Bigartm

Please refer to [BigARTM v0.7.3 Release notes](http://docs.bigartm.org/en/latest/release_notes/v0.7.3.html).
- New CLI for BigARTM (`bigartm` on Linux, `bigartm.exe` on Windows)
- Support for classification in BigARTM CLI
- Support for asynchronous processing of batches
- Improvements in coherence regularizer and coherence score
- New `TopicMass` score for phi matrix
- Support for documents markup (aka `ptdw` matrices)
- New API for importing batches through memory
- Other changes in https://github.com/bigartm/bigartm/compare/v0.7.2...v0.7.3

Questions? Ask [bigartm-usersgooglegroups.com](https://groups.google.com/forum/!forum/bigartm-users)!

Please refer to [BigARTM v0.7.2 Release notes](http://docs.bigartm.org/en/latest/release_notes/v0.7.2.html).
- Enhancements in high-level python API (`ArtmModel` -> `ARTM`)
- Enhancements in low-level python API (`library.py` -> `master_component.py`)
- Enhancements in CLI interface (`cpp_client`)
- New feature: status retrieval from BigARTM
- New feature: float token counts (`token_count` -> `token_weight`)
- Allow custom weights for each batch (`ProcessBatchesArgs.batch_weight`)
- Bug fixes and cleanup in the online documentation
- Other changes in https://github.com/bigartm/bigartm/compare/v0.7.1...v0.7.2

Questions? Ask [bigartm-usersgooglegroups.com](https://groups.google.com/forum/!forum/bigartm-users)!

Please refer to [BigARTM v0.7.1 Release notes](http://docs.bigartm.org/en/latest/release_notes/v0.7.1.html).

**New features**
- BigARTM notebooks in [English](http://nbviewer.ipython.org/github/bigartm/bigartm-book/blob/master/Topic_Modeling_with_BigARTM.ipynb) and in [Russian](http://nbviewer.ipython.org/github/bigartm/bigartm-book/blob/master/Topic_Modeling_with_BigARTM_RU.ipynb) --- new source of information about BigARTM
- ArtmModel --- a brand new Python API, documented in [English](http://nbviewer.ipython.org/github/bigartm/bigartm-book/blob/master/BigARTM_example_EN.ipynb) and in [Russian](http://nbviewer.ipython.org/github/bigartm/bigartm-book/blob/master/BigARTM_example_RU.ipynb)
- Much faster retrieval of Phi and Theta matrices from Python
- Much faster dictionary imports from Python
- Auto-detect and use all CPU cores by default
- Fixed Import/Export of topic models (was broken in v0.7.0)
- New capability to implement Phi-regularizers in Python code
- Improvements in Coherence score
- Other changes in https://github.com/bigartm/bigartm/compare/v0.7.0...v0.7.1

**New examples**
- [example20_attach_model.py](https://raw.githubusercontent.com/bigartm/bigartm/master/src/python/examples/example20_attach_model.py)

Questions? Ask [bigartm-usersgooglegroups.com](https://groups.google.com/forum/!forum/bigartm-users)!

Please refer to [BigARTM v0.7.0 Release notes](http://docs.bigartm.org/en/latest/release_notes/v0.7.0.html).

**New features**
- New-style models (ProcessBatchese / MergeModel / RegularizeModel / NormalizeModel APIs)
- Network modus operandi had been removed. Let us know if need it back!
- Coherence regularizer and scores (experimental)
- Other changes in https://github.com/bigartm/bigartm/compare/v0.6.4...v0.7.0

**New examples**
- [example17_process_batches.py](https://raw.githubusercontent.com/bigartm/bigartm/master/src/python/examples/example17_process_batches.py)
- [example18_merge_model.py](https://raw.githubusercontent.com/bigartm/bigartm/master/src/python/examples/example18_merge_model.py)
- [example19_regularize_model.py](https://raw.githubusercontent.com/bigartm/bigartm/master/src/python/examples/example19_regularize_model.py)

Questions? Ask [bigartm-usersgooglegroups.com](https://groups.google.com/forum/!forum/bigartm-users)!

Changes in https://github.com/bigartm/bigartm/compare/v0.6.3...v0.6.4
1. New option to export and import topic model in binary format
Example is available in [example15_import_export_topic_model](https://github.com/bigartm/bigartm/blob/master/src/python/examples/example15_import_export_topic_model.py)
2. New option of relative regularization for Phi matrix
3. Remove MasterProxy feature (please let us know if you need it back!)

Changes in https://github.com/bigartm/bigartm/compare/v0.6.2...v0.6.3
1. New option to initialize topic model based on folders with batches.
Example is available in [example14_initialize_topic_model](https://github.com/bigartm/bigartm/blob/master/src/python/examples/example14_initialize_topic_model.py)
2. New options to retrieve and restore topic model into BigARTM library.
Example is available in [example13_overwrite_topic_model](https://github.com/bigartm/bigartm/blob/master/src/python/examples/example13_overwrite_topic_model.py)
3. New regularizer LabelRegularizationPhi is added
4. Better error and info logging, and better validation of input data

Check documentation for new fields:
- ModelConfig.use_new_tokens
- TopicModel.operation_type
- GetTopicModelArgs.request_type
- ModelConfig.score_name (deprecated option)