pierreandLetta Code
d401f6b07e
- restricted pickle VM (whitelisted globals — cannot execute arbitrary Python, unlike torch.load) - zip container reader (store + raw-deflate) - Tensor views: dtype/offset/sizes/strides, f32 materialization - layers: conv2d f32x8 FMA, linear, relu, maxpool2, adaptiveAvgPool2d, softmax, padInput - examples/stripsolver: real .pt forward, 3777 @ 1.0 👾 Generated with [Letta Code](https://letta.com) Co-Authored-By: Letta Code <noreply@letta.com>
63 lines
2.2 KiB
Markdown
63 lines
2.2 KiB
Markdown
# crucible
|
||
|
||
CPU-only inference library for PyTorch `.pt` state dicts, in pure Zig.
|
||
No Python, no venv, no torch — the archive reader and the math both live here.
|
||
|
||
```zig
|
||
const crucible = @import("crucible");
|
||
|
||
var sd = try crucible.loadStateDict(alloc, io, "model.pt");
|
||
defer sd.deinit();
|
||
|
||
const weights = try (sd.get("trunk.0.weight") orelse return error.Missing).toF32(alloc);
|
||
try crucible.layers.conv2d(alloc, &input, h, w, in_c, weights, bias, &out, out_c);
|
||
```
|
||
|
||
## What it is
|
||
|
||
- **Restricted pickle VM** — decodes the data-loading subset of pickle
|
||
(protocol 2) used by `.pt` state dicts. GLOBAL resolution is whitelisted:
|
||
it reads tensors, storages and OrderedDicts, and refuses everything else.
|
||
Unlike `torch.load`, it structurally cannot execute arbitrary Python.
|
||
- **ZIP container reader** — store + raw-deflate entries via `std.zip`.
|
||
- **Tensor views** — dtype (f32/f64/i64/i32/u8), storage offset, sizes,
|
||
strides; contiguous and strided materialization to f32.
|
||
- **Layers (CPU, inference)** — `conv2d` (f32x8 FMA over output width),
|
||
`linear`, `relu`, `maxpool2`, `adaptiveAvgPool2d`, `softmax`, `padInput`.
|
||
|
||
## What it is not
|
||
|
||
- Not a training library. No autograd, no CUDA, no NPU.
|
||
- Not a full pickle implementation — unsupported opcodes and globals are
|
||
errors, not imports.
|
||
- Not a model format converter. state_dict-style checkpoints only.
|
||
|
||
## Usage
|
||
|
||
Add to your `build.zig.zon`:
|
||
|
||
```sh
|
||
zig fetch --save git+https://git.chaosmith.systems/pierre/crucible
|
||
```
|
||
|
||
```zig
|
||
const crucible = b.dependency("crucible", .{});
|
||
exe_mod.addImport("crucible", crucible.module("crucible"));
|
||
```
|
||
|
||
See `examples/stripsolver.zig` for a complete CNN: loads a real trained
|
||
`.pt`, runs a 3×conv + pose-conditioned multi-head forward.
|
||
|
||
## Performance
|
||
|
||
On a 132k-parameter CNN (44×100 input, 3× conv3x3, pose-conditioned
|
||
4-head FC): ~5 ms per forward pass with f32x8 FMA (`-Dcpu=x86_64_v3`),
|
||
2.9 MB RSS. Validated at 99.95% digit accuracy against the reference
|
||
PyTorch implementation — identical predictions.
|
||
|
||
## Status
|
||
|
||
v0.1 — working for real state dicts (Conv2d/Linear/MaxPool2d/
|
||
AdaptiveAvgPool2d/ReLU/Softmax). Layer set grows on demand.
|
||
AVX-512 path: when hardware that has it does.
|