huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/muse
+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/26f575585ec649d88d09a1e402bb936b.1000x1000x1.jpg&#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">Muse</div>
+    <a href="https://genius.com/artists/muse">
+    	<div style="text-align: center; font-size: 14px;">@muse</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 Muse.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/muse).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/muse")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3w58rwod/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 Muse's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3j03atcr) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3j03atcr/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/muse')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/muse")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/muse")
+```
+
+## 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": "4.10.3",
+  "use_cache": true,
+  "vocab_size": 50257
+}

evaluation.txt

@@ -0,0 +1 @@
+{"eval_loss": 2.3269803524017334, "eval_runtime": 2.0389, "eval_samples_per_second": 20.6, "eval_steps_per_second": 2.943, "epoch": 100.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b056be70daa0c9b157c0e26ac48b2f06c4e5212c52b9f67a36a83f017417f67d
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e13d67d67e40251ee94e84a5741483ffc44566e986b76e3c0e77d65319fd10b3
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5f36b4aa1dea8901090d21c164984bd0ab489dc6a881344c90af543c12dcfee
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82f87ce21bae77d6d5460bfce067ada28ee1585e023f4af7e6e79850f7cfa5b9
+size 14567

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94ddf4d086675f377d675c419847ad97dc0ca3884a3889ded4e2611e2e9cf05c
+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 +1,320 @@
+{
+  "best_metric": 2.3269803524017334,
+  "best_model_checkpoint": "output/muse/checkpoint-200",
+  "epoch": 8.0,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0001240985658141214,
+      "loss": 3.1279,
+      "step": 5
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 8.97985658141214e-05,
+      "loss": 2.8709,
+      "step": 10
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.740143418587861e-05,
+      "loss": 2.8349,
+      "step": 15
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.3101434185878613e-05,
+      "loss": 2.5989,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0,
+      "loss": 2.849,
+      "step": 25
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.7842092514038086,
+      "eval_runtime": 1.8206,
+      "eval_samples_per_second": 23.069,
+      "eval_steps_per_second": 3.296,
+      "step": 25
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3101434185878598e-05,
+      "loss": 2.4736,
+      "step": 30
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.74014341858786e-05,
+      "loss": 2.628,
+      "step": 35
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.979856581412138e-05,
+      "loss": 2.6224,
+      "step": 40
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.0001240985658141214,
+      "loss": 2.6052,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0001372,
+      "loss": 2.4193,
+      "step": 50
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.6401476860046387,
+      "eval_runtime": 1.8685,
+      "eval_samples_per_second": 22.478,
+      "eval_steps_per_second": 3.211,
+      "step": 50
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 2.1791,
+      "step": 55
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 8.979856581412141e-05,
+      "loss": 2.2449,
+      "step": 60
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.7401434185878625e-05,
+      "loss": 2.1171,
+      "step": 65
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.310143418587862e-05,
+      "loss": 2.1624,
+      "step": 70
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 2.2355,
+      "step": 75
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.5702805519104004,
+      "eval_runtime": 1.8707,
+      "eval_samples_per_second": 22.451,
+      "eval_steps_per_second": 3.207,
+      "step": 75
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 1.310143418587859e-05,
+      "loss": 2.1067,
+      "step": 80
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 4.740143418587858e-05,
+      "loss": 2.1021,
+      "step": 85
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 8.979856581412137e-05,
+      "loss": 2.0876,
+      "step": 90
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 2.0067,
+      "step": 95
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.8774,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.4998016357421875,
+      "eval_runtime": 1.8837,
+      "eval_samples_per_second": 22.296,
+      "eval_steps_per_second": 3.185,
+      "step": 100
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 0.00012409856581412142,
+      "loss": 1.8595,
+      "step": 105
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 8.97985658141213e-05,
+      "loss": 1.8355,
+      "step": 110
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 4.740143418587876e-05,
+      "loss": 1.8151,
+      "step": 115
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 1.3101434185878628e-05,
+      "loss": 1.6825,
+      "step": 120
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 1.828,
+      "step": 125
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.455634355545044,
+      "eval_runtime": 1.8736,
+      "eval_samples_per_second": 22.417,
+      "eval_steps_per_second": 3.202,
+      "step": 125
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 1.3101434185878583e-05,
+      "loss": 1.7296,
+      "step": 130
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 4.740143418587856e-05,
+      "loss": 1.5323,
+      "step": 135
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 8.979856581412136e-05,
+      "loss": 1.7239,
+      "step": 140
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 1.7961,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.5753,
+      "step": 150
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.415633201599121,
+      "eval_runtime": 1.8716,
+      "eval_samples_per_second": 22.44,
+      "eval_steps_per_second": 3.206,
+      "step": 150
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 0.00012409856581412142,
+      "loss": 1.5374,
+      "step": 155
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 8.979856581412144e-05,
+      "loss": 1.5005,
+      "step": 160
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 4.740143418587866e-05,
+      "loss": 1.3827,
+      "step": 165
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 1.3101434185878644e-05,
+      "loss": 1.5368,
+      "step": 170
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0,
+      "loss": 1.4028,
+      "step": 175
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.395447254180908,
+      "eval_runtime": 1.8726,
+      "eval_samples_per_second": 22.429,
+      "eval_steps_per_second": 3.204,
+      "step": 175
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 1.3101434185878574e-05,
+      "loss": 1.4826,
+      "step": 180
+    },
+    {
+      "epoch": 7.4,
+      "learning_rate": 4.7401434185878544e-05,
+      "loss": 1.3107,
+      "step": 185
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 8.979856581412133e-05,
+      "loss": 1.3342,
+      "step": 190
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 0.00012409856581412136,
+      "loss": 1.3195,
+      "step": 195
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.2922,
+      "step": 200
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.3269803524017334,
+      "eval_runtime": 1.8745,
+      "eval_samples_per_second": 22.406,
+      "eval_steps_per_second": 3.201,
+      "step": 200
+    }
+  ],
+  "max_steps": 2500,
+  "num_train_epochs": 100,
+  "total_flos": 209033625600000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6271ac7a629656367a6532235ac16fadc9c75e5013ae73d36e2e159560675238
+size 2671