File size: 5,478 Bytes
ce60168 | 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 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 | import os
import sys
from huggingface_hub import HfApi, create_repo
def load_naman_token():
env_file = "/home/adminuser/.env"
token = None
if os.path.exists(env_file):
with open(env_file, "r", encoding="utf-8") as f:
for line in f:
line = line.strip()
if line.startswith("NAMAN_HF_TOKEN="):
_, val = line.split("=", 1)
val = val.strip().strip("'").strip('"')
if val:
token = val
break
return token
def create_readme_card(repo_id):
return f"""---
license: apache-2.0
tags:
- mamba
- codestral-mamba
- qwen2.5
- state-space-model
- benchmark
- evaluation
---
# MAMBA_7B: Codestral Mamba 7B Installation, Benchmarks & Architectural Research
This repository contains the complete codebase, benchmark suite, evaluation datasets, research documents, and visual PowerPoint presentations for **MAMBA_7B**.
Published to **namanadep** Hugging Face profile using `NAMAN_HF_TOKEN`.
---
## 📊 Summary Benchmark Metrics (Codestral Mamba 7B vs. Qwen 2.5 7B)
| Metric | Codestral Mamba 7B | Qwen 2.5 7B Instruct | Takeaway |
| :--- | :---: | :---: | :--- |
| **Architecture** | **Selective State Space Model (SSM S6)** | **Multi-Head Self-Attention Transformer** | Mamba eliminates $O(N^2)$ quadratic KV-cache memory scaling. |
| **Average Latency** | **4.28s** | **7.40s** | **42.2% faster completion** for Codestral Mamba. |
| **Generation Speed** | **194.8 t/s** | **193.3 t/s** | Identical throughput on NVIDIA H200 GPUs. |
| **Memory Footprint** | **Constant $O(1)$ Memory State** | $O(N)$ Growth | Fixed VRAM up to 256k long-context reasoning. |
---
## 📂 Repository Layout & Uploaded Artifacts
- `docs/CODESTRAL_MAMBA_7B_VS_QWEN_7B_COMPARISON.md`: 71 KB Exhaustive 10-Prompt Benchmark Report.
- `docs/CODESTRAL_MAMBA_7B_VS_QWEN_7B_COMPARISON.pptx`: 9-Slide Visual Benchmark Comparison Deck.
- `docs/MAMBA_7B_INSTALLATION_AND_ARCHITECTURE_GUIDE.pptx`: 8-Slide Hands-on Installation & Architecture Journey Deck.
- `docs/MAMBA_MODELS_RESEARCH_OLLAMA_HUGGINGFACE.md`: State Space Models Architectural Research Document.
- `data/mamba_vs_qwen_results.json`: Raw Evaluation JSON transcripts across 10 technical categories.
- `src/`: Complete Python benchmark test harness and slide generation scripts.
"""
def main():
base_dir = "/home/adminuser/aiops_pocs/MAMBA_7B"
print("Loading NAMAN_HF_TOKEN from /home/adminuser/.env...")
token = load_naman_token()
if not token:
print("Error: Could not find NAMAN_HF_TOKEN in /home/adminuser/.env")
sys.exit(1)
api = HfApi(token=token)
user_info = api.whoami()
username = user_info.get("name", "namanadep")
repo_id = f"{username}/MAMBA_7B"
print(f"Authenticated as '{username}'. Creating repository: https://huggingface.co/{repo_id}...")
try:
create_repo(repo_id, token=token, exist_ok=True, repo_type="model")
print(f"Repository ready: https://huggingface.co/{repo_id}")
except Exception as e:
print(f"Repo creation status: {e}")
# Write README.md locally
readme_path = os.path.join(base_dir, "README.md")
with open(readme_path, "w", encoding="utf-8") as f:
f.write(create_readme_card(repo_id))
print(f"Uploading files from {base_dir} to Hugging Face repository https://huggingface.co/{repo_id}...")
# 1. Upload README.md
api.upload_file(
path_or_fileobj=readme_path,
path_in_repo="README.md",
repo_id=repo_id,
token=token
)
print(" -> Uploaded README.md")
# 2. Upload Docs Folder
docs_dir = os.path.join(base_dir, "docs")
if os.path.exists(docs_dir):
for fname in os.listdir(docs_dir):
fpath = os.path.join(docs_dir, fname)
if os.path.isfile(fpath):
api.upload_file(
path_or_fileobj=fpath,
path_in_repo=f"docs/{fname}",
repo_id=repo_id,
token=token
)
print(f" -> Uploaded docs/{fname}")
# 3. Upload Src Folder
src_dir = os.path.join(base_dir, "src")
if os.path.exists(src_dir):
for fname in os.listdir(src_dir):
fpath = os.path.join(src_dir, fname)
if os.path.isfile(fpath):
api.upload_file(
path_or_fileobj=fpath,
path_in_repo=f"src/{fname}",
repo_id=repo_id,
token=token
)
print(f" -> Uploaded src/{fname}")
# 4. Upload Data Folder
data_dir = os.path.join(base_dir, "data")
if os.path.exists(data_dir):
for fname in os.listdir(data_dir):
fpath = os.path.join(data_dir, fname)
if os.path.isfile(fpath):
api.upload_file(
path_or_fileobj=fpath,
path_in_repo=f"data/{fname}",
repo_id=repo_id,
token=token
)
print(f" -> Uploaded data/{fname}")
print("==========================================================================")
print(f" Successfully published MAMBA_7B to Hugging Face: https://huggingface.co/{repo_id} ")
print("==========================================================================")
if __name__ == "__main__":
main()
|