Upload generation and checking scripts
Browse files
README.md
CHANGED
|
@@ -23,14 +23,14 @@ Each row has the form
|
|
| 23 |
123 = (x1)^3 + (x2)^3 + (x3)^3 + (x4)^3
|
| 24 |
```
|
| 25 |
|
| 26 |
-
The files are sharded by intervals of length `10,000,000`
|
|
|
|
| 27 |
|
| 28 |
```text
|
| 29 |
cubes-00.txt 0 <= n <= 9,999,999
|
| 30 |
cubes-01.txt 10,000,000 <= n <= 19,999,999
|
| 31 |
...
|
| 32 |
-
cubes-99.txt 990,000,000 <= n <=
|
| 33 |
-
cubes-100.txt n = 1,000,000,000
|
| 34 |
```
|
| 35 |
|
| 36 |
The generation code uses a congruence-filtered generalized Pell search. For
|
|
@@ -56,6 +56,7 @@ sage -python scripts/generate_hf_dataset.py \
|
|
| 56 |
--max-in-flight 30 \
|
| 57 |
--out-dir generated/sum4cubes \
|
| 58 |
--overwrite \
|
|
|
|
| 59 |
--progress-every 10000000
|
| 60 |
```
|
| 61 |
|
|
|
|
| 23 |
123 = (x1)^3 + (x2)^3 + (x3)^3 + (x4)^3
|
| 24 |
```
|
| 25 |
|
| 26 |
+
The files are sharded by intervals of length `10,000,000`, with the endpoint
|
| 27 |
+
`1,000,000,000` appended to the last shard:
|
| 28 |
|
| 29 |
```text
|
| 30 |
cubes-00.txt 0 <= n <= 9,999,999
|
| 31 |
cubes-01.txt 10,000,000 <= n <= 19,999,999
|
| 32 |
...
|
| 33 |
+
cubes-99.txt 990,000,000 <= n <= 1,000,000,000
|
|
|
|
| 34 |
```
|
| 35 |
|
| 36 |
The generation code uses a congruence-filtered generalized Pell search. For
|
|
|
|
| 56 |
--max-in-flight 30 \
|
| 57 |
--out-dir generated/sum4cubes \
|
| 58 |
--overwrite \
|
| 59 |
+
--merge-final-singleton \
|
| 60 |
--progress-every 10000000
|
| 61 |
```
|
| 62 |
|