"source": "## MANDATORY | difficulty 3 | Exercise 1\n\nBefore running the next cell, predict which of these bug labels goes with each fingerprint:\n\n- shuffled but otherwise familiar values\n- values that look uniformly too large\n- values broken already in a local stage pair\n"
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"pedagogy": {
"role": "mandatory",
"difficulty": 3,
"kind": "exercise",
"title": "Prediction Check"
}
},
"outputs": [],
"source": "# MANDATORY | difficulty 3 | Prediction Check\n\nfingerprints = {\n \"wrong_order\": \"shuffled but same value set\",\n \"missing_scale\": \"same shape but uniformly off by a factor\",\n \"wrong_zeta\": \"local pair outputs go bad immediately\",\n}\nprint(fingerprints)\n"
},
{
"cell_type": "markdown",
"metadata": {
"pedagogy": {
"role": "mandatory",
"difficulty": 3,
"kind": "exercise",
"title": "Exercise 2"
}
},
"source": "## MANDATORY | difficulty 3 | Exercise 2\n\nExplain why \u201calmost right\u201d is a useless debugging description unless you also specify whether the issue is:\n\n- sign\n- order\n- zeta\n- scaling\n"
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"pedagogy": {
"role": "mandatory",
"difficulty": 3,
"kind": "exercise",
"title": "A Small Debugging Drill"
}
},
"outputs": [],
"source": "# MANDATORY | difficulty 3 | A Small Debugging Drill\n\nfrom ntt_learning.toy_ntt import fast_intt_psi_gs_trace, fast_ntt_psi_ct_trace\n\nforward_trace = fast_ntt_psi_ct_trace([5, 6, 7, 8], 7681, 1925)\ninverse_trace = fast_intt_psi_gs_trace(forward_trace.raw_output, 7681, 1925)\n\nprint(\"forward BO output:\", forward_trace.raw_output)\nprint(\"forward NO output:\", forward_trace.normal_order_output)\nprint(\"inverse unscaled:\", inverse_trace.raw_output)\nprint(\"inverse scaled:\", inverse_trace.scaled_output)\n"
},
{
"cell_type": "markdown",
"metadata": {
"pedagogy": {
"role": "mandatory",
"difficulty": 2,
"kind": "reflection",
"title": "Reflection"
}
},
"source": "## MANDATORY | difficulty 2 | Reflection\n\nReflection prompt:\n\n- Which bug fingerprint feels easiest to recognize now?\n- Which one still needs more repetition?\n- What is your debugging order of operations now?\n"
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"pedagogy": {
"role": "facultative",
"difficulty": 4,
"kind": "exploration",
"title": "Optional: Build Your Own Fingerprint Table"
