ce-checkpoints / v14p16 /ckpt_018000.json
icarus112's picture
Upload v14p16/ckpt_018000.json with huggingface_hub
393b0e9 verified
Raw History Blame Contribute Delete
6.89 kB
{"sent": 18000, "total_steps": 290806, "eval_results": [{"sent": 2000, "train_loss": 2.126433063993644, "train_acc": 0.4420721146276177, "eval_loss": 1.5603604731278675, "eval_acc": 0.4864864864864865, "ppl": 4.760536981536906, "neural_p_loss": 2.8910552147071416, "neural_p_acc": 0.2702702702702703, "neural_p_ppl": 18.012306427574416, "neural_w_loss": 2.8706762847838783, "neural_w_acc": 0.3108108108108108, "neural_w_ppl": 17.648949881031317, "w_freq": 0.0006606546977323693, "w_bigram": 0.0038864667250278445, "w_trigram": 0.02143677854837551, "w_skip_bi": 0.24398733039007514, "w_skip_tri": 0.462865588351597, "w_neural_phoneme": 0.4554046285000462, "w_neural_word": 0.4562885175308269, "word_vocab": 1593, "elapsed_s": 27.154215097427368, "steps_per_sec": 1169.4317028153953}, {"sent": 4000, "train_loss": 2.1017483443530787, "train_acc": 0.44808349146110055, "eval_loss": 1.9341591509055989, "eval_acc": 0.45121951219512196, "ppl": 6.91822442626254, "neural_p_loss": 4.007785948793894, "neural_p_acc": 0.17073170731707318, "neural_p_ppl": 55.02490763408735, "neural_w_loss": 3.170647255726072, "neural_w_acc": 0.32926829268292684, "neural_w_ppl": 23.82289887503131, "w_freq": 0.001927260451727949, "w_bigram": 0.00889674903929824, "w_trigram": 0.037892929246574926, "w_skip_bi": 0.16459510100377556, "w_skip_tri": 0.4651920923309866, "w_neural_phoneme": 0.47869333217593085, "w_neural_word": 0.47262632796182374, "word_vocab": 2198, "elapsed_s": 49.305346727371216, "steps_per_sec": 1178.4725969231197}, {"sent": 6000, "train_loss": 2.2799464338812525, "train_acc": 0.42062484996056104, "eval_loss": 3.458364858909436, "eval_acc": 0.32653061224489793, "ppl": 31.7649937801258, "neural_p_loss": 4.097416845083684, "neural_p_acc": 0.23469387755102042, "neural_p_ppl": 60.184620429567694, "neural_w_loss": 3.376138009974124, "neural_w_acc": 0.29591836734693877, "neural_w_ppl": 29.2575602385435, "w_freq": 0.0015334145068691678, "w_bigram": 0.024028759208300738, "w_trigram": 0.0899106976154774, "w_skip_bi": 0.24758365343631503, "w_skip_tri": 0.4640016784013768, "w_neural_phoneme": 0.45104562978213464, "w_neural_word": 0.4677985784696181, "word_vocab": 2819, "elapsed_s": 73.34146499633789, "steps_per_sec": 1189.831700312467}, {"sent": 8000, "train_loss": 2.4013737561295847, "train_acc": 0.4018227756722613, "eval_loss": 1.8160536244520216, "eval_acc": 0.4067796610169492, "ppl": 6.147549975344211, "neural_p_loss": 3.230951564350047, "neural_p_acc": 0.23728813559322035, "neural_p_ppl": 25.303723639994033, "neural_w_loss": 2.710904980247823, "neural_w_acc": 0.3050847457627119, "neural_w_ppl": 15.042882868619948, "w_freq": 9.645189222412739e-05, "w_bigram": 2.6840420531435904e-05, "w_trigram": 0.1254234983275961, "w_skip_bi": 0.21649016476116295, "w_skip_tri": 0.5314471793700382, "w_neural_phoneme": 0.4531481845227636, "w_neural_word": 0.45319692631572206, "word_vocab": 3459, "elapsed_s": 100.56465125083923, "steps_per_sec": 1198.3335943701622}, {"sent": 10000, "train_loss": 2.383003929912084, "train_acc": 0.4034812425536861, "eval_loss": 1.6604919251303263, "eval_acc": 0.5, "ppl": 5.261898668075032, "neural_p_loss": 3.342523949635387, "neural_p_acc": 0.2647058823529412, "neural_p_ppl": 28.290440318093488, "neural_w_loss": 2.8277236063727016, "neural_w_acc": 0.29411764705882354, "neural_w_ppl": 16.906930157629425, "w_freq": 