Upload with huggingface_hub

README.md

@@ -0,0 +1,64 @@
+---
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: xlnet-large_spell_5k_2_p3
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# xlnet-large_spell_5k_2_p3
+
+This model is a fine-tuned version of [model_saves/xlnet-large_spell_5k_2_p2](https://huggingface.co/model_saves/xlnet-large_spell_5k_2_p2) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4726
+- Accuracy: 0.9405
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 128
+- eval_batch_size: 128
+- seed: 52
+- distributed_type: multi-GPU
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: constant
+- num_epochs: 15.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| No log        | 1.0   | 268  | 0.4353          | 0.9405   |
+| No log        | 2.0   | 536  | 0.4413          | 0.9400   |
+| No log        | 3.0   | 804  | 0.4726          | 0.9405   |
+| 0.3275        | 4.0   | 1072 | 0.5153          | 0.9397   |
+| 0.3275        | 5.0   | 1340 | 0.5466          | 0.9391   |
+| 0.3275        | 6.0   | 1608 | 0.5922          | 0.9385   |
+
+
+### Framework versions
+
+- Transformers 4.21.0.dev0
+- Pytorch 1.11.0+cu113
+- Datasets 2.3.2
+- Tokenizers 0.12.1

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.9405013472869778,
+    "eval_loss": 0.4725654423236847,
+    "eval_runtime": 4.757,
+    "eval_samples": 4382,
+    "eval_samples_per_second": 921.177,
+    "eval_steps_per_second": 7.358,
+    "train_loss": 0.2767679157541759,
+    "train_runtime": 670.3814,
+    "train_samples": 34304,
+    "train_samples_per_second": 767.563,
+    "train_steps_per_second": 5.997
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.9405013472869778,
+    "eval_loss": 0.4725654423236847,
+    "eval_runtime": 4.757,
+    "eval_samples": 4382,
+    "eval_samples_per_second": 921.177,
+    "eval_steps_per_second": 7.358
+}

inference_tweak_params.json

@@ -0,0 +1 @@
+{"min_error_probability": 0.64, "additional_confidence": 0.0}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e890f332a0d2bf4b96870aeb809e3515924e70496f36b69186d14ef7e772089c
+size 1461731013

special_tokens_map.json

@@ -0,0 +1,18 @@
+{
+  "additional_special_tokens": [
+    "$START"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<cls>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,28 @@
+{
+  "add_prefix_space": true,
+  "add_special_tokens": false,
+  "additional_special_tokens": [
+    "<eop>",
+    "<eod>"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<cls>",
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "keep_accents": false,
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "name_or_path": "model_saves/xlnet-large_spell_5k_2_p2",
+  "pad_token": "<pad>",
+  "remove_space": true,
+  "sep_token": "<sep>",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "XLNetTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.2767679157541759,
+    "train_runtime": 670.3814,
+    "train_samples": 34304,
+    "train_samples_per_second": 767.563,
+    "train_steps_per_second": 5.997
+}

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_metric": 0.9405013472869778,
+  "best_model_checkpoint": "model_saves/xlnet-large_spell_5k_2_p3/checkpoint-804",
+  "epoch": 6.0,
+  "global_step": 1608,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9404566240678004,
+      "eval_loss": 0.43533819913864136,
+      "eval_runtime": 4.8033,
+      "eval_samples_per_second": 912.282,
+      "eval_steps_per_second": 7.287,
+      "step": 268
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9400093918760273,
+      "eval_loss": 0.44132065773010254,
+      "eval_runtime": 4.7938,
+      "eval_samples_per_second": 914.089,
+      "eval_steps_per_second": 7.301,
+      "step": 536
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9405013472869778,
+      "eval_loss": 0.4725654423236847,
+      "eval_runtime": 4.8004,
+      "eval_samples_per_second": 912.849,
+      "eval_steps_per_second": 7.291,
+      "step": 804
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 1e-05,
+      "loss": 0.3275,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9397354621585662,
+      "eval_loss": 0.5153043270111084,
+      "eval_runtime": 4.8338,
+      "eval_samples_per_second": 906.524,
+      "eval_steps_per_second": 7.241,
+      "step": 1072
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9390534330661121,
+      "eval_loss": 0.5466041564941406,
+      "eval_runtime": 4.7698,
+      "eval_samples_per_second": 918.69,
+      "eval_steps_per_second": 7.338,
+      "step": 1340
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9384832120216013,
+      "eval_loss": 0.5921659469604492,
+      "eval_runtime": 4.746,
+      "eval_samples_per_second": 923.305,
+      "eval_steps_per_second": 7.375,
+      "step": 1608
+    },
+    {
+      "epoch": 6.0,
+      "step": 1608,
+      "total_flos": 2.551532092391424e+16,
+      "train_loss": 0.2767679157541759,
+      "train_runtime": 670.3814,
+      "train_samples_per_second": 767.563,
+      "train_steps_per_second": 5.997
+    }
+  ],
+  "max_steps": 4020,
+  "num_train_epochs": 15,
+  "total_flos": 2.551532092391424e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e216ec84e57661f3069adfd7931d9f52ae57db9b18ecfa83c8606fa3fbae6d2
+size 3439