ce-checkpoints / v14p21 /ckpt_042000.json
icarus112's picture
Upload v14p21/ckpt_042000.json with huggingface_hub
bbce209 verified
Raw
History Blame Contribute Delete
16.2 kB
{"sent": 42000, "total_steps": 766169, "eval_results": [{"sent": 2000, "train_loss": 2.034756302735293, "train_acc": 0.45750275547157926, "eval_loss": 1.352896087921218, "eval_acc": 0.5675675675675675, "ppl": 3.8686131665524304, "neural_p_loss": 2.6708052447679975, "neural_p_acc": 0.35135135135135137, "neural_p_ppl": 14.451601585278466, "neural_w_loss": 2.6059707874432254, "neural_w_acc": 0.35135135135135137, "neural_w_ppl": 13.544367624531349, "w_freq": 0.02762921568485441, "w_bigram": 0.05736157689369715, "w_trigram": 0.2556483969794982, "w_skip_bi": 0.06602301041330799, "w_skip_tri": 0.7029348773222942, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 1593, "elapsed_s": 33.13985323905945, "steps_per_sec": 958.2118475579963}, {"sent": 4000, "train_loss": 1.9871752307832833, "train_acc": 0.45798861480075903, "eval_loss": 1.3621331294360826, "eval_acc": 0.5609756097560976, "ppl": 3.904513257031836, "neural_p_loss": 3.0075657267441636, "neural_p_acc": 0.3048780487804878, "neural_p_ppl": 20.238074909812454, "neural_w_loss": 2.853650904935056, "neural_w_acc": 0.3048780487804878, "neural_w_ppl": 17.351013244292663, "w_freq": 0.041958338932862195, "w_bigram": 0.04556890695642839, "w_trigram": 0.25534014281640255, "w_skip_bi": 0.21001471098001054, "w_skip_tri": 0.5418681264423018, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 2198, "elapsed_s": 60.683359146118164, "steps_per_sec": 957.5112653221819}, {"sent": 6000, "train_loss": 1.9963971557578784, "train_acc": 0.45615418910113514, "eval_loss": 1.5623044802657207, "eval_acc": 0.5, "ppl": 4.769800500665318, "neural_p_loss": 3.1973851453388473, "neural_p_acc": 0.2755102040816326, "neural_p_ppl": 24.468464993275628, "neural_w_loss": 2.9878445692114024, "neural_w_acc": 0.29591836734693877, "neural_w_ppl": 19.842866439365235, "w_freq": 0.05217992852142049, "w_bigram": 0.10062519196738448, "w_trigram": 0.23244254603635675, "w_skip_bi": 0.10069160130565644, "w_skip_tri": 0.6199743643753467, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 2819, "elapsed_s": 90.59811115264893, "steps_per_sec": 963.1988889146786}, {"sent": 8000, "train_loss": 1.9600586370216024, "train_acc": 0.46167960055345003, "eval_loss": 1.459280730407222, "eval_acc": 0.5084745762711864, "ppl": 4.302863496163115, "neural_p_loss": 2.7714404639924903, "neural_p_acc": 0.3220338983050847, "neural_p_ppl": 15.981638411983257, "neural_w_loss": 2.831258966643224, "neural_w_acc": 0.3220338983050847, "neural_w_ppl": 16.96680802936297, "w_freq": 0.01200933225018967, "w_bigram": 0.08472161232066765, "w_trigram": 0.19069835185323336, "w_skip_bi": 0.23624146063096285, "w_skip_tri": 0.6688825266958343, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 3459, "elapsed_s": 124.2912380695343, "steps_per_sec": 969.5775975180253}, {"sent": 10000, "train_loss": 1.928708914269561, "train_acc": 0.4654636329294162, "eval_loss": 1.3571800480176663, "eval_acc": 0.6176470588235294, "ppl": 3.885221700734973, "neural_p_loss": 3.0372149639363846, "neural_p_acc": 0.29411764705882354, "neural_p_ppl": 20.847102378535944, "neural_w_loss": 2.997333934640992, "neural_w_acc": 0.29411764705882354, "neural_w_ppl": 20.032058888623784, "w_freq": 0.004312531445526712, "w_bigram": 0.04938791094812298, "w_trigram": 0.15022623023631929, "w_skip_bi": 0.08103724610297994, "w_skip_tri": 0.9961421230232116, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 4067, "elapsed_s": 159.87436866760254, "steps_per_sec": 969.0296280206303}, {"sent": 12000, "train_loss": 1.7988620508447908, "train_acc": 0.5024344275168839, "eval_loss": 1.5005923709647202, "eval_acc": 