FlowSteer/requirements.txt at main · beita6969/FlowSteer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
# FlowSteer Training Dependencies
# Tested on: NVIDIA A100 80GB, CUDA 12.x

# ========== PyTorch (CUDA 12.x) ==========
# Install: pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126
torch>=2.0.0
torchvision>=0.15.0
torchaudio>=2.0.0

# ========== Transformers & LoRA ==========
transformers>=4.40.0
accelerate>=0.27.0
peft>=0.10.0
bitsandbytes>=0.42.0
safetensors>=0.4.0

# ========== OpenAI API (LLM Judge) ==========
openai>=1.0.0
httpx>=0.25.0
socksio>=1.0.0

# ========== Data Processing ==========
datasets>=2.16.0
sentencepiece>=0.1.99
tiktoken>=0.5.0
pandas>=2.0.0
scikit-learn>=1.3.0
nltk>=3.8.0

# ========== Utilities ==========
pyyaml>=6.0
tqdm>=4.66.0
scipy>=1.11.0
numpy>=1.24.0
sympy>=1.12
antlr4-python3-runtime>=4.11,<4.12
regex>=2023.0.0
tenacity>=8.2.0
matplotlib>=3.8.0
seaborn>=0.13.0

# ========== Monitoring (Optional) ==========
wandb>=0.16.0

# ========== HuggingFace Hub ==========
huggingface-hub>=0.20.0