-
# We have seen what "production" can mean this week...more (huge) and strange than many have ever seen!
So we need to start not only thinking of making our docker-compose AWS smarter but actually o…
-
Hi, I want to fine-tuning XNLI-15 pertained model for NER downstream task, so i added a BiLSTM and CRF architecture.
But a weird thing happened, the loss of my model keep decreasing, but the F1 sc…
-
A blog post talking about the container and Kubernetes credentials of Liberty, such as:
- Container images - where to get them, prebuilt for production, maybe talk about how to use them, signing, etc…
-
### 🚀 The feature
**TL;DR \-** We want to lean into **modular Multi-Threading/Multi-Processing** instead of the current monolithic Multi-Processing, and steer users away from the monolithic Datase…
-
I have an RX Vega 56 and my performance is way too low. I took the measurements by starting lczero and running "go infinite" command until depth 26 (Id 211, 10x128). I got 1200 nps, which is a terribl…
-
## Enhancement
In the performance overview panel, the color tuning method is [introduced](https://docs.pingcap.com/tidb/dev/performance-tuning-methods), the idea is that if the overall database tim…
-
Recently, we have been looking into optimizing tvm models with tuning and quantization. There should be a way to handle the optimized models. I don't think we should try to tune the model with CI/CD p…
-
When I use "basic" strategy tuning to quantize my model, I ran into this issue during one of the phases:
```
...
2024-02-21 23:25:49 [INFO] Tune 73 result is: [Accuracy (int8|fp32): 0.0035|0.0000…
-
### 🐛 Describe the bug
I want to use profile guided optimization to solve the following long standing problems in PT2 for dynamic shapes:
* With automatic dynamic, we must first uselessly compile …
-
I'm trying to fine-tune BGE-M3 based on the README here: https://github.com/FlagOpen/FlagEmbedding/tree/master/examples/finetune
I originally started with the latest transformers version a few week…