huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/radiohead
+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/593c69b2e4bb8eb47801ce1952c5d30b.600x600x184.gif&#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">Radiohead</div>
+    <a href="https://genius.com/artists/radiohead">
+    	<div style="text-align: center; font-size: 14px;">@radiohead</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 Radiohead.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/radiohead).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/radiohead")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/35vxvq9n/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 Radiohead's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2bulf32i) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2bulf32i/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/radiohead')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/radiohead")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/radiohead")
+```
+
+## 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 +1,42 @@
+{
+  "_name_or_path": "radiohead",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "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": "4.17.0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

evaluation.txt

@@ -0,0 +1 @@
+{"eval_loss": 2.588986396789551, "eval_runtime": 2.2503, "eval_samples_per_second": 20.886, "eval_steps_per_second": 2.666, "epoch": 10.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cd239cc8addb71ee18a5f198cb109315f36fd08edfb5ec0fc239849bf7f4d12
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4d98cf3591d6c065ea13072f4ac3f498eb65a2cfce7305d069519c25f85ffce
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28fbaae1bf45583e5cb67ad32c1d88c478e1557757fa838131a33632b71301c7
+size 510404393

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acf452d303eb4056fcb17fa527a28038577d972c97ad69222bf34a63d415312b
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c35e4601ff589d570a695430c49d5028d31da3a18518f505c21842dd34307806
+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 +1,368 @@
+{
+  "best_metric": 2.588986396789551,
+  "best_model_checkpoint": "output/radiohead/checkpoint-240",
+  "epoch": 8.0,
+  "global_step": 240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0001280093426996125,
+      "loss": 3.2073,
+      "step": 5
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00010290000000000001,
+      "loss": 2.9899,
+      "step": 10
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 6.86e-05,
+      "loss": 2.8292,
+      "step": 15
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.4300000000000014e-05,
+      "loss": 2.8291,
+      "step": 20
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.190657300387505e-06,
+      "loss": 2.8479,
+      "step": 25
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0,
+      "loss": 2.9261,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.8060529232025146,
+      "eval_runtime": 2.0848,
+      "eval_samples_per_second": 22.545,
+      "eval_steps_per_second": 2.878,
+      "step": 30
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.190657300387513e-06,
+      "loss": 2.6517,
+      "step": 35
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.429999999999997e-05,
+      "loss": 2.7329,
+      "step": 40
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 6.859999999999999e-05,
+      "loss": 2.7524,
+      "step": 45
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00010290000000000001,
+      "loss": 2.5591,
+      "step": 50
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00012800934269961248,
+      "loss": 2.6201,
+      "step": 55
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0001372,
+      "loss": 2.5207,
+      "step": 60
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.7088782787323,
+      "eval_runtime": 2.1182,
+      "eval_samples_per_second": 22.189,
+      "eval_steps_per_second": 2.833,
+      "step": 60
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00012800934269961253,
+      "loss": 2.3985,
+      "step": 65
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00010289999999999998,
+      "loss": 2.3027,
+      "step": 70
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 6.860000000000001e-05,
+      "loss": 2.1903,
+      "step": 75
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 3.4300000000000054e-05,
+      "loss": 2.2458,
+      "step": 80
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 9.190657300387498e-06,
+      "loss": 2.2898,
+      "step": 85
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 2.2945,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.6637511253356934,
+      "eval_runtime": 2.1338,
+      "eval_samples_per_second": 22.027,
+      "eval_steps_per_second": 2.812,
+      "step": 90
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 9.190657300387474e-06,
+      "loss": 2.1611,
+      "step": 95
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.4300000000000014e-05,
+      "loss": 2.0392,
+      "step": 100
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 6.859999999999997e-05,
+      "loss": 2.3157,
+      "step": 105
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 0.00010289999999999994,
+      "loss": 2.1789,
+      "step": 110
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 0.0001280093426996125,
+      "loss": 2.1359,
+      "step": 115
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0001372,
+      "loss": 2.0964,
+      "step": 120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.617748737335205,
+      "eval_runtime": 2.1438,
+      "eval_samples_per_second": 21.924,
+      "eval_steps_per_second": 2.799,
+      "step": 120
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 0.00012800934269961248,
+      "loss": 1.8717,
+      "step": 125
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 0.00010290000000000009,
+      "loss": 1.9838,
+      "step": 130
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 6.860000000000003e-05,
+      "loss": 2.0446,
+      "step": 135
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.429999999999996e-05,
+      "loss": 1.8428,
+      "step": 140
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 9.190657300387574e-06,
+      "loss": 1.9594,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 2.0192,
+      "step": 150
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6335015296936035,
+      "eval_runtime": 2.1534,
+      "eval_samples_per_second": 21.826,
+      "eval_steps_per_second": 2.786,
+      "step": 150
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 9.190657300387467e-06,
+      "loss": 1.7401,
+      "step": 155
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 3.429999999999989e-05,
+      "loss": 1.8589,
+      "step": 160
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 6.859999999999984e-05,
+      "loss": 1.7454,
+      "step": 165
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 0.00010290000000000003,
+      "loss": 1.8506,
+      "step": 170
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 0.00012800934269961248,
+      "loss": 1.7711,
+      "step": 175
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.6952,
+      "step": 180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.604886293411255,
+      "eval_runtime": 2.1277,
+      "eval_samples_per_second": 22.09,
+      "eval_steps_per_second": 2.82,
+      "step": 180
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 0.00012800934269961253,
+      "loss": 1.7717,
+      "step": 185
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 0.00010290000000000012,
+      "loss": 1.577,
+      "step": 190
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 6.859999999999993e-05,
+      "loss": 1.5197,
+      "step": 195
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 3.429999999999998e-05,
+      "loss": 1.6176,
+      "step": 200
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 9.19065730038752e-06,
+      "loss": 1.562,
+      "step": 205
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0,
+      "loss": 1.6157,
+      "step": 210
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.6068503856658936,
+      "eval_runtime": 2.1309,
+      "eval_samples_per_second": 22.056,
+      "eval_steps_per_second": 2.816,
+      "step": 210
+    },
+    {
+      "epoch": 7.17,
+      "learning_rate": 9.190657300387459e-06,
+      "loss": 1.5445,
+      "step": 215
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 3.429999999999988e-05,
+      "loss": 1.5187,
+      "step": 220
+    },
+    {
+      "epoch": 7.5,
+      "learning_rate": 6.859999999999982e-05,
+      "loss": 1.4775,
+      "step": 225
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 0.00010290000000000002,
+      "loss": 1.3245,
+      "step": 230
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 0.00012800934269961248,
+      "loss": 1.6077,
+      "step": 235
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.0001372,
+      "loss": 1.5085,
+      "step": 240
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.588986396789551,
+      "eval_runtime": 2.1323,
+      "eval_samples_per_second": 22.042,
+      "eval_steps_per_second": 2.814,
+      "step": 240
+    }
+  ],
+  "max_steps": 300,
+  "num_train_epochs": 10,
+  "total_flos": 243524173824000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fbef6ebe47b44e8f7ce6b04240d2f0cdf8816562a86c75bf5961e0d35045126
+size 3055