Update README.md
Browse files
README.md
CHANGED
|
@@ -42,6 +42,9 @@ A fast and efficient ~1.5B model optimized for CPU inference. The model was refa
|
|
| 42 |
- Current quantizations were done from the FP16 model, but the model allows for more compression thanks to its ternary architecture.
|
| 43 |
- If you need to do ternary compression, please write to me and I'll perform QAT from your dataset, tailored specifically to your task.
|
| 44 |
- Plus double QAT via ONNX QAT.
|
|
|
|
|
|
|
|
|
|
| 45 |
|
| 46 |
|
| 47 |
# JiRack RoboTech
|
|
|
|
| 42 |
- Current quantizations were done from the FP16 model, but the model allows for more compression thanks to its ternary architecture.
|
| 43 |
- If you need to do ternary compression, please write to me and I'll perform QAT from your dataset, tailored specifically to your task.
|
| 44 |
- Plus double QAT via ONNX QAT.
|
| 45 |
+
- Adapt train process to avoid catastrophic forgetting with NDA
|
| 46 |
+
- Adapt train process to avoid fast plato in training with NDA
|
| 47 |
+
- Convert model to TQ2_0 with support AVX2 and AVX-512 CPU instructions for high performance on CPU
|
| 48 |
|
| 49 |
|
| 50 |
# JiRack RoboTech
|