0.5263157894736842, "ppl": 4.484344679291401, "neural_p_loss": 3.1631840533862685, "neural_p_acc": 0.2894736842105263, "neural_p_ppl": 23.645765573425734, "neural_w_loss": 2.995208448675175, "neural_w_acc": 0.34210526315789475, "neural_w_ppl": 19.98952624586967, "w_freq": 0.0014326126602689187, "w_bigram": 0.013075289355986236, "w_trigram": 0.15218068265622336, "w_skip_bi": 0.019038405740286334, "w_skip_tri": 1.0453717343635036, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 4455, "elapsed_s": 193.94419884681702, "steps_per_sec": 962.9470801934483}, {"sent": 14000, "train_loss": 1.8944028245350273, "train_acc": 0.47702280356522747, "eval_loss": 1.8254705634127903, "eval_acc": 0.5211267605633803, "ppl": 6.205714514347845, "neural_p_loss": 3.2968962108804827, "neural_p_acc": 0.30985915492957744, "neural_p_ppl": 27.028617466863356, "neural_w_loss": 3.1950254247213654, "neural_w_acc": 0.323943661971831, "neural_w_ppl": 24.41079432204936, "w_freq": 0.003375580926078459, "w_bigram": 0.020003362793722404, "w_trigram": 0.02415162957412654, "w_skip_bi": 0.020991522651497722, "w_skip_tri": 0.9537335828623482, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 4946, "elapsed_s": 229.31658124923706, "steps_per_sec": 965.1024744672113}, {"sent": 16000, "train_loss": 1.7916851813835162, "train_acc": 0.4957093134901584, "eval_loss": 1.592353545175441, "eval_acc": 0.6103896103896104, "ppl": 4.915303714980986, "neural_p_loss": 3.064480695378446, "neural_p_acc": 0.3116883116883117, "neural_p_ppl": 21.42333386170474, "neural_w_loss": 3.041215345583282, "neural_w_acc": 0.3116883116883117, "neural_w_ppl": 20.930665775582888, "w_freq": 0.0024329292643762088, "w_bigram": 0.018976655950950386, "w_trigram": 0.09692114021536256, "w_skip_bi": 0.1338353646211841, "w_skip_tri": 0.872418871759108, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 5338, "elapsed_s": 263.2176640033722, "steps_per_sec": 967.4198764895112}, {"sent": 18000, "train_loss": 1.8017173520407517, "train_acc": 0.49939166021457804, "eval_loss": 1.5351673653480973, "eval_acc": 0.5512820512820513, "ppl": 4.642102392024997, "neural_p_loss": 3.0312173832893987, "neural_p_acc": 0.32051282051282054, "neural_p_ppl": 20.722444397583914, "neural_w_loss": 2.9961780250033767, "neural_w_acc": 0.32051282051282054, "neural_w_ppl": 20.008917016226523, "w_freq": 0.0021521366947774073, "w_bigram": 0.0003686755876820033, "w_trigram": 0.08770442976179554, "w_skip_bi": 0.05860403255922668, "w_skip_tri": 1.0285428568659212, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 5752, "elapsed_s": 299.0590331554413, "steps_per_sec": 972.4033309799687}, {"sent": 20000, "train_loss": 1.7430485202986175, "train_acc": 0.5092219191080536, "eval_loss": 1.3647259324529617, "eval_acc": 0.6126126126126126, "ppl": 3.9146500264284394, "neural_p_loss": 2.8856868652908805, "neural_p_acc": 0.3063063063063063, "neural_p_ppl": 17.915869158959204, "neural_w_loss": 2.914304333995505, "neural_w_acc": 0.3063063063063063, "neural_w_ppl": 18.435982654262375, "w_freq": 0.023178140802986327, "w_bigram": 0.06471716918991578, "w_trigram": 0.05405259436521933, "w_skip_bi": 0.05532135423068759, "w_skip_tri": 0.8886153881047714, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 6098, "elapsed_s": 332.4108462333679, "steps_per_sec": 976.2918499120357}, {"sent": 22000, "train_loss": 1.782738727002634, "train_acc": 0.4991849302662561, "eval_loss": 1.7438273745376343, "eval_acc": 0.4782608695652174, "ppl": 5.719191072719693, "neural_p_loss": 3.0476664680146386, "neural_p_acc": 0.2608695652173913, "neural_p_ppl": 21.066128536703815, "neural_w_loss": 3.04421873711849, "neural_w_acc": 0.2463768115942029, "neural_w_ppl": 20.99362325565598, "w_freq": 0.0071663158933466965, "w_bigram": 0.00930713677634622, "w_trigram": 