gesh75 commited on
Commit
a27656d
·
verified ·
1 Parent(s): 273cb45

Card: add published retrieval+validator numbers and Intent finder link

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -75,7 +75,7 @@ MIT, matching the repo. CLI command names and syntax are functional facts belong
75
  ## Derived
76
 
77
  - [gesh75/netops-intent2cli](https://huggingface.co/datasets/gesh75/netops-intent2cli): synthetic operator intents paired with verbatim commands from this corpus. Default config is v1 (565 train / 102 validation / 108 test). Config `v2` adds 650 training pairs and an 88-row `operator_test`.
78
- - [gesh75/netops-intent2cli-qwen0.5b-lora](https://huggingface.co/gesh75/netops-intent2cli-qwen0.5b-lora): experimental Qwen2.5-0.5B LoRA trained on v2. Published exact match: **12/108 (11.1%)** on the frozen v1 test, **9/88 (10.2%)** on `operator_test`. Not production-safe.
79
 
80
  ## Load
81
 
@@ -97,7 +97,7 @@ df = pd.read_parquet("hf://datasets/gesh75/multivendor-cli-commands/data/train.p
97
  | | |
98
  |---|---|
99
  | 📡 **[netlog-ai](https://huggingface.co/spaces/gesh75/netlog-ai)** | Sanitize-first network log analysis — browser lite demo of [netlog-ai](https://pypi.org/project/netlog-ai/) (PyPI 2.1.0) |
100
- | 🛠️ **[multivendor-cli-configurator](https://huggingface.co/spaces/gesh75/multivendor-cli-configurator)** | 70k+ CLI commands across 17 vendors — cheatsheet, N-vendor compare, CLI Studio |
101
  | 🗂️ **[multivendor-cli-commands](https://huggingface.co/datasets/gesh75/multivendor-cli-commands)** | The same 70,006 commands as a Parquet dataset |
102
  | 🧭 **[netops-intent2cli](https://huggingface.co/datasets/gesh75/netops-intent2cli)** | v1 + v2 synthetic operator intents → corpus command pairs |
103
  | 🤖 **[netops-intent2cli-qwen0.5b-lora](https://huggingface.co/gesh75/netops-intent2cli-qwen0.5b-lora)** | LoRA adapter: English intent → vendor CLI (experimental) |
 
75
  ## Derived
76
 
77
  - [gesh75/netops-intent2cli](https://huggingface.co/datasets/gesh75/netops-intent2cli): synthetic operator intents paired with verbatim commands from this corpus. Default config is v1 (565 train / 102 validation / 108 test). Config `v2` adds 650 training pairs and an 88-row `operator_test`.
78
+ - [gesh75/netops-intent2cli-qwen0.5b-lora](https://huggingface.co/gesh75/netops-intent2cli-qwen0.5b-lora): experimental Qwen2.5-0.5B LoRA trained on v2. Adapter-only exact match: **12/108 (11.1%)** on the frozen v1 test, **9/88 (10.2%)** on `operator_test`. Optional retrieval+validator lookup mode (same card): **63/108** and **58/88** when the gold command is already in this corpus; it cannot return a command that is missing from the base. Not production-safe. Browser demo: [Intent finder](https://gesh75-multivendor-cli-configurator.static.hf.space/intent-finder.html) (retrieval only, no model).
79
 
80
  ## Load
81
 
 
97
  | | |
98
  |---|---|
99
  | 📡 **[netlog-ai](https://huggingface.co/spaces/gesh75/netlog-ai)** | Sanitize-first network log analysis — browser lite demo of [netlog-ai](https://pypi.org/project/netlog-ai/) (PyPI 2.1.0) |
100
+ | 🛠️ **[multivendor-cli-configurator](https://huggingface.co/spaces/gesh75/multivendor-cli-configurator)** | 70k+ CLI commands across 17 vendors — cheatsheet, N-vendor compare, CLI Studio, Intent finder |
101
  | 🗂️ **[multivendor-cli-commands](https://huggingface.co/datasets/gesh75/multivendor-cli-commands)** | The same 70,006 commands as a Parquet dataset |
102
  | 🧭 **[netops-intent2cli](https://huggingface.co/datasets/gesh75/netops-intent2cli)** | v1 + v2 synthetic operator intents → corpus command pairs |
103
  | 🤖 **[netops-intent2cli-qwen0.5b-lora](https://huggingface.co/gesh75/netops-intent2cli-qwen0.5b-lora)** | LoRA adapter: English intent → vendor CLI (experimental) |