SSVEN_pasqwerty / applied /opt09_srif_softreset /control_baseline.summary.json
p-j-r-1-2-3's picture
add run control_baseline
b3faaec verified
Raw
History Blame Contribute Delete
12.6 kB
Invalid JSON:Unexpected token 'N', ..."_recall": NaN, ""... is not valid JSON
{
"key": "opt09_srif_softreset/control_baseline",
"tag": "control_baseline",
"script": "opt09_srif_softreset.py",
"script_sha256_16": "dd62e980e1fe3fd1",
"hf_folder": "applied/opt09_srif_softreset",
"priority": 0,
"note": "control: hard reset, fp32 membrane",
"phase": 2,
"n_phases": 10,
"method": "harmonic_matched",
"command": [
"/usr/bin/python3",
"-m",
"torch.distributed.run",
"--standalone",
"--nproc_per_node=2",
"/kaggle/working/scripts/opt09_srif_softreset.py",
"--method",
"harmonic_matched",
"--data-root",
"/kaggle/working/data",
"--epochs",
"60",
"--seed",
"1234",
"--batch-size",
"64",
"--event-eval",
"--no-resume",
"--decision-margin",
"2.0",
"--hub-dir",
"/kaggle/working/hub/opt09_srif_softreset/control_baseline",
"--keep-epoch-weights",
"1",
"--resplit",
"block",
"--drop-rest",
"--bench",
"--baseline"
],
"flags": [
"--baseline"
],
"epochs_requested": 60,
"gpus": 2,
"relative_epoch_cost": 1.0,
"exit_code": 0,
"status": "ok",
"duration_s": 338.0,
"started_utc": "2026-08-23T04:39:14.264194+00:00",
"finished_utc": "2026-08-23T04:44:52.262866+00:00",
"optimization_summary": {
"optimization": "srif_softreset",
"research_section": "4. Membrane-Potential Memory Wall -- Soft Reset Integrate-and-Fire (SRIF), dynamic reset potential",
"reference": "Guo et al., 'Reducing Information Loss for Spiking Neural Networks' (SRIF), ECCV 2022; research.md S4 (stacks with MINT / SpQuant-SNN membrane quantization)",
"script": "opt09_srif_softreset.py",
"baseline_mode": true,
"method": "harmonic_matched",
"accuracy": {
"test_acc": 0.9464285714285714,
"test_target_acc": 0.9464285714285714,
"best_val_acc": 0.9428571428571428
},
"efficiency": {
"spikes_per_inference": 7588.84375,
"syn_ops_per_inference": 3247531.347402598,
"dense_macs_per_inference": 122675200.0,
"event_fraction_of_dense": 0.026472598759998744,
"syn_ops_per_inference_dense_fanout": 64824149.13392857,
"weight_sparsity": 0.7324707846410685,
"effective_ticks": 12.189285714285715,
"num_ticks": 100,
"membrane_bytes_per_inference": 451200,
"weight_bytes": 4907008,
"input_raster_bytes": null,
"train_ms_per_step": 95.90769074993659,
"infer_ms_per_sample": 0.6329408046872231,
"peak_vram_mb": 2828.263424,
"discarded_charge_per_reset": 0.27173534693841395
},
"optimization_specific": {
"reset_mode": "hard",
"learn_reset": false,
"lam_init": 0.9,
"v_reset_init": 0.0,
"u_bits": 0,
"u_scale": 1.0,
"u_headroom_requested": 2.0,
"u_headroom_actual": 0.0,
"u_threshold_level": 0,
"u_levels_above_threshold": null,
"reset_family": {
"reset_mode": "hard",
"learn_reset": false,
"num_spiking_sites": 4,
"extra_params": 0,
"extra_params_frac_of_weights": 0.0,
"extra_flops_per_tick_per_neuron": 0,
"learned_lam_per_layer": [
0.0,
0.0,
0.0,
0.0
],
"learned_v_reset_per_layer": [
0.0,
0.0,
0.0,
0.0
]
},
"reset_stats": {
"reset_mode": "hard",
"learn_reset": false,
"discarded_charge_per_reset": 0.27173534693841395,
"spikes_per_inference": 7588.84375,
"syn_per_inference": 3309132.71875,
"lam_per_site": [
0.0,
0.0,
0.0,
0.0
],
"v_reset_per_site": [
0.0,
0.0,
0.0,
0.0
],
"u_bits": 0,
"membrane_saturation_frac": null,
"batches": 4,
"grid_headroom": 0.0,
"grid_levels_above_thresh": null
},
"reset_x_bits_sweep": {
"rows": [
{
"reset_mode": "hard",
"u_bits": 0,
"acc": 0.9464285714285714,
"target_acc": 0.9464285714285714,
"grid_headroom": 0.0,
"grid_levels_above_thresh": null,
"discarded_charge_per_reset": 0.2740163210672487,
"spikes_per_inference": 7531.203125,
"membrane_saturation_frac": null,
"membrane_memory_reduction_x": 1.0,
"acc_drop_vs_fp32": 0.0
},
{
"reset_mode": "hard",
"u_bits": 4,
"acc": 0.96,
"target_acc": 0.96,
"grid_headroom": 1.75,
"grid_levels_above_thresh": 3,
"discarded_charge_per_reset": 0.16058549060468713,
"spikes_per_inference": 9036.453125,
"membrane_saturation_frac": 0.30208174035904256,
"membrane_memory_reduction_x": 8.0,
"acc_drop_vs_fp32": -0.013571428571428568
},
{
"reset_mode": "hard",
"u_bits": 2,
"acc": 0.9007142857142857,
"target_acc": 0.9007142857142857,
"grid_headroom": 1.0,
"grid_levels_above_thresh": 0,
"discarded_charge_per_reset": 0.0,
"spikes_per_inference": 9342.71875,
"membrane_saturation_frac": 0.5301250831117021,
"membrane_memory_reduction_x": 16.0,
"acc_drop_vs_fp32": 0.04571428571428571
},
{
"reset_mode": "soft",
"u_bits": 0,
"acc": 0.8421428571428572,
"target_acc": 0.8421428571428572,
"grid_headroom": 0.0,
"grid_levels_above_thresh": null,
"discarded_charge_per_reset": 0.0,
"spikes_per_inference": 10064.375,
"membrane_saturation_frac": null,
"membrane_memory_reduction_x": 1.0,
"acc_drop_vs_fp32": 0.0
},
{
"reset_mode": "soft",
"u_bits": 4,
"acc": 0.9585714285714285,
"target_acc": 0.9585714285714285,
"grid_headroom": 1.75,
"grid_levels_above_thresh": 3,
"discarded_charge_per_reset": 0.0,
"spikes_per_inference": 10738.90625,
"membrane_saturation_frac": 0.33801418439716313,
"membrane_memory_reduction_x": 8.0,
"acc_drop_vs_fp32": -0.11642857142857133
},
{
"reset_mode": "soft",
"u_bits": 2,
"acc": 0.9007142857142857,
"target_acc": 0.9007142857142857,
"grid_headroom": 1.0,
"grid_levels_above_thresh": 0,
"discarded_charge_per_reset": 0.0,
"spikes_per_inference": 9342.71875,
"membrane_saturation_frac": 0.5301250831117021,
"membrane_memory_reduction_x": 16.0,
"acc_drop_vs_fp32": -0.0585714285714285
},
{
"reset_mode": "trained",
"u_bits": 0,
"acc": 0.9464285714285714,
"target_acc": 0.9464285714285714,
"grid_headroom": 0.0,
"grid_levels_above_thresh": null,
"discarded_charge_per_reset": 0.27401632159846223,
"spikes_per_inference": 7531.203125,
"membrane_saturation_frac": null,
"membrane_memory_reduction_x": 1.0,
"acc_drop_vs_fp32": 0.0
},
{
"reset_mode": "trained",
"u_bits": 4,
"acc": 0.96,
"target_acc": 0.96,
"grid_headroom": 1.75,
"grid_levels_above_thresh": 3,
"discarded_charge_per_reset": 0.1605852697683357,
"spikes_per_inference": 9036.453125,
"membrane_saturation_frac": 0.30208174035904256,
"membrane_memory_reduction_x": 8.0,
"acc_drop_vs_fp32": -0.013571428571428568
},
{
"reset_mode": "trained",
"u_bits": 2,
"acc": 0.9007142857142857,
"target_acc": 0.9007142857142857,
"grid_headroom": 1.0,
"grid_levels_above_thresh": 0,
"discarded_charge_per_reset": 0.0,
"spikes_per_inference": 9342.71875,
"membrane_saturation_frac": 0.5301250831117021,
"membrane_memory_reduction_x": 16.0,
"acc_drop_vs_fp32": 0.04571428571428571
}
],
"by_mode": {