0.0002577675415842086, "w_bigram": 0.006858605857105867, "w_trigram": 0.057255928614800174, "w_skip_bi": 0.3631300953533809, "w_skip_tri": 0.55, "w_neural_phoneme": 0.4536226366219268, "w_neural_word": 0.45337210895852126, "word_vocab": 4067, "elapsed_s": 129.5134997367859, "steps_per_sec": 1196.1919052056703}, {"sent": 12000, "train_loss": 2.166187917651436, "train_acc": 0.43024972514528037, "eval_loss": 1.8950823302834507, "eval_acc": 0.4473684210526316, "ppl": 6.653096133375717, "neural_p_loss": 3.3284754046012965, "neural_p_acc": 0.2894736842105263, "neural_p_ppl": 27.89577948935034, "neural_w_loss": 2.9314989125500834, "neural_w_acc": 0.27631578947368424, "neural_w_ppl": 18.755722625620383, "w_freq": 0.001830647670653317, "w_bigram": 0.010751529318916112, "w_trigram": 0.014880119970841477, "w_skip_bi": 0.4, "w_skip_tri": 0.55, "w_neural_phoneme": 0.4507301277323498, "w_neural_word": 0.460157926054588, "word_vocab": 4455, "elapsed_s": 155.7714912891388, "steps_per_sec": 1198.922848169598}, {"sent": 14000, "train_loss": 2.0723396371472873, "train_acc": 0.4303449473318671, "eval_loss": 2.1827389525727696, "eval_acc": 0.4647887323943662, "ppl": 8.870569084203826, "neural_p_loss": 3.8119023625106525, "neural_p_acc": 0.19718309859154928, "neural_p_ppl": 45.236413119727516, "neural_w_loss": 3.022336178946411, "neural_w_acc": 0.29577464788732394, "neural_w_ppl": 20.539218969970463, "w_freq": 0.0004532895354388001, "w_bigram": 0.0007364133673548604, "w_trigram": 0.0008602769491051947, "w_skip_bi": 0.24866369647317582, "w_skip_tri": 0.5007494316937173, "w_neural_phoneme": 0.4518035771741291, "w_neural_word": 0.45181336182425547, "word_vocab": 4946, "elapsed_s": 183.64421343803406, "steps_per_sec": 1205.123732769705}, {"sent": 16000, "train_loss": 1.9488462174984795, "train_acc": 0.45082213154104656, "eval_loss": 1.9366909511786845, "eval_acc": 0.4935064935064935, "ppl": 6.9357621804315155, "neural_p_loss": 3.387368825035995, "neural_p_acc": 0.2857142857142857, "neural_p_ppl": 29.587998559371375, "neural_w_loss": 2.8479335491562274, "neural_w_acc": 0.3246753246753247, "neural_w_ppl": 17.252094375092256, "w_freq": 0.00047762504251971287, "w_bigram": 0.0044797096534032015, "w_trigram": 0.006856315750209463, "w_skip_bi": 0.2592850819473108, "w_skip_tri": 0.5225626006070283, "w_neural_phoneme": 0.45585823993661867, "w_neural_word": 0.456167286146228, "word_vocab": 5338, "elapsed_s": 210.90789437294006, "steps_per_sec": 1207.3611599844937}, {"sent": 18000, "train_loss": 1.9461497683425277, "train_acc": 0.44903771706669615, "eval_loss": 1.801561795109538, "eval_acc": 0.4230769230769231, "ppl": 6.059103156241893, "neural_p_loss": 3.4020551483139614, "neural_p_acc": 0.11538461538461539, "neural_p_ppl": 30.02574403916486, "neural_w_loss": 3.010162640129175, "neural_w_acc": 0.3076923076923077, "neural_w_ppl": 20.29069973891918, "w_freq": 0.00015452889642913296, "w_bigram": 0.0002021176419538591, "w_trigram": 0.00016794538757380464, "w_skip_bi": 0.3227165168969855, "w_skip_tri": 0.55, "w_neural_phoneme": 0.4508332848479917, "w_neural_word": 0.45082707259464005, "word_vocab": 5752, "elapsed_s": 240.33339738845825, "steps_per_sec": 1210.0107732008687}], "mixing_weights": {"w_freq": 0.00015452889642913296, "w_bigram": 0.0002021176419538591, "w_trigram": 0.00016794538757380464, "w_skip_bi": 0.3227165168969855, "w_skip_tri": 0.55, "w_neural_phoneme": 0.4508332848479917, "w_neural_word": 0.45082707259464005}, "word_vocab": 5752, "elapsed_s": 240.33339738845825, "steps_per_sec": 1210.0107732008687}