From 06f303562aed2c9349a7fa0ae7b88d444950a103 Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Tue, 16 Jun 2026 05:32:16 +0800 Subject: [PATCH] =?UTF-8?q?=E5=88=9D=E5=A7=8B=E5=8C=96=E9=A1=B9=E7=9B=AE?= =?UTF-8?q?=EF=BC=8C=E7=94=B1ModelHub=20XC=E7=A4=BE=E5=8C=BA=E6=8F=90?= =?UTF-8?q?=E4=BE=9B=E6=A8=A1=E5=9E=8B?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Model: kartikraut09/ecocloud-grpo-qwen Source: Original Platform --- .gitattributes | 36 +++++++++++++ README.md | 114 +++++++++++++++++++++++++++++++++++++++++ chat_template.jinja | 54 +++++++++++++++++++ config.json | 57 +++++++++++++++++++++ generation_config.json | 13 +++++ model.safetensors | 3 ++ tokenizer.json | 3 ++ tokenizer_config.json | 29 +++++++++++ 8 files changed, 309 insertions(+) create mode 100644 .gitattributes create mode 100644 README.md create mode 100644 chat_template.jinja create mode 100644 config.json create mode 100644 generation_config.json create mode 100644 model.safetensors create mode 100644 tokenizer.json create mode 100644 tokenizer_config.json diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..52373fe --- /dev/null +++ b/.gitattributes @@ -0,0 +1,36 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..80a579a --- /dev/null +++ b/README.md @@ -0,0 +1,114 @@ +--- +license: mit +language: +- en +tags: +- reinforcement-learning +- grpo +- cloud-management +- multi-agent +- sustainability +- openenv +base_model: Qwen/Qwen2.5-0.5B-Instruct +pipeline_tag: text-generation +--- + +# ⚡ CloudEdge GRPO Controller + +**A Qwen2.5-0.5B model fine-tuned with Group Relative Policy Optimization (GRPO) to manage cloud infrastructure crises.** + +Built for the **Meta PyTorch OpenEnv Hackathon Grand Finale**. + +## Model Description + +This model is a reinforcement-learning-trained controller for the **CloudEdge** cloud crisis simulator. It learns to select optimal infrastructure actions (scale_up, scale_down, optimize_energy, migrate_region) by balancing three competing objectives: + +| Objective | Target | Agent | +|-----------|--------|-------| +| Latency | < 150ms | ResourceAgent | +| Cost | < $400/hr | CostAgent | +| Carbon | < 220 units | SustainabilityAgent | + +### Training Method + +- **Algorithm:** GRPO (Group Relative Policy Optimization) via [TRL](https://github.com/huggingface/trl) +- **Base Model:** [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) +- **Training Steps:** 512 +- **Generations per prompt:** 4 +- **Reward Function:** Shaped multi-objective reward with gap closure + worst-metric bonus + +### Shaped Reward Function + +``` +reward = Σ (gap_closure × weight) + worst_metric_bonus +``` + +| Action | Reward (crisis state) | Model Learns | +|--------|----------------------|-------------| +| `optimize_energy` | **+7.5** | "Best action — addresses cost + carbon simultaneously" | +| `scale_down` | **+5.75** | "Good — reduces cost effectively" | +| `migrate_region` | **+3.75** | "Moderate — helps carbon but hurts cost" | +| `scale_up` | **+1.5** | "Worst — increases cost and carbon" | + +### Training Results + +The model converged to selecting `optimize_energy` as the dominant policy when all metrics are above target — which is the mathematically optimal action given the shaped reward function. + +## How to Use + +```python +from transformers import AutoModelForCausalLM, AutoTokenizer + +model = AutoModelForCausalLM.from_pretrained("kartikraut09/ecocloud-grpo-qwen") +tokenizer = AutoTokenizer.from_pretrained("kartikraut09/ecocloud-grpo-qwen") + +prompt = """<|im_start|>system +You are the CloudEdge controller managing a cloud platform in crisis. +Pick the BEST single action for the current state. Respond with ONLY the action name. + +Actions: + scale_up → latency -40, cost +30, carbon +20 + scale_down → latency +25, cost -35, carbon -15 + optimize_energy → latency +10, cost -20, carbon -40 + migrate_region → latency +15, cost +10, carbon -50 + +Targets: latency<150ms, cost<$400, carbon<220<|im_end|> +<|im_start|>user +Cloud state: latency=280ms, cost=$620/hr, carbon=380, load=critical. Best action?<|im_end|> +<|im_start|>assistant +""" + +inputs = tokenizer(prompt, return_tensors="pt") +outputs = model.generate(**inputs, max_new_tokens=16, temperature=0.1) +print(tokenizer.decode(outputs[0], skip_special_tokens=True)) +# Output: optimize_energy +``` + +## Technical Details + +- **Architecture:** Qwen2 (0.5B parameters) +- **Framework:** PyTorch + HuggingFace Transformers + TRL +- **Environment:** OpenEnv-compatible Gymnasium-style simulator +- **Training Hardware:** Google Colab T4 GPU +- **Training Time:** ~15 minutes (512 steps) + +## Project Links + +- **GitHub:** [KartikRaut09/ecocloud-war-room](https://github.com/KartikRaut09/ecocloud-war-room) +- **Hackathon:** Meta PyTorch OpenEnv Hackathon Grand Finale +- **Themes:** Multi-Agent Interactions · Long-Horizon Planning · World Modeling + +## Citation + +```bibtex +@misc{cloudedge2026, + title={CloudEdge: Multi-Agent LLM Simulator for Sustainable Cloud Crisis Management}, + author={Kartik Raut}, + year={2026}, + url={https://github.com/KartikRaut09/ecocloud-war-room} +} +``` + +## License + +MIT License diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..bdf7919 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,54 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0]['role'] == 'system' %} + {{- messages[0]['content'] }} + {%- else %} + {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }} + {%- endif %} + {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within XML tags:\n" }} + {%- for tool in tools %} + {{- "\n" }} + {{- tool | tojson }} + {%- endfor %} + {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }} +{%- else %} + {%- if messages[0]['role'] == 'system' %} + {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }} + {%- else %} + {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %} + {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role }} + {%- if message.content %} + {{- '\n' + message.content }} + {%- endif %} + {%- for tool_call in message.tool_calls %} + {%- if tool_call.function is defined %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {{- tool_call.arguments | tojson }} + {{- '}\n' }} + {%- endfor %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {{- message.content }} + {{- '\n' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|>\n' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant\n' }} +{%- endif %} diff --git a/config.json b/config.json new file mode 100644 index 0000000..c920646 --- /dev/null +++ b/config.json @@ -0,0 +1,57 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "bfloat16", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 896, + "initializer_range": 0.02, + "intermediate_size": 4864, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 21, + "model_type": "qwen2", + "num_attention_heads": 14, + "num_hidden_layers": 24, + "num_key_value_heads": 2, + "pad_token_id": 151643, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "rope_theta": 1000000.0, + "rope_type": "default" + }, + "sliding_window": null, + "tie_word_embeddings": true, + "transformers_version": "5.0.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..d58659f --- /dev/null +++ b/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.1, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.0.0" +} diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..1fa4f75 --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e5ee3baaaefadcbdd372c850438e531a1ecab63808f3ea7f1087487c133cb0df +size 988097824 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..34510ff --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..9fd0fb4 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,29 @@ +{ + "add_prefix_space": false, + "backend": "tokenizers", + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "is_local": true, + "model_max_length": 131072, + "pad_token": "<|endoftext|>", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +}