SPIRA v2: entropy-patch tokenization + hierarchical dopamine-gated spiking cortex + hippocampal one-shot memory with replay consolidation
032a961 verified | #!/usr/bin/env python3 | |
| """Probe reservoir scaling (E/I balance) + resulting in-vocab accuracy.""" | |
| import sys | |
| import time | |
| import numpy as np | |
| sys.path.insert(0, "/home/z/my-project/download/spira") | |
| from spira import SPIRA, SPIRAConfig | |
| from spira.spiking import SpikingCore | |
| data = open("/home/z/my-project/data/train.bin", "rb").read() | |
| cfg = SPIRAConfig(snn_tokens=40_000, am_tokens=1000, epochs=1) | |
| model = SPIRA.build(data, cfg, log=lambda m: None) | |
| ids = model._ids_train | |
| N = 40_000 | |
| neurons = SPIRA.input_neurons_stream(ids[:N], cfg.d_in, cfg.k_in) | |
| cols = model.cols_of(ids[: N + 1]) | |
| cov = (cols[1:] != cfg.vocab_size)[: N - 1].mean() | |
| print(f"coverage: {cov:.3f}") | |
| variants = [ | |
| # (w_in_scale, w_rec_scale, inh_frac, lam) | |
| (1.5, 0.35, 0.50, 0.90), | |
| (1.2, 0.30, 0.50, 0.90), | |
| (1.5, 0.25, 0.50, 0.90), | |
| (1.0, 0.35, 0.50, 0.90), | |
| (1.5, 0.35, 0.50, 0.70), | |
| (2.0, 0.40, 0.50, 0.90), | |
| ] | |
| for w_in, w_rec, inh, lam in variants: | |
| core = SpikingCore(n_out=cfg.vocab_size + 1, seed=cfg.seed, | |
| w_in_scale=w_in, w_rec_scale=w_rec, inh_frac=inh, lam=lam, | |
| eta_r=0.002) | |
| # firing rate over first 2000 ticks | |
| core.reset() | |
| sp = 0 | |
| for t in range(2000): | |
| core.step(neurons[t]) | |
| sp += int(core.s.sum()) | |
| rate = sp / 2000 / core.n_res | |
| res = core.train_pass(neurons[: N - 1], cols[1:N], log_every=20_000, log_fn=lambda d: None) | |
| print(f"w_in={w_in:.1f} w_rec={w_rec:.2f} inh={inh:.2f} lam={lam:.2f} | " | |
| f"rate={rate:.3f} overall={res['accuracy']:.4f} in-vocab~={res['accuracy']/cov:.4f}") | |