0.11245278322593703, "w_skip_bi": 0.047501737959639405, "w_skip_tri": 0.9773995332825159, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 6393, "elapsed_s": 365.43560695648193, "steps_per_sec": 978.7114150663256}, {"sent": 24000, "train_loss": 1.7889105976312585, "train_acc": 0.500903683743551, "eval_loss": 1.4512491458118961, "eval_acc": 0.6067415730337079, "ppl": 4.268443094204289, "neural_p_loss": 3.035928115431413, "neural_p_acc": 0.2808988764044944, "neural_p_ppl": 20.820292569787995, "neural_w_loss": 3.0262349836095326, "neural_w_acc": 0.29213483146067415, "neural_w_ppl": 20.61945368076461, "w_freq": 0.002684633632562478, "w_bigram": 0.01927017451844042, "w_trigram": 0.11261337570860293, "w_skip_bi": 0.040353453336810875, "w_skip_tri": 0.8845841577659969, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 6673, "elapsed_s": 396.2719557285309, "steps_per_sec": 979.345104769064}, {"sent": 26000, "train_loss": 1.8331113641062142, "train_acc": 0.4845958902018259, "eval_loss": 1.688622110182927, "eval_acc": 0.47580645161290325, "ppl": 5.412018400151453, "neural_p_loss": 3.1265865610833665, "neural_p_acc": 0.24193548387096775, "neural_p_ppl": 22.79603371783048, "neural_w_loss": 3.097342767248777, "neural_w_acc": 0.24193548387096775, "neural_w_ppl": 22.13904445751863, "w_freq": 0.00028980688963234586, "w_bigram": 0.00024669458178313045, "w_trigram": 0.10911837798256947, "w_skip_bi": 0.07219731349502292, "w_skip_tri": 0.9082050864494834, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 6984, "elapsed_s": 433.18279933929443, "steps_per_sec": 971.5021017498314}, {"sent": 28000, "train_loss": 2.0354776854698717, "train_acc": 0.4362919489701117, "eval_loss": 1.6900754654721977, "eval_acc": 0.49079754601226994, "ppl": 5.419889704234146, "neural_p_loss": 3.22381893289846, "neural_p_acc": 0.24539877300613497, "neural_p_ppl": 25.12388363359761, "neural_w_loss": 3.1798019336429295, "neural_w_acc": 0.24539877300613497, "neural_w_ppl": 24.041991170838386, "w_freq": 0.06120326262084201, "w_bigram": 0.045085546804469814, "w_trigram": 0.1710564710560188, "w_skip_bi": 0.1060605483013625, "w_skip_tri": 0.6121802197201829, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 7648, "elapsed_s": 478.1594407558441, "steps_per_sec": 972.4141371443101}, {"sent": 30000, "train_loss": 2.1245365424623555, "train_acc": 0.40860972544838353, "eval_loss": 1.689479278763187, "eval_acc": 0.46357615894039733, "ppl": 5.416659401055769, "neural_p_loss": 3.1900912710482014, "neural_p_acc": 0.23841059602649006, "neural_p_ppl": 24.290644374495738, "neural_w_loss": 3.1764933166031972, "neural_w_acc": 0.23841059602649006, "neural_w_ppl": 23.962576877369067, "w_freq": 0.009440722899971252, "w_bigram": 0.019175844665860246, "w_trigram": 0.18273217169529207, "w_skip_bi": 0.10399082560349197, "w_skip_tri": 0.6298211246801771, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 8515, "elapsed_s": 530.3793792724609, "steps_per_sec": 975.6996976576647}, {"sent": 32000, "train_loss": 2.0812049548426583, "train_acc": 0.419365814929436, "eval_loss": 1.505766159612692, "eval_acc": 0.5726495726495726, "ppl": 4.507605853197383, "neural_p_loss": 3.1877225018788993, "neural_p_acc": 0.23076923076923078, "neural_p_ppl": 24.233173539442777, "neural_w_loss": 3.150081177075829, "neural_w_acc": 0.23076923076923078, "neural_w_ppl": 23.337959011317864, "w_freq": 0.0013836769782898067, "w_bigram": 0.00900241312642705, "w_trigram": 0.1704882340853834, "w_skip_bi": 0.11902786103131086, "w_skip_tri": 0.7180221255922719, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 9486, "elapsed_s": 591.339371919632, "steps_per_sec": 979.4849244009399}, {"sent": 34000, "train_loss": 2.0695108420662978, "train_acc": 0.4232196982597146, "eval_loss": 