e2323d579c
Dataset on Weibo topic's popularity over time.
戒酒的李白
2025-08-11 21:40:28 +08:00
de68be59f5
Updated the topic prediction script to support Top-K predictions and added an interactive selection feature for optional Chinese foundation models.
戒酒的李白
2025-08-09 23:16:05 +08:00
d726941d95
Fix label mapping and boost training with batch_size=64.
戒酒的李白
2025-08-09 19:42:09 +08:00
4373f31dcf
Train and prediction script for a topic classification model based on bert-chinese.
戒酒的李白
2025-08-09 19:34:44 +08:00
20222bbdda
Add the topic identification dataset.
戒酒的李白
2025-08-09 09:59:42 +08:00
d88d5edd99
The LLM-based topic recognition model is complete and adapted to quickly updating Weibo topics.
戒酒的李白
2025-08-07 11:14:38 +08:00
ad27791e1c
Optimize the model loading logic of Qwen3.
戒酒的李白
2025-08-05 11:40:12 +08:00
c4af1eff37
Add the base class, configuration file, and training script for the Qwen3 sentiment analysis model, with support for various model sizes (0.6B, 4B, 8B).
戒酒的李白
2025-08-05 01:44:22 +08:00
43525c5ca8
Added a base model class and training scripts for various sentiment analysis models, including Naive Bayes, SVM, XGBoost, LSTM, and BERT. Also, improved prediction functionality and the model loading mechanism.
戒酒的李白
2025-08-04 22:07:30 +08:00
bd60e2ed1b
A new multilingual sentiment analysis module has been added.
戒酒的李白
2025-08-04 19:49:59 +08:00
645242a552
Updated how the fine-tuned BERT model is stored.
戒酒的李白
2025-08-04 14:06:45 +08:00
b504808745
Update the.gitAttributes file, add processing rules for text and binary files, and ensure the correct identification of file types.
戒酒的李白
2025-08-04 00:20:29 +08:00
9a5168e7b2
Add contributor code of conduct, license files and ICONS, and update.gitAttributes to mark.js files as Python language.
戒酒的李白
2025-08-04 00:00:58 +08:00
14d8ac6e71
Completed fine-tuning for several models and prepared the foundational components for an upcoming refactor.
戒酒的李白
2025-08-03 23:08:30 +08:00
ed23ab1908
A new microblog sentiment recognition model has been added, based on the fine-tuning of GPT2, but it has not yet been adapted to the system.
戒酒的李白
2025-05-18 22:07:18 +08:00
bad84f5476
Enhance LSTM model for small datasets and improve performance.
戒酒的李白
2025-03-29 11:43:27 +08:00
a614bca835
Enhanced public opinion prediction system by integrating LSTM model.
戒酒的李白
2025-03-27 16:25:22 +08:00
c9fe3312a8
Add spider and workflow visualization module entries and related route configurations.
戒酒的李白
2025-03-21 19:38:58 +08:00
d4d22f726e
The newly added crawler visualization settings and the process visualization orchestration module have been adapted to support bilingual switching.
戒酒的李白
2025-03-18 10:18:52 +08:00
54412b7f2a
Process visualization orchestration BUG fix, feature enhancement, and support for export and import of orchestration processes.
戒酒的李白
2025-03-17 18:50:56 +08:00
231d533ece
Optimize the crawler configuration page, add multi-account parallel functionality, adapt AI configuration features, and include database configuration options.
戒酒的李白
2025-03-15 13:19:41 +08:00
5630b30002
Comprehensive enhancement of user authentication system security, including password storage, session management, XSS/CSRF protection, etc.
戒酒的李白
2025-03-07 19:32:19 +08:00
戒酒的李白