amaye15 commited on
Commit
f22025a
·
verified ·
1 Parent(s): 58a3fc3

Upload model files

Browse files
README.md CHANGED
@@ -1,19 +1,28 @@
1
  ---
2
  license: mit
3
  library_name: gguf
 
 
 
 
 
 
4
  tags:
5
  - gguf
6
  - time-series
7
  - forecasting
 
 
 
8
  - rust
9
- base_model: moment-research/MOMENT-1-large
10
  ---
11
 
12
  # moment-rs
13
 
14
  Pure Rust converter and inference engine for [moment-research/MOMENT-1-large](https://huggingface.co/moment-research/MOMENT-1-large).
15
 
16
- Produces GGUF v3 files and runs native forecasting — no Python required.
17
 
18
  ## Build
19
 
@@ -58,7 +67,7 @@ Print all tensor names and shapes from a `.safetensors` checkpoint:
58
 
59
  ## Infer
60
 
61
- Run forecasting from comma-separated context values:
62
 
63
  ```bash
64
  echo '{"context": [1.0, 1.2, 1.5, 1.3, 1.8, 2.0, 1.9, 2.1], "horizon": 96}' \
@@ -85,9 +94,10 @@ Output is JSON in an OpenAI-compatible forecast format:
85
  }
86
  ```
87
 
88
- **Batch inference** — pass multiple series as a nested array to get one `Choice` per series:
89
 
90
  ```bash
 
91
  echo '{"context": [[1.0, 1.2, 1.5], [2.0, 2.2, 2.5]], "horizon": 96}' \
92
  | ./target/release/moment-rs infer --gguf gguf/moment-f16.gguf
93
  ```
 
1
  ---
2
  license: mit
3
  library_name: gguf
4
+ pipeline_tag: time-series-forecasting
5
+ language:
6
+ - en
7
+ base_model: moment-research/MOMENT-1-large
8
+ base_model_relation: quantized
9
+ quantized_by: amaye15
10
  tags:
11
  - gguf
12
  - time-series
13
  - forecasting
14
+ - zero-shot
15
+ - transformer
16
+ - masked-encoder
17
  - rust
18
+ inference: false
19
  ---
20
 
21
  # moment-rs
22
 
23
  Pure Rust converter and inference engine for [moment-research/MOMENT-1-large](https://huggingface.co/moment-research/MOMENT-1-large).
24
 
25
+ Pre-converted GGUF files are available at [amaye15/moment-gguf](https://huggingface.co/amaye15/moment-gguf). Produces GGUF v3 files and runs native forecasting — no Python required.
26
 
27
  ## Build
28
 
 
67
 
68
  ## Infer
69
 
70
+ Run forecasting from stdin JSON:
71
 
72
  ```bash
73
  echo '{"context": [1.0, 1.2, 1.5, 1.3, 1.8, 2.0, 1.9, 2.1], "horizon": 96}' \
 
94
  }
95
  ```
96
 
97
+ **Batch / Multivariate inference** — Moment is channel-independent: each variate is encoded as an independent series. Pass a batch of univariate series to get one `Choice` per series, or use the batch mode to handle multiple variates of a multivariate dataset by submitting each variate as a separate item:
98
 
99
  ```bash
100
+ # Two independent series — one Choice each
101
  echo '{"context": [[1.0, 1.2, 1.5], [2.0, 2.2, 2.5]], "horizon": 96}' \
102
  | ./target/release/moment-rs infer --gguf gguf/moment-f16.gguf
103
  ```
moment-rs/.cargo/config.toml ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ [build]
2
+ rustflags = ["-C", "target-cpu=native"]
moment-rs/Cargo.toml CHANGED
@@ -26,7 +26,11 @@ base64 = "0.22"
26
  tokio-util = { version = "0.7", features = ["io"] }
27
 
28
  [profile.release]
29
- opt-level = 3
 
 
 
 
30
 
31
 
32
  [features]
 
26
  tokio-util = { version = "0.7", features = ["io"] }
27
 
28
  [profile.release]
29
+ opt-level = 3
30
+ lto = "fat"
31
+ codegen-units = 1
32
+ strip = true
33
+ panic = "abort"
34
 
35
 
36
  [features]