初始化项目，由ModelHub XC社区提供模型

Model: hard007ik/shopmanager-grpo-smoke-l4-v2 Source: Original Platform
2026-05-02 12:37:45 +08:00
commit b8ef3bde9b
19 changed files with 152349 additions and 0 deletions
--- a/training_summary.json
+++ b/training_summary.json
@@ -0,0 +1,63 @@
+{
+  "loss": {
+    "final": 0.0,
+    "max": -0.0,
+    "min": -0.0,
+    "mean": 0.0,
+    "n": 8
+  },
+  "reward_total": {
+    "final": 0.8438500165939331,
+    "max": 0.9020000100135803,
+    "min": 0.800000011920929,
+    "mean": 0.8473062515258789,
+    "n": 8
+  },
+  "reward_market": {
+    "final": 0.0,
+    "max": 0.0,
+    "min": 0.0,
+    "mean": 0.0,
+    "n": 0
+  },
+  "reward_warehouse": {
+    "final": 0.0,
+    "max": 0.0,
+    "min": 0.0,
+    "mean": 0.0,
+    "n": 0
+  },
+  "reward_showroom": {
+    "final": 0.0,
+    "max": 0.0,
+    "min": 0.0,
+    "mean": 0.0,
+    "n": 0
+  },
+  "n_log_rows": 9,
+  "output_dir": "/ws/sm/shopmanager-grpo-smoke-l4-v2",
+  "run_config": {
+    "model": "Qwen/Qwen3-0.6B",
+    "env_url": "https://hard007ik-shopmanagereng.hf.space",
+    "dataset_size": 300,
+    "num_generations": 2,
+    "per_device_batch": 2,
+    "grad_accum": 1,
+    "max_completion_length": 64,
+    "max_turns": 15,
+    "lr": 5e-06,
+    "warmup_steps": 10,
+    "max_steps": 8,
+    "epochs": 1,
+    "vllm_gpu_mem": 0.3,
+    "reward_weights": [
+      1.0,
+      0.0,
+      0.0,
+      0.0
+    ],
+    "precision": {
+      "bf16": true
+    }
+  }
+}