swiss training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 1.0,
+    "init_mem_cpu_alloc_delta": 347038,
+    "init_mem_cpu_peaked_delta": 18306,
+    "init_mem_gpu_alloc_delta": 1109893120,
+    "init_mem_gpu_peaked_delta": 0,
+    "train_mem_cpu_alloc_delta": 894426,
+    "train_mem_cpu_peaked_delta": 4141937,
+    "train_mem_gpu_alloc_delta": 3345894400,
+    "train_mem_gpu_peaked_delta": 3577686016,
+    "train_runtime": 4566.3645,
+    "train_samples": 90000,
+    "train_samples_per_second": 2.464
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "xlm-roberta-base",
+  "architectures": [
+    "XLMRobertaForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.4.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:168d04b8498547b1894465057675cc4d839ddd777f6df71b291bceef7abeeb6b
+size 1109906127

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "xlm-roberta-base"}

train_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 1.0,
+    "init_mem_cpu_alloc_delta": 347038,
+    "init_mem_cpu_peaked_delta": 18306,
+    "init_mem_gpu_alloc_delta": 1109893120,
+    "init_mem_gpu_peaked_delta": 0,
+    "train_mem_cpu_alloc_delta": 894426,
+    "train_mem_cpu_peaked_delta": 4141937,
+    "train_mem_gpu_alloc_delta": 3345894400,
+    "train_mem_gpu_peaked_delta": 3577686016,
+    "train_runtime": 4566.3645,
+    "train_samples": 90000,
+    "train_samples_per_second": 2.464
+}

trainer_state.json

@@ -0,0 +1,155 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "global_step": 11250,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.7777777777777784e-05,
+      "loss": 0.7338,
+      "step": 500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.555555555555556e-05,
+      "loss": 0.1426,
+      "step": 1000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.3333333333333334e-05,
+      "loss": 0.0418,
+      "step": 1500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.111111111111111e-05,
+      "loss": 0.0146,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 0.0263,
+      "step": 2500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 3.6666666666666666e-05,
+      "loss": 0.0448,
+      "step": 3000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.444444444444445e-05,
+      "loss": 0.0153,
+      "step": 3500
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.222222222222223e-05,
+      "loss": 0.0115,
+      "step": 4000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3e-05,
+      "loss": 0.0194,
+      "step": 4500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.0112,
+      "step": 5000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 2.5555555555555554e-05,
+      "loss": 0.0423,
+      "step": 5500
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 0.0059,
+      "step": 6000
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.111111111111111e-05,
+      "loss": 0.0053,
+      "step": 6500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.888888888888889e-05,
+      "loss": 0.0029,
+      "step": 7000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0145,
+      "step": 7500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4444444444444444e-05,
+      "loss": 0.0017,
+      "step": 8000
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2222222222222222e-05,
+      "loss": 0.0021,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 0.0042,
+      "step": 9000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 7.777777777777777e-06,
+      "loss": 0.0049,
+      "step": 9500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 0.002,
+      "step": 10000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.0058,
+      "step": 10500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.1111111111111112e-06,
+      "loss": 0.0035,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0,
+      "step": 11250,
+      "total_flos": 5.753298461184e+16,
+      "train_runtime": 4566.3645,
+      "train_samples_per_second": 2.464
+    }
+  ],
+  "max_steps": 11250,
+  "num_train_epochs": 1,
+  "total_flos": 5.753298461184e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3982c327d454b9125813c818afe9bac4e2b6410c04d8a9415cc10500508a8f42
+size 2287