Model update

New model files after re-training.

README.md

@@ -1,4 +1,9 @@
 ---
+metrics:
+- accuracy
+model-index:
+- name: opt-small-sw
+  results: []
 language:
 - en
 widget:
@@ -6,4 +11,45 @@ widget:
   example_title: "SW Example 1"
 ---
 
-OPT small fine-tuned on the SwitchBoard corpus. More details soon.
+# opt-small-sw
+
+This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the SwitchBoard dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.7166
+- Accuracy: 0.4663
+
+## Model description
+
+More details soon.
+
+## Intended uses & limitations
+
+More details soon.
+
+## Training and evaluation data
+
+More details soon.
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 7
+- eval_batch_size: 7
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 20.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.26.0
+- Pytorch 1.13.1
+- Datasets 2.9.0
+- Tokenizers 0.13.0.dev0

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.46629797756620556,
+    "eval_loss": 2.7165656089782715,
+    "eval_runtime": 13.0377,
+    "eval_samples": 219,
+    "eval_samples_per_second": 16.797,
+    "eval_steps_per_second": 2.454,
+    "perplexity": 15.12827650639429,
+    "train_loss": 1.609383287502609,
+    "train_runtime": 6211.091,
+    "train_samples": 1828,
+    "train_samples_per_second": 5.886,
+    "train_steps_per_second": 0.844
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.46629797756620556,
+    "eval_loss": 2.7165656089782715,
+    "eval_runtime": 13.0377,
+    "eval_samples": 219,
+    "eval_samples_per_second": 16.797,
+    "eval_steps_per_second": 2.454,
+    "perplexity": 15.12827650639429
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3772e3c544f22f54463450169d30345852a35cbc6d07c902be35f48b2686214c
+oid sha256:e7111d9cb1613dcaf14ba32ba3adc776ab0a315c21915cb99c6451552da3c13f
 size 501023389

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 1.609383287502609,
+    "train_runtime": 6211.091,
+    "train_samples": 1828,
+    "train_samples_per_second": 5.886,
+    "train_steps_per_second": 0.844
+}

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "global_step": 5240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.91,
+      "learning_rate": 4.522900763358779e-05,
+      "loss": 2.2998,
+      "step": 500
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 4.0458015267175576e-05,
+      "loss": 2.0546,
+      "step": 1000
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 3.568702290076336e-05,
+      "loss": 1.8768,
+      "step": 1500
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.091603053435115e-05,
+      "loss": 1.7311,
+      "step": 2000
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 2.6145038167938934e-05,
+      "loss": 1.6031,
+      "step": 2500
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 2.1374045801526718e-05,
+      "loss": 1.495,
+      "step": 3000
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 1.6603053435114505e-05,
+      "loss": 1.404,
+      "step": 3500
+    },
+    {
+      "epoch": 15.27,
+      "learning_rate": 1.1832061068702292e-05,
+      "loss": 1.3284,
+      "step": 4000
+    },
+    {
+      "epoch": 17.18,
+      "learning_rate": 7.061068702290078e-06,
+      "loss": 1.2688,
+      "step": 4500
+    },
+    {
+      "epoch": 19.08,
+      "learning_rate": 2.2900763358778625e-06,
+      "loss": 1.2258,
+      "step": 5000
+    },
+    {
+      "epoch": 20.0,
+      "step": 5240,
+      "total_flos": 1.910567337984e+16,
+      "train_loss": 1.609383287502609,
+      "train_runtime": 6211.091,
+      "train_samples_per_second": 5.886,
+      "train_steps_per_second": 0.844
+    }
+  ],
+  "max_steps": 5240,
+  "num_train_epochs": 20,
+  "total_flos": 1.910567337984e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:670b5e761cb7eca2eb51facca3ec0d49c913ce671c790ac1d8c1595ac7186a70
-size 3579
+oid sha256:427110b16d5e4b9dfcca9d31e951734a4782d334b29c648b733aa04313056c52
+size 3515