-
I think that you can replace most of the logic in `_get_batch_logps` with `torch.nn.CrossEntropyLoss` and then to get the average instead of the sum divide by the number of tokens that are not ignored…
-
I've tried to run the code after compiling as follows - any help would be appreciated.
Windows:
```bash
g++ -O3 class1.cpp modelclasses.cpp -I . -o lf -lgsl -lopenblas
```
Ubuntu:
```bash
g…
-
For see instantiation of nested static classes a qualified `new` operator may be needed (see e.g. https://stackoverflow.com/questions/2863157/how-does-object_new-work-does-java-have-a-new-operator).
…
-
Performance optimization
## Targets
- [x] `forEach` to `for` loop in `_preprocess()` - slight insignificant performance gain
- [x] cache `espree.parse()` results in `_preprocess()` - 30%~40% redu…
-
parser.add_argument('--seq_len', type=int, default=1440, help='input sequence length for global_model')
parser.add_argument('--label_len', type=int, default=10, help='start token length') #720
parse…
-
New Modules: Smart Target + (one more I forget the name)
This should allow for easy targeting (left-click your target, alt-click for Self)
Currently confirmed we can enumerate the targets properly, s…
-
Traceback (most recent call last):
File "main.py", line 949, in
main()
File "main.py", line 652, in main
train_metrics = train_one_epoch(epoch,
File "main.py", line 784, in train_o…
-
For the following parser definition (`syntax`):
```
%lexer
\n+ %restart
A A
B B
X X
%parser
document:
foo:
$$.sem = [$1.sem]
document X foo:
$$.sem = $1.sem
$…
-
First, thanks for putting together this reference material.
I would like to contribute with some additional threaded interpreter optimizations.
1. Token mapping. Merge call into the inner inte…
-
## Main Objectives/Goals
- Empower Consumer Participation: Develop a platform that allows consumers to engage directly in product creation, ensuring their contributions are meaningful and rewarded.
- …