report

README.md

@@ -4,6 +4,7 @@
 Model takes text message and classifies it's mood. Part of virtual AI companion creation.
 - Link to the dataset - https://www.kaggle.com/shrivastava/isears-dataset
 - Semestral work for NI-MVI course (FIT CTU)
+The source code available for testing and interaction is at https://colab.research.google.com/drive/1Es9stqxrdxpLuolWxPcPMpuQz-iRg3rd#scrollTo=uI2QgGstEGzW
 
 ## Tech stack
 - Python 3, Keras, Tensorflow, pandas, numpy
@@ -15,15 +16,13 @@ Model takes text message and classifies it's mood. Part of virtual AI companion
 - Completed BERT contextual embeddings layer
 - Builded BERT-biLSTM classification model
 - Started training
-
-## To do
-- Finish training
-- Evaluate results
-- Save model
+- Finished training
+- Evaluated results
+- Saved model
 
 ## Papers
 My research journey started with:
 - https://iopscience.iop.org/article/10.1088/1742-6596/971/1/012049/pdf
 - https://towardsdatascience.com/sentiment-analysis-comparing-3-common-approaches-naive-bayes-lstm-and-vader-ab561f834f89
 - https://www.sciencedirect.com/science/article/pii/S1532046420301672
-- https://towardsdatascience.com/from-pre-trained-word-embeddings-to-pre-trained-language-models-focus-on-bert-343815627598
\ No newline at end of file
+- https://towardsdatascience.com/from-pre-trained-word-embeddings-to-pre-trained-language-models-focus-on-bert-343815627598

metrics.json

+{"loss": [1.9433, 1.864, 1.7994, 1.7219, 1.6296, 1.5297, 1.4467, 1.3725, 1.3203, 1.2754, 1.2361, 1.2103, 1.1867, 1.1541, 1.1259, 1.1104, 1.0994, 1.0662, 1.0575, 1.0339, 1.0141, 1.0059, 0.9908, 0.9668, 0.9571, 0.9514, 0.9327, 0.9175, 0.9012, 0.9, 0.8714, 0.8575, 0.8551, 0.8383, 0.8179, 0.8066], "accuracy": [0.1786, 0.2584, 0.3292, 0.3746, 0.4232, 0.4562, 0.4782, 0.4974, 0.5143, 0.5337, 0.5402, 0.5538, 0.5649, 0.5778, 0.5847, 0.5908, 0.5964, 0.6087, 0.61, 0.6215, 0.6394, 0.6394, 0.6411, 0.6511, 0.6597, 0.6524, 0.6699, 0.6724, 0.6727, 0.6775, 0.6883, 0.7004, 0.695, 0.7023, 0.7052, 0.7186], "val_loss": [1.877, 1.8029, 1.7153, 1.6201, 1.5145, 1.4201, 1.3478, 1.289, 1.2532, 1.241, 1.2177, 1.1979, 1.1759, 1.1719, 1.1691, 1.1708, 1.163, 1.1652, 1.1461, 1.1781, 1.1526, 1.1559, 1.1505, 1.148, 1.151, 1.1491, 1.1516, 1.1607, 1.1549, 1.1796, 1.1751, 1.1847, 1.1974, 1.2018, 1.1906, 1.1949], "val_accuracy": [0.2439, 0.3584, 0.4077, 0.4503, 0.4787, 0.4937, 0.5121, 0.523, 0.5297, 0.5414, 0.5505, 0.5597, 0.5664, 0.5714, 0.5664, 0.5664, 0.5773, 0.5739, 0.5815, 0.5773, 0.5781, 0.5873, 0.5873, 0.5965, 0.5906, 0.5998, 0.5982, 0.6007, 0.6057, 0.599, 0.609, 0.5931, 0.5982, 0.6023, 0.6032, 0.5998]}
\ No newline at end of file