(GH159) 1.10.2 Performance improvements • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Intercept __builtin__.sum in groupby (GH885) 1.11 v.0.7.1 (February 29, 2012) This • Add dropna method to Panel (GH171) Improvements to existing features • Use moving min/max algorithms from Bottleneck in rolling_min/rolling_max for > 100x speedup. (GH1504, GH50) • Add Cython group use groups dict in Grouper.size (GH860) • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Enable column access via attributes on GroupBy (GH882) • Enable setting existing

(GH159) 1.13.2 Performance improvements • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Intercept __builtin__.sum in groupby (GH885) 1.14 v.0.7.1 (February 29, 2012) This argpartition Indirect partition. sort Full sort. Notes See np.partition for notes on the different algorithms. Examples >>> a = np.array([3, 4, 2, 1]) >>> a.partition(a, 3) >>> a array([2, 1, 3, 4]) >>> argpartition Indirect partition. sort Full sort. Notes See np.partition for notes on the different algorithms. Examples >>> a = np.array([3, 4, 2, 1]) >>> a.partition(a, 3) >>> a array([2, 1, 3, 4]) >>>

(GH159) 1.12.2 Performance improvements • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Intercept __builtin__.sum in groupby (GH885) 1.13 v.0.7.1 (February 29, 2012) This dropna method to Panel (GH171) 30.9.2 Improvements to existing features • Use moving min/max algorithms from Bottleneck in rolling_min/rolling_max for > 100x speedup. (GH1504, GH50) • Add Cython group use groups dict in Grouper.size (GH860) • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Enable column access via attributes on GroupBy (GH882) • Enable setting existing

(GH159) 1.1.2 Performance improvements • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Intercept __builtin__.sum in groupby (GH885) 1.2 v.0.7.1 (February 29, 2012) This

sum, var and std methods may suffer from numerical im- precision due to the underlying windowing algorithms accumulating sums. When values differ with magnitude 1/??.?????(??.??????).??? this results in these cases, you may be better switching to a different library that implements these out-of-core algorithms for you. 2.24.4 Use other libraries pandas is just one library offering a DataFrame API. Because algorithm. See numpy.sort() for more information. ‘mergesort’ and ‘stable’ are the only stable algorithms. order [None] Has no effect but is accepted for compatibility with numpy. Returns Series[np

sum, var and std methods may suffer from numerical im- precision due to the underlying windowing algorithms accumulating sums. When values differ with magnitude 1/??.?????(??.??????).??? this results in these cases, you may be better switching to a different library that implements these out-of-core algorithms for you. 2.24. Scaling to large datasets 955 pandas: powerful Python data analysis toolkit, Release algorithm. See numpy.sort() for more information. ‘mergesort’ and ‘stable’ are the only stable algorithms. order [None] Has no effect but is accepted for compatibility with numpy. Returns Series[np

sum, var and std methods may suffer from numerical im- precision due to the underlying windowing algorithms accumulating sums. When values differ with magnitude 1/??.?????(??.??????).??? this results in these cases, you may be better switching to a different library that implements these out-of-core algorithms for you. 956 Chapter 2. User Guide pandas: powerful Python data analysis toolkit, Release 1.3 algorithm. See numpy.sort() for more information. ‘mergesort’ and ‘stable’ are the only stable algorithms. order [None] Has no effect but is accepted for compatibility with numpy. Returns Series[np

(GH159) 1.2.2 Performance improvements • Use khash for Series.value_counts, add raw function to algorithms.py (GH861) • Intercept __builtin__.sum in groupby (GH885) 1.3 v.0.7.1 (February 29, 2012) This

sum, var and std methods may suffer from numerical im- precision due to the underlying windowing algorithms accumulating sums. When values differ with magnitude 1/??.?????(??.??????).??? this results in these cases, you may be better switching to a different library that implements these out-of-core algorithms for you. 964 Chapter 2. User Guide pandas: powerful Python data analysis toolkit, Release 1.4 algorithm. See numpy.sort() for more information. ‘mergesort’ and ‘stable’ are the only stable algorithms. order [None] Has no effect but is accepted for compatibility with numpy. Returns Series[np

sum, var and std methods may suffer from numerical im- precision due to the underlying windowing algorithms accumulating sums. When values differ with magnitude 1/??.?????(??.??????).??? this results in these cases, you may be better switching to a different library that implements these out-of-core algorithms for you. 964 Chapter 2. User Guide pandas: powerful Python data analysis toolkit, Release 1.4 algorithm. See numpy.sort() for more information. ‘mergesort’ and ‘stable’ are the only stable algorithms. order [None] Has no effect but is accepted for compatibility with numpy. Returns Series[np