anima_closeout v6: CELL 1's shard loop variable was `sh`, clobbering CELL 0's sh() shell helper in the shared notebook namespace -> CELL 2 smoke crashed with 'str' object is not callable. Renamed + builder now AST-checks every cell for cross-cell function-name clobbering (proven to flag the exact bug).
Browse files
colab/anima_closeout.ipynb
CHANGED
|
@@ -243,13 +243,13 @@
|
|
| 243 |
" n_seen = n_dropped = n_dup = 0\n",
|
| 244 |
" shards_used, local_paths = [], {}\n",
|
| 245 |
" THIN = [\"id\", \"caption_joycaption\", \"prompt_fused\", \"age_audit\", \"audit\"]\n",
|
| 246 |
-
" for
|
| 247 |
" if len(pool_ids) >= TARGET:\n",
|
| 248 |
" break\n",
|
| 249 |
-
" local = hf_hub_download(ANIMA[\"DATASET\"],
|
| 250 |
" local_dir=str(SRC))\n",
|
| 251 |
-
" local_paths[
|
| 252 |
-
" shards_used.append(
|
| 253 |
" pf = pq.ParquetFile(local)\n",
|
| 254 |
" cols = [c for c in THIN if c in pf.schema_arrow.names]\n",
|
| 255 |
" for rg in range(pf.num_row_groups):\n",
|
|
@@ -266,7 +266,7 @@
|
|
| 266 |
" elif len(pool_ids) < TARGET:\n",
|
| 267 |
" seen.add(rid)\n",
|
| 268 |
" pool_ids.append(rid)\n",
|
| 269 |
-
" print(f\"[scan] {
|
| 270 |
" f\"(seen {n_seen}, dropped {n_dropped}, dup {n_dup})\",\n",
|
| 271 |
" flush=True)\n",
|
| 272 |
"\n",
|
|
@@ -304,8 +304,8 @@
|
|
| 304 |
" if name in t.schema.names\n",
|
| 305 |
" else pa.array([\"\"] * t.num_rows))\n",
|
| 306 |
"\n",
|
| 307 |
-
" for
|
| 308 |
-
" pf = pq.ParquetFile(local_paths[
|
| 309 |
" cols = [c for c in HEAVY if c in pf.schema_arrow.names]\n",
|
| 310 |
" for rg in range(pf.num_row_groups):\n",
|
| 311 |
" tbl = pf.read_row_group(rg, columns=cols)\n",
|
|
|
|
| 243 |
" n_seen = n_dropped = n_dup = 0\n",
|
| 244 |
" shards_used, local_paths = [], {}\n",
|
| 245 |
" THIN = [\"id\", \"caption_joycaption\", \"prompt_fused\", \"age_audit\", \"audit\"]\n",
|
| 246 |
+
" for shard in shard_order: # NOT `sh` \u2014 that is the shell fn\n",
|
| 247 |
" if len(pool_ids) >= TARGET:\n",
|
| 248 |
" break\n",
|
| 249 |
+
" local = hf_hub_download(ANIMA[\"DATASET\"], shard, repo_type=\"dataset\",\n",
|
| 250 |
" local_dir=str(SRC))\n",
|
| 251 |
+
" local_paths[shard] = local\n",
|
| 252 |
+
" shards_used.append(shard)\n",
|
| 253 |
" pf = pq.ParquetFile(local)\n",
|
| 254 |
" cols = [c for c in THIN if c in pf.schema_arrow.names]\n",
|
| 255 |
" for rg in range(pf.num_row_groups):\n",
|
|
|
|
| 266 |
" elif len(pool_ids) < TARGET:\n",
|
| 267 |
" seen.add(rid)\n",
|
| 268 |
" pool_ids.append(rid)\n",
|
| 269 |
+
" print(f\"[scan] {shard}: pool {len(pool_ids)}/{TARGET} \"\n",
|
| 270 |
" f\"(seen {n_seen}, dropped {n_dropped}, dup {n_dup})\",\n",
|
| 271 |
" flush=True)\n",
|
| 272 |
"\n",
|
|
|
|
| 304 |
" if name in t.schema.names\n",
|
| 305 |
" else pa.array([\"\"] * t.num_rows))\n",
|
| 306 |
"\n",
|
| 307 |
+
" for shard in shards_used:\n",
|
| 308 |
+
" pf = pq.ParquetFile(local_paths[shard])\n",
|
| 309 |
" cols = [c for c in HEAVY if c in pf.schema_arrow.names]\n",
|
| 310 |
" for rg in range(pf.num_row_groups):\n",
|
| 311 |
" tbl = pf.read_row_group(rg, columns=cols)\n",
|