pull request to update tutorial

[dbullock1086]

I made a lot of these changes while trying to optimize an update to the tutorial. Specifically, I added a Z->ll topological model for a classification study. Major updates include:

• implementing a choice of weight normalization for trimming and range
• adding 'arrType' to Config to avoid the np.float32 vs float conflict
• output CSV standardized so it can be used in pandas, PyTables, h5py, R, Julia, Matlab, etc.

Changes also include general cleanup and simplification.

[afarbin]

Hi Daniel, Wow that’s great. Thanks. Sol has been trying to exactly reproduce D. Whiteson et al paper on their data. She also cleaned up the tutorial. Meanwhile I finally got Jupyter notebooks to work the way I want for my class… It would be great if all of these converge… I’m traveling this week, but we should pick up meetings next week. I’ll let everyone know. Amir

…

On Jan 10, 2017, at 10:13 PM, dbullock1086 ***@***.***> wrote: I made a lot of these changes while trying to optimize an update to the tutorial. Specifically, I added a Z->ll topological model for a classification study. Major updates include: implementing a choice of weight normalization for trimming and range adding 'arrType' to Config to avoid the np.float32 vs float conflict output CSV standardized so it can be used in pandas, PyTables, h5py, R, Julia, Matlab, etc. Changes also include general cleanup and simplification. You can view, comment on, or merge this pull request online at: #2 <#2> Commit Summary header changed to comma-separated for h5py access using lxplus samples capitalization fix capitalization fix capitalization fix capitalization fix revert np.float32 to float sample directories test CSV output refork refork redundancy cleanup regroup trying different grouping missing comma trying new style simple pedagogy models name fix missing comma missing slash missing comma missing variable reorder for color new configs new tutorial test tested lxplus, checking uta tested lxplus, checking uta actually quick optimizing testing new normalization IQR weight scale/shift capitalization set for tutorial again retest retest File Changes M Analysis.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-0> (87) M Arguments.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-1> (68) M CSVWriter.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-2> (33) M Classification.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-3> (2) M Experiment.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-4> (190) D InputFiles.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-5> (37) D InputVars.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-6> (23) A SUSYConfig.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-7> (204) D ScanConfig.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-8> (62) A ZllConfig.py <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2/files#diff-9> (76) Patch Links: https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2.patch <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2.patch> https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2.diff <https://github.com/UTA-HEP-Computing/EventClassificationDNN/pull/2.diff> — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AOaeCq-b3AvGyl0GqL3LGkKKijZUK3jOks5rRFcAgaJpZM4LgLXC>.