"hard": {
"fp32_acc": 0.9464285714285714,
"acc_by_bits": {
"0": 0.9464285714285714,
"4": 0.96,
"2": 0.9007142857142857
}
},
"soft": {
"fp32_acc": 0.8421428571428572,
"acc_by_bits": {
"0": 0.8421428571428572,
"4": 0.9585714285714285,
"2": 0.9007142857142857
}
},
"trained": {
"fp32_acc": 0.9464285714285714,
"acc_by_bits": {
"0": 0.9464285714285714,
"4": 0.96,
"2": 0.9007142857142857
}
}
}
},
"mechanism": "V = u - s*((1-lam)*u + lam*thresh) + s*v_r; lam=0,v_r=0 is the baseline hard reset exactly, lam=1,v_r=0 discards zero charge",
"cost": "2 scalars per spiking layer; identical FMA count per tick, so the reset change is free at inference time",
"caveat": "the sweep swaps the reset on ALREADY-TRAINED weights, which under-states what training with each reset would give; and a charge-conserving reset can raise the spike count, so read spikes_per_inference next to the accuracy"
},
"config": {
"method": "harmonic_matched",
"data_root": "/kaggle/working/data",
"subjects": "",
"blocks": "",
"limit": 0,
"drop_rest": true,
"resplit": "block",
"test_blocks": "6",
"class_weight": false,
"epochs": 60,
"batch_size": 64,
"lr": 0.002,
"weight_decay": 0.0005,
"grad_clip": 5.0,
"num_bins": 100,
"window_sec": 5.0,
"hidden": "512,256",
"freq_groups": 8,
"dropout": 0.3,
"spike_drop": 0.1,
"time_jitter": 2,
"label_smoothing": 0.05,
"no_batchnorm": false,
"binary_input": false,
"beta": 1.0,
"thresh": 1.0,
"window": 0.5,
"gain": 1.0,
"alpha": 0.9,
"readout": "spikecount",
"fake_quant": 0,
"amp": false,
"spike_reg": 0.0,
"event_eval": true,
"decision_margin": 2.0,
"export": false,
"seed": 1234,
"hub_dir": "/kaggle/working/hub/opt09_srif_softreset/control_baseline",
"repo_id": "UWU-R-13/SSVEP-SNN",
"push_hf": false,
"keep_epoch_weights": 1,
"no_resume": true,
"selftest": false,
"reset_mode": "srif",
"learn_reset": true,
"lam_init": 0.9,
"v_reset_init": 0.0,
"u_bits": 0,
"u_headroom": 2.0,
"reset_sweep": "hard,soft,trained",
"bits_sweep": "0,4,2",
"baseline": true,
"bench": true,
"bench_warmup": 10,
"bench_iters": 50,
"bench_train_steps": 20,
"sampling_rate": 250.0
}
},
"final_summary": {
"final_test_acc": 0.9464285714285714,
"final_test_target_acc": 0.9464285714285714,
"final_test_rest_recall": NaN,
"best_acc": 0.9428571428571428,
"best_epoch": 60,
"resplit": "block",
"class_weight": false,
"drop_rest": true,
"epochs": 60,
"method": "harmonic_matched",
"n_in": 2880,
"n_classes": 40,
"num_bins": 100
},
"event_driven_summary": {
"acc": 0.9464285714285714,
"anytime_acc": {
"0.25": 0.6157142857142858,
"0.5": 0.855,
"0.75": 0.9192857142857143,
"1.0": 0.9464285714285714
},
"syn_per_inference": 3247531.347402598,
"dense_per_inference": 122675200.0,
"event_fraction_of_dense": 0.026472598759998744,
"syn_per_inference_dense_fanout": 64824149.13392857,
"weight_sparsity": 0.7324707846410685,
"num_ticks": 100,
"dt_ms": 50.0,
"margin_decision_acc": 0.45071428571428573,
"mean_decision_tick": 12.189285714285715,
"mean_decision_ms": 609.4642857142858,
"dense_forward_test_acc": 0.9464285714285714
},
"extra_sweep_file_present": true,
"selftest": {
"mode": "gate",
"passed": true
},
"driver_version": 1,
"zip": {
"name": "control_baseline.zip",
"files": 78,
"bytes": 7975348,
"MB": 8.0,
"sha256": "4186ff3945005fa15ca75a27d3ef1dcacb7656ee24e417d75076703e65e0cf74"
}
}