the file GFG.html | |
| html_template = """<html> | |
| <head> | |
| <title></title> | |
| </head> | |
| <body> | |
| <h2>VSB Power Line Fault Detection:Classification</h2> |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| plt.figure(figsize=(16,8)) | |
| plt.title("Number of words in a question and it's frequency for challenge set") | |
| sns.barplot([i[0] for i in sorted_x],[i[1] for i in sorted_x]) | |
| plt.xlabel("Number of words") | |
| plt.ylabel("Frequency") | |
| plt.show() |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def Answer_graph(type_): | |
| print(type_+" dataset Analysis ...............") | |
| plt.figure(figsize=(16,4.5)) | |
| for i,env in enumerate(["Train","Test","Dev"],start = 1): | |
| Answers = eval(env+"_multiple_main_"+type_+"[\'AnswerKey\'].value_counts().to_dict()") | |
| plt.subplot(1,3,i) | |
| plt.title(env+"_Answers Options vs Frequency("+type_+")") | |
| sns.barplot([i for i in Answers.keys()],[i for i in Answers.values()]) | |
| plt.xlabel("Answer Options") | |
| plt.ylabel("Frequency") |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Test SchoolGrade | |
| def plot_subplot_bar(data,type_): | |
| print(type_+" dataset Analysis ...............") | |
| plt.figure(figsize=(16,4.5)) | |
| for i,v in enumerate(["Train","Test","Dev"],start = 1): | |
| plt.subplot(1,3,i) | |
| plt.title(v+"_schoolGrade vs No_of_times_used("+type_+")") | |
| sns.distplot(eval(v+data)) | |
| plt.xticks([i for i in range(11)]) | |
| plt.ylabel("No_of_times_used") |