81 commits

Author	SHA1	Message	Date
Stefano Pigozzi	a850cde76b	end	2023-02-15 18:57:30 +01:00
Stefano Pigozzi	7645596007	Remove unused _adapt_string_lookup_layer	2023-02-15 17:29:58 +01:00
Stefano Pigozzi	1c839aecb6	Progress!	2023-02-15 03:34:39 +01:00
Stefano Pigozzi	21871f4e06	Fix docstring	2023-02-14 17:51:30 +01:00
Stefano Pigozzi	58554c84e0	Do not stringify processed tokens	2023-02-14 17:51:24 +01:00
Stefano Pigozzi	b777236735	Various upgrades	2023-02-14 02:25:38 +01:00
Stefano Pigozzi	0a0e82d48a	Remove superfluous comment	2023-02-13 23:09:41 +01:00
Stefano Pigozzi	474107ca5e	Fix doc inaccuracy	2023-02-13 18:58:39 +01:00
Stefano Pigozzi	c5b487f350	Remove double newlines from config docstrings	2023-02-13 18:47:47 +01:00
Stefano Pigozzi	292d618e00	Add reminder to myself	2023-02-13 18:47:29 +01:00
Stefano Pigozzi	c82352c7dc	Fix PottsTokenizer failing to train NLTK model	2023-02-13 17:45:28 +01:00
Stefano Pigozzi	597020ac2d	Improve rendering of evaluation steps	2023-02-13 17:14:56 +01:00
Stefano Pigozzi	119483594a	Remove unused imports	2023-02-13 15:59:03 +01:00
Stefano Pigozzi	c992c04264	Manually run garbage collection after each iteration	2023-02-13 15:57:37 +01:00
Stefano Pigozzi	e93a8ddaf7	Fix type annotation of PottsTokenizerWithNegation	2023-02-13 15:55:45 +01:00
Stefano Pigozzi	d4373f8c02	Fix access to ratings instead of tokens	2023-02-13 15:43:46 +01:00
Stefano Pigozzi	bf193bcd93	Remove unused imports	2023-02-13 15:42:45 +01:00
Stefano Pigozzi	76f20059a6	Fix mse rendering	2023-02-13 15:42:35 +01:00
Stefano Pigozzi	62e8c6fa9f	Final improvements	2023-02-13 15:40:22 +01:00
Stefano Pigozzi	4e8aa68db3	CODE IS DONE	2023-02-12 05:11:58 +01:00
Stefano Pigozzi	ae2cf563e6	more things	2023-02-11 05:57:14 +01:00
Stefano Pigozzi	7778c648c1	fix and patch things	2023-02-11 04:32:17 +01:00
Stefano Pigozzi	ade3a6bdc7	Normalize mean value of tensorflow models	2023-02-10 17:33:54 +01:00
Stefano Pigozzi	07be21e809	idk something more & more	2023-02-10 06:21:50 +01:00
Stefano Pigozzi	03c6123d85	idk something more	2023-02-10 06:02:37 +01:00
Stefano Pigozzi	87d9d7299d	remove useless metric	2023-02-10 05:57:57 +01:00
Stefano Pigozzi	34eec56ada	Something was missing	2023-02-10 05:53:48 +01:00
Stefano Pigozzi	c979699ff1	Add polar model	2023-02-10 05:52:13 +01:00
Stefano Pigozzi	4f40aa44b4	Fix some leftover bugs	2023-02-10 05:18:24 +01:00
Stefano Pigozzi	3d9eeecb2a	Improve the tokenizer situation	2023-02-10 05:12:07 +01:00
Stefano Pigozzi	0ce584e856	Fix tensorspec error	2023-02-10 04:21:08 +01:00
Stefano Pigozzi	0a4ce38982	Use the correct sentiment analyzer in main	2023-02-10 04:17:50 +01:00
Stefano Pigozzi	8907272002	Configure file logging	2023-02-10 04:07:34 +01:00
Stefano Pigozzi	2bd97cbad2	Count runs	2023-02-10 03:32:31 +01:00
Stefano Pigozzi	b4573d5eab	Improve logging	2023-02-10 03:30:41 +01:00
Stefano Pigozzi	14dd8045e0	Add docstring for the root module	2023-02-10 03:27:35 +01:00
Stefano Pigozzi	48ac00b548	Document and fix imports in .tokenizer	2023-02-10 03:19:17 +01:00
Stefano Pigozzi	1294335da0	Document and fix imports in .database	2023-02-10 03:18:45 +01:00
Stefano Pigozzi	218c91bcc1	Document and fix imports in .analysis	2023-02-10 03:17:36 +01:00
Stefano Pigozzi	1809db5f00	Refactor code	2023-02-09 18:54:58 +01:00
Stefano Pigozzi	704624507a	Add missing file	2023-02-08 19:48:41 +01:00
Stefano Pigozzi	e3005ab8b0	enough	2023-02-08 19:46:05 +01:00
Stefano Pigozzi	4d6c8f0fee	stuff's working	2023-02-08 10:54:14 +01:00
Stefano Pigozzi	c31743f066	back to i have no idea of what's happening, but at least it works	2023-02-07 10:22:09 +01:00
Stefano Pigozzi	e9a4421acd	Now I understand text vectorization (but this still does not work)	2023-02-06 01:12:30 +01:00
Stefano Pigozzi	3abba24ca2	Made good progress ... How does text vectorization in tensorflow work?	2023-02-05 17:40:22 +01:00
Stefano Pigozzi	dcfc4fbc3b	Getting closer...	2023-02-04 06:14:24 +01:00
Stefano Pigozzi	02f10e6ae4	Use a class as DataTuple	2023-02-04 05:34:56 +01:00
Stefano Pigozzi	4af654a2fa	Make the namedtuple verbose	2023-02-04 05:29:03 +01:00
Stefano Pigozzi	2675f5ead8	Use float instead of str as Category	2023-02-04 05:28:18 +01:00