End of training

README.md

@@ -0,0 +1,69 @@
+---
+language:
+- eng
+license: mit
+base_model: evie-8/speaker-segmentation-fine-tuned-callhome-eng
+tags:
+- speaker-diarization
+- speaker-segmentation
+- generated_from_trainer
+datasets:
+- evie-8/backup_uganda
+model-index:
+- name: speaker-segmentation-fine-tuned-backup-uganda-eng
+  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. -->
+
+# speaker-segmentation-fine-tuned-backup-uganda-eng
+
+This model is a fine-tuned version of [evie-8/speaker-segmentation-fine-tuned-callhome-eng](https://huggingface.co/evie-8/speaker-segmentation-fine-tuned-callhome-eng) on the evie-8/backup_uganda dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3139
+- Der: 0.1059
+- False Alarm: 0.0200
+- Missed Detection: 0.0339
+- Confusion: 0.0520
+
+## 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: 0.001
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Der    | False Alarm | Missed Detection | Confusion |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:-----------:|:----------------:|:---------:|
+| 0.0661        | 1.0   | 1065 | 0.3346          | 0.1149 | 0.0132      | 0.0510           | 0.0507    |
+| 0.1333        | 2.0   | 2130 | 0.3214          | 0.1089 | 0.0194      | 0.0367           | 0.0528    |
+| 0.2857        | 3.0   | 3195 | 0.3139          | 0.1059 | 0.0200      | 0.0339           | 0.0520    |
+
+
+### Framework versions
+
+- Transformers 4.42.4
+- Pytorch 2.3.1+cu121
+- Datasets 2.20.0
+- Tokenizers 0.19.1

config.json

@@ -0,0 +1,19 @@
+{
+  "_name_or_path": "evie-8/speaker-segmentation-fine-tuned-callhome-eng",
+  "architectures": [
+    "SegmentationModel"
+  ],
+  "chunk_duration": 10.0,
+  "max_speakers_per_chunk": 3,
+  "max_speakers_per_frame": 2,
+  "min_duration": null,
+  "model_type": "pyannet",
+  "sample_rate": 16000,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.4",
+  "warm_up": [
+    0.0,
+    0.0
+  ],
+  "weigh_by_cardinality": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:165533d6d7a5e7a739afd1088016adecca853db1fc29ce0598423480d60390f3
+size 5899124

runs/Jul30_07-58-19_6ba39b71e623/events.out.tfevents.1722326320.6ba39b71e623.572.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7311d6052b51150c14c7fae4694c4b6d878e93d8805df6ed4729f7f063b51984
+size 680167

runs/Jul30_07-58-19_6ba39b71e623/events.out.tfevents.1722327291.6ba39b71e623.572.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a81496b300fd5e69fa369cc340506c203242a7b2a7883ea8b93872119422aef9
+size 574

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93434d278a5b7bf47443774718e74ae7a6d32f43c877eb3ffb469bfc81d0be89
+size 5176