Skip to main content

GSoC 2018: Week 2

Hello everyone. Here is the brief work done and progress in my project (Rubi) in the second week.

This week was quite frustrating for me. I gave a lot of time but the outcome was little less. Initially my aim was to solve the problem of the order in which rule was matched. But I have to revert back to my initial structure as the new structure was extremely time consuming. Next I thought to keep track of the rules applied to an expression. This could be a great help in debugging too. So I implemented it.

Next I moved on to testing trinomial_products . But here also results were not good. Our rules are of older version (4.10.8) , but we did not have the test-suite of that version. So I tried extremely hard to test those tests with rubi 4.10.8 . For 3 - 4 days , I put a great effort. At last I realised, things will not work this way. So, I then mailed to Albert, the maintainer of original Rubi, asking him for the test-suite of older version. And the good part is that he has provided me the test-suite today. So now I will move to test our rubi with this test-suite.

In between all this, I found some errors in utility function . I corrected them and updated the PR.
Here are the works done by me this week :

  • Made necessary changes in the parsetools and rubi , so that it keeps track of the rules applied to an expression.
  • test_1_2 of algebraic rules completely passes without any failure. (Removed all @XFAIL from test_1_2 )
  • Corrected AppellF1, NegQ, PositiveQ, BinomialMatchQ, TrinomialMatchQ in the utility functions.
For the upcoming week I plan to make test suits of trinomial and miscellaneous algebra


Popular posts from this blog

GSoC 2018: Week 9

Hello everyone. Here is the progress of rubi module in the 9th week of GSoC period.
In the very beginning of the week, the PR for appelf1 was merged. My aim for this week was to load all rules and test at least a small part for each type of rule. Due to the time factor, its not possible to test all. Moreover, rubi needs to be updated to the latest version, so the test suite too will be updated.  
I started with trigo rules. They are huge in number. They were never tested before, so some utility_functions like `TrigReduce`, `KnownTrigIntegrandQ`, `FunctionOfTrig` etc were wrong. They were corrected. Soon I realised a difference between sympy and Mathematica. In mathematica, `1/Cot[x]` is automatically transformed to `Tan[x]` and `1/Tan[x]` too was transformed to `Cot[x]`. But in sympy, these are different. So, `tan(x)` didn't match to `cot(x)` . Francesco suggested to replace all `cot` with `1/tan`  and similarly for `sec` and `csc` (in the code generator too). 
Next, I moved to `i…

GSoC 2018: Week 10

Hello everyone. Here is the brief progress of Rubi module in the 10th week of GSoC period.
Last week, we were able to load all rules. Rubi has now come in a very stable and working condition. One Missing thing in the main PR was missing instructions and documentation. This week I tried my best to write instructions, so that a new user can easily get familiar with it. I have kept it really simple.
Initially, I wrote a script in Mathematica to generate downvalues in separate files. This was done till now manually, but now it's automated. Then I wrote a sympy script to generate all rules and constraints in proper files directly just by running a function. For a more easier understanding of steps, I created `` explaining all steps.
Next thing I realised that it would be good if we create a parsing script for the tests too. Till now I used maple format. But a lot of manual things were required. So I wrote a script which directly parses tests into sympy format. All this…

GSoC 2018: Week 8

Hello everyone. Here is the brief summary of the progress of rubi integration module in the 8th week of GSoC period.

We completed the testing of logarithmic test cases. Some of the tests are failing as they depend on trigonometric rules. The rules and tests have been updated in this PR.

Next, I loaded all trigonometric rules. They are huge in number. So the loading time increased from 1 min 30 sec to around 6 min. For now, I have not loaded them. I also updated the test cases for parsetools and fixed code quality issues. Currently, the travis is failing due to a bug, which has been fixed in branch 1.2. In the upcoming week, 1.2 probably will get merged in master.

Then I moved on to implementing appellf1 in sympy. Progress can be seen in this PR.

We will be updating our utility_functions to the latest version of RUBI.