huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/madonna
+tags:
+- huggingartists
+- lyrics
+- lm-head
+- causal-lm
+widget:
+- text: "I am"
+---
+
+<div class="inline-flex flex-col" style="line-height: 1.5;">
+    <div class="flex">
+        <div
+			style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/5cad1fbda7135f40c49ea35ad60ba4f5.215x215x1.png&#39;)">
+        </div>
+    </div>
+    <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
+    <div style="text-align: center; font-size: 16px; font-weight: 800">Madonna</div>
+    <a href="https://genius.com/artists/madonna">
+    	<div style="text-align: center; font-size: 14px;">@madonna</div>
+    </a>
+</div>
+
+I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists).
+
+Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)!
+
+## How does it work?
+
+To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist).
+
+## Training data
+
+The model was trained on lyrics from Madonna.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/madonna).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/madonna")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1kywkdsd/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
+
+## Training procedure
+
+The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Madonna's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/kfajgkjk) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/kfajgkjk/artifacts) is logged and versioned.
+
+## How to use
+
+You can use this model directly with a pipeline for text generation:
+
+```python
+from transformers import pipeline
+generator = pipeline('text-generation',
+                     model='huggingartists/madonna')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/madonna")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/madonna")
+```
+
+## Limitations and bias
+
+The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
+
+In addition, the data present in the user's tweets further affects the text generated by the model.
+
+## About
+
+*Built by Aleksey Korshuk*
+
+[![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk)
+
+[![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk)
+
+[![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky)
+
+For more details, visit the project repository.
+
+[![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "resid_pdrop": 0.1,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 200,
+      "min_length": 100,
+      "temperature": 1.0,
+      "top_p": 0.95
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

evaluation.txt

@@ -0,0 +1 @@
+{"eval_loss": 1.924851655960083, "eval_runtime": 14.3209, "eval_samples_per_second": 22.485, "eval_steps_per_second": 2.863, "epoch": 1.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9b6a9dde4f77fe7e5286411ef04c90642d3d5632397b8526fcc446a7864b4f1
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ceae229ca904bed5d1bb6cd9932ed1f1ce4b1f959602365ba45f365e7ae157b8
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a28db54f1ddd728c41eb92a39efd92a5721e39626fd0317625976c4d82ef50fb
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d2957881bb5b55492f8fff01b392d863d09b0e66192bf730f50b9b2472170d6
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33ebdf640e584fb96b06bd917ad6a00182e9ef4871b5e81f588509ccfb511410
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -0,0 +1,288 @@
+{
+  "best_metric": 1.924851655960083,
+  "best_model_checkpoint": "output/madonna/checkpoint-222",
+  "epoch": 1.0,
+  "global_step": 222,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0001370283488463827,
+      "loss": 2.5046,
+      "step": 5
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00013651425439773298,
+      "loss": 2.4348,
+      "step": 10
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00013566028939181128,
+      "loss": 2.5907,
+      "step": 15
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00013447072741690948,
+      "loss": 2.4827,
+      "step": 20
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00013295152152507697,
+      "loss": 2.4145,
+      "step": 25
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0001311102744406261,
+      "loss": 2.1817,
+      "step": 30
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00012895620051300705,
+      "loss": 2.3325,
+      "step": 35
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0001265000796044546,
+      "loss": 2.0885,
+      "step": 40
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00012375420314317213,
+      "loss": 2.1847,
+      "step": 45
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00012073231261202464,
+      "loss": 2.1728,
+      "step": 50
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00011744953078056836,
+      "loss": 2.0993,
+      "step": 55
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00011392228602455959,
+      "loss": 2.3687,
+      "step": 60
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0001101682301116794,
+      "loss": 1.9625,
+      "step": 65
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0001062061498649071,
+      "loss": 2.0389,
+      "step": 70
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00010205587314561534,
+      "loss": 2.098,
+      "step": 75
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.773816962688569e-05,
+      "loss": 2.1281,
+      "step": 80
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.327464685361516e-05,
+      "loss": 1.9415,
+      "step": 85
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 8.868764210957135e-05,
+      "loss": 2.3485,
+      "step": 90
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 8.400011063253747e-05,
+      "loss": 2.1515,
+      "step": 95
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 7.923551073696397e-05,
+      "loss": 1.8548,
+      "step": 100
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 7.441768641901993e-05,
+      "loss": 1.969,
+      "step": 105
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 6.957074803153629e-05,
+      "loss": 2.0221,
+      "step": 110
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 6.471895162599185e-05,
+      "loss": 2.0988,
+      "step": 115
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 5.9886577565364115e-05,
+      "loss": 1.7345,
+      "step": 120
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 5.509780901531541e-05,
+      "loss": 1.9466,
+      "step": 125
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 5.037661092179389e-05,
+      "loss": 2.1125,
+      "step": 130
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.574661008069485e-05,
+      "loss": 2.0488,
+      "step": 135
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.123097689976244e-05,
+      "loss": 2.0849,
+      "step": 140
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.685230944444331e-05,
+      "loss": 1.8891,
+      "step": 145
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.263252034797391e-05,
+      "loss": 1.8136,
+      "step": 150
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.8592727151649647e-05,
+      "loss": 1.9213,
+      "step": 155
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.475314662405725e-05,
+      "loss": 1.9865,
+      "step": 160
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 2.113299358814034e-05,
+      "loss": 2.2742,
+      "step": 165
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7750384762408767e-05,
+      "loss": 1.8535,
+      "step": 170
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4622248097508915e-05,
+      "loss": 2.1606,
+      "step": 175
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.1764238061872434e-05,
+      "loss": 1.7923,
+      "step": 180
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.190657300387505e-06,
+      "loss": 1.8476,
+      "step": 185
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 6.91438505814559e-06,
+      "loss": 1.9976,
+      "step": 190
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.9468127274594585e-06,
+      "loss": 1.9706,
+      "step": 195
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.2977868407019207e-06,
+      "loss": 1.982,
+      "step": 200
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.975559794252193e-06,
+      "loss": 1.9318,
+      "step": 205
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.867485501471999e-07,
+      "loss": 2.035,
+      "step": 210
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.363015221117509e-07,
+      "loss": 1.9653,
+      "step": 215
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.7473811683140093e-08,
+      "loss": 1.8408,
+      "step": 220
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.924851655960083,
+      "eval_runtime": 14.2025,
+      "eval_samples_per_second": 22.672,
+      "eval_steps_per_second": 2.887,
+      "step": 222
+    }
+  ],
+  "max_steps": 222,
+  "num_train_epochs": 1,
+  "total_flos": 232027324416000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83c9e6cfb63c365b4c49a065df2aef2c18c8bd82cedbf8864df569b4417ff289
+size 2671