[Diff since v0.18.1](TARGENE/TMLE.jl@v0.18.1...v0.19.0)

- Weighted TMLE is now the default
- Renaming: TMLEE -> Tmle, OSE -> Ose, NAIVE -> Plugin
- Datasets must be a `DataFrame`

- Multithreaded acceleration for nuisance function estimation
- Two C-TMLE flavours
- A new resampling strategy taking into account the treatment variables

**Merged pull requests:**
- Ol ctmle template (TARGENE#125) (@olivierlabayle)
- Update README.md (TARGENE#133) (@olivierlabayle)

**Closed issues:**
- Paper submission (TARGENE#53)
- Multiple steps TMLE (TARGENE#90)
- C-TMLE (TARGENE#92)
- Documentation needs some love (TARGENE#108)
- GLM for ill-formed matrices (TARGENE#116)
- Bug if `factorialEstimand` is built from a dataset with categorical treatments (TARGENE#124)
- Acceleration (TARGENE#126)
- Greedy C-TMLE (TARGENE#127)
- Documentation (TARGENE#128)
- Cleaning (TARGENE#129)
- Write JOSS paper and submit (TARGENE#130)

[Diff since v0.18.0](TARGENE/TMLE.jl@v0.18.0...v0.18.1)

- Fix independence assumption in propensity score evaluation. Instead of assuming ``p(T1, ...Tp) = p(T1)*...*p(Tp)``, we now have a general factorization rule ``p(T1, ...Tp) =  TT p(Ti|T<i)``.

**Merged pull requests:**
- Dependent ps (TARGENE#123) (@olivierlabayle)

**Closed issues:**
- Link to documentation in repo description (TARGENE#122)

[Diff since v0.17.1](TARGENE/TMLE.jl@v0.17.1...v0.18.0)

- Multiple steps TMLE

- change from AbstractDifferentiation to DifferentiationInterface

**Merged pull requests:**
- Adds possibility to target more than once (TARGENE#119) (@olivierlabayle)
- Changr from AD to DI (TARGENE#121) (@olivierlabayle)

**Closed issues:**
- IPW and AIPW (TARGENE#93)

[Diff since v0.17.0](TARGENE/TMLE.jl@v0.17.0...v0.17.1)

- New [CausalTables](https://github.com/salbalkus/CausalTables.jl) @salbalkus
- Add documentation for YAML/JSON integrations @olivierlabayle

**Merged pull requests:**
- implement function to fit estimators on a CausalTable (TARGENE#118) (@salbalkus)

**Closed issues:**
- Generate all parameters from a parameter family given data (TARGENE#77)
- Find the time to rework docs (TARGENE#95)
- Bring the FailedEstimate from TMLECI here (TARGENE#113)

[Diff since v0.16.1](TARGENE/TMLE.jl@v0.16.1...v0.17.0)

**Merged pull requests:**
- Agnostic composed (TARGENE#110) (@olivierlabayle)
- For 0.17.0 release (TARGENE#112) (@olivierlabayle)

**Closed issues:**
- ComposedEstimands shouldn't take a function (TARGENE#102)

[Diff since v0.16.0](TARGENE/TMLE.jl@v0.16.0...v0.16.1)

**Merged pull requests:**
- Cv check and doc (TARGENE#109) (@olivierlabayle)

[Diff since v0.15.0](TARGENE/TMLE.jl@v0.15.0...v0.16.0)

- Support for OrderedFactor{2}, Multiclass should be deprecated in the future
- Replace TreatmentTransformer (deprecated) with MLJModels.ContinuousEncoder

**Merged pull requests:**
- Ordered factor (TARGENE#107) (@olivierlabayle)

**Closed issues:**
- No issue, just thank you! (TARGENE#105)

[Diff since v0.14.2](TARGENE/TMLE.jl@v0.14.2...v0.15.0)

Add `factorialEstimand` and `factorialEstimands` for all `CM`, `ATE` and `IATE`.

**Merged pull requests:**
- generalize methods to factorialEstimand [BREAKING] (TARGENE#106) (@olivierlabayle)

[Diff since v0.14.1](TARGENE/TMLE.jl@v0.14.1...v0.14.2)

- Fix the "show" method issues arising from the previous version
- add sorting of treatment values for maximizing joint estimand size

[Diff since v0.14.0](TARGENE/TMLE.jl@v0.14.0...v0.14.1)

- Add significance_test function
- Change show methods of Estimates to return test result

**Merged pull requests:**
- Significance test (TARGENE#103) (@olivierlabayle)