1.621684443678378, "eval_acc": 0.5324675324675324, "ppl": 5.061609135324841, "neural_p_loss": 3.1374996262597983, "neural_p_acc": 0.24675324675324675, "neural_p_ppl": 23.046170717748474, "neural_w_loss": 3.1132382719530605, "neural_w_acc": 0.24675324675324675, "neural_w_ppl": 22.49376752506418, "w_freq": 9.916455516824266e-05, "w_bigram": 8.35405273647475e-05, "w_trigram": 0.17236827072800834, "w_skip_bi": 0.09765749259802337, "w_skip_tri": 0.6993597636982568, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 10546, "elapsed_s": 649.9585614204407, "steps_per_sec": 981.4994952998821}, {"sent": 36000, "train_loss": 2.0547800714302227, "train_acc": 0.43052444138089496, "eval_loss": 1.7366793290628082, "eval_acc": 0.5660377358490566, "ppl": 5.678455797099701, "neural_p_loss": 3.1228900480480046, "neural_p_acc": 0.27358490566037735, "neural_p_ppl": 22.71192343519108, "neural_w_loss": 3.064126468904378, "neural_w_acc": 0.27358490566037735, "neural_w_ppl": 21.415746493590788, "w_freq": 0.0032742319178578685, "w_bigram": 0.00034408106221136636, "w_trigram": 0.15804274182457703, "w_skip_bi": 0.08579290588797428, "w_skip_tri": 0.7018275488548228, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 11088, "elapsed_s": 685.151654958725, "steps_per_sec": 981.9023206483067}, {"sent": 38000, "train_loss": 1.9853641625893987, "train_acc": 0.4523450405803347, "eval_loss": 1.710390196121058, "eval_acc": 0.4883720930232558, "ppl": 5.531119277901323, "neural_p_loss": 3.1088682149345503, "neural_p_acc": 0.20930232558139536, "neural_p_ppl": 22.395682952103403, "neural_w_loss": 3.0374060582180853, "neural_w_acc": 0.23255813953488372, "neural_w_ppl": 20.85108652125183, "w_freq": 0.0009121755009112233, "w_bigram": 0.0013836601599476418, "w_trigram": 0.1578999480743385, "w_skip_bi": 0.0743155762903845, "w_skip_tri": 0.7836194955059553, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 11395, "elapsed_s": 714.8133630752563, "steps_per_sec": 980.1131822520787}, {"sent": 40000, "train_loss": 2.0214433718229845, "train_acc": 0.4380983335303156, "eval_loss": 1.6040378898833512, "eval_acc": 0.6129032258064516, "ppl": 4.97307265681044, "neural_p_loss": 3.1321680673776577, "neural_p_acc": 0.25806451612903225, "neural_p_ppl": 22.923625669911818, "neural_w_loss": 3.0258002186931168, "neural_w_acc": 0.25806451612903225, "neural_w_ppl": 20.610491014176226, "w_freq": 0.0018632907220916834, "w_bigram": 0.0021607310842909075, "w_trigram": 0.14627372308973077, "w_skip_bi": 0.092392635667519, "w_skip_tri": 0.7967095220326451, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 11827, "elapsed_s": 749.6809797286987, "steps_per_sec": 979.6727139399834}, {"sent": 42000, "train_loss": 1.9718594535512548, "train_acc": 0.4470009770857629, "eval_loss": 1.7715192240820656, "eval_acc": 0.45023696682464454, "ppl": 5.879779281711265, "neural_p_loss": 3.264132151489252, "neural_p_acc": 0.2559241706161137, "neural_p_ppl": 26.15740047655673, "neural_w_loss": 3.189327556639467, "neural_w_acc": 0.2559241706161137, "neural_w_ppl": 24.272100341463435, "w_freq": 0.0016707190222532405, "w_bigram": 0.0071221960214408716, "w_trigram": 0.1372427925562817, "w_skip_bi": 0.08546929567544227, "w_skip_tri": 0.8450323951900028, "w_neural_phoneme": 0.0, "w_neural_word": 0.0, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 12148, "elapsed_s": 781.2321453094482, "steps_per_sec": 980.7187333497629}], "mixing_weights": {"w_freq": 0.0016707190222532405, "w_bigram": 0.0071221960214408716, "w_trigram": 0.1372427925562817, "w_skip_bi": 0.08546929567544227, "w_skip_tri": 0.8450323951900028, "w_neural_phoneme": 0.0, "w_neural_word": 0.0}, "phoneme_proto_slots": 1024, "word_proto_slots": 1024, "word_vocab": 12148, "elapsed_s": 781.2321453094482, "steps_per_sec": 980.7187333497629}