Escribí un modelo NN que analiza una imagen y extrae 8 números flotantes al final. El modelo funciona bien (pero lentamente) en mi computadora, así que lo pruebo en la nube de TPU y allí BAM. Tengo un error:

I1008 12: 58: 47.077905 140221679261440 tf_logging.py:115] Error registrado de training_loop: esquema de sistema de archivos '[local]' no implementado (archivo: '/home/gcloud_iba/Data/CGTR/model/GA_subset/model.ckpt- 0_temp_e840841d93124a67b54074b1c0fd7ae4 ') [[{{node save / SaveV2}} = SaveV2 [dtypes = [DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT. trabajador / réplica: 0 / tarea: 0 / dispositivo: CPU: 0 "] (guardar / ShardedFilename, save / SaveV2 / tensor_names, save / SaveV2 / shape_and_slices, batch_normalization / beta / Read / ReadVariableOp, batch_normalization / beta / Momentum / Read_1 / ReadVariableOp, batch_normalization / gamma / Read / ReadVariableOp, batch_normalization / gamma / Momentum / Read_1 / ReadVariableOp, batch_normalization / moving_mean / Read / ReadVariableOp, batch_normalization / moving_variance / Read / ReadVariableOp, batch_normalization_1 / beta / Readum / Beta / Read / Moment / Beta / Readum / Read_1 / ReadVariableOp, normalización_batch_1 / gamma / Read / ReadVariableOp, batch_n ormalization_1 / gamma / Momentum / Read_1 / ReadVariableOp, batch_normalization_1 / moving_mean / Read / ReadVariableOp, batch_normalization_1 / moving_variance / Read / ReadVariableOp, conv2d / kernel / Read / ReadVariableOp, conv2d / kernel / Momentum / Read_1p / Read2 / Read2 / Read2 / Read2 / Read2 / Read2 / Read2 ReadVariableOp, conv2d_1 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_2 / kernel / Read / ReadVariableOp, conv2d_2 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_3 / kernel / Read / ReadVariableOp, conv2d_3 / Kernel / ReadOd2 / Kernel / Read2d_2 / 2 kernel / Read / ReadVariableOp, conv2d_4 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_5 / kernel / Read / ReadVariableOp, conv2d_5 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_6 / kernel / Read / ReadVariableOp / conv2 Readm / O2 / conv2 / M2 / /2 ReadVariableOp, conv2d_7 / kernel / Read / ReadVariableOp, conv2d_7 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_8 / kernel / Read / ReadVariableOp, conv2d_8 / kernel / Momentum / Read_1 / ReadVariableOp, conv2d_ / kernel / V2, Lectura / conv. Momentum / Read_1 / Rea dVariableOp, dense / bias / Read / ReadVariableOp, dense / bias / Momentum / Read_1 / ReadVariableOp, dense / kernel / Read / ReadVariableOp, dense / kernel / Momentum / Read_1 / ReadVariableOp, dense_1 / bias / Read / ReadVariableOp / dense_1 Momentum / Read_1 / ReadVariableOp, dense_1 / kernel / Read / ReadVariableOp, dense_1 / kernel / Momentum / Read_1 / ReadVariableOp, dense_2 / bias / Read / ReadVariableOp, dense_2 / bias / Momentum / Read_1 / ReadVariableOpnelOp / Onse, /nsense2, /nsense2 dense_2 / kernel / Momentum / Read_1 / ReadVariableOp, dense_3 / bias / Read / ReadVariableOp, dense_3 / bias / Momentum / Read_1 / ReadVariableOp, dense_3 / kernel / Read / ReadVariableOp, dense_3 / kernel / Momentum / ReadO / Read / Global / ReadV / Read_1 / Read_1 / Read / Global / Read / ReadVariableOp)]]

Verifiqué que el TPU tiene acceso al disco duro y funciona (tengo otro código que lee el mismo conjunto de datos con otro modelo). No guardo en caché mis datos (todavía) pero hago algunas búsquedas previas. ¿Aparte de esto, no veo lo que no funciona?

Gracias por cualquier ayuda que puedas brindar!

Pi-r