Datasets:
p-doom
/
crowd-code-0.1

Dataset card Data Studio Files
xet
Community
1
Sequence
int64
1
25.2k
Time
int64
1
858M
File
stringclasses
830 values
RangeOffset
int64
0
2.21M
RangeLength
int64
0
168k
Text
stringlengths
1
4.7M
⌀
Language
stringclasses
20 values
Type
stringclasses
9 values
242
466,998
utils/dataloader_new.py
3,414
0
null
python
selection_command
243
467,079
utils/dataloader_new.py
3,415
0
null
python
selection_command
244
904,870
utils/dataloader_new.py
3,487
0
null
python
selection_mouse
245
906,435
TERMINAL
0
0
bash
null
terminal_focus
246
919,192
models/tokenizer.py
0
0
from typing import Dict, Any, Tuple\n\nimport flax.linen as nn\n\nfrom utils.preprocess import patchify, unpatchify\nfrom utils.nn import STTransformer, VectorQuantizer\n\n\nclass TokenizerVQVAE(nn.Module):\n """ST-ViVit VQ-VAE"""\n\n in_dim: int\n model_dim: int\n latent_dim: int\n num_latents: int\n patch_size: int\n num_blocks: int\n num_heads: int\n dropout: float\n codebook_dropout: float\n\n def setup(self):\n self.encoder = STTransformer(\n self.model_dim,\n self.latent_dim,\n self.num_blocks,\n self.num_heads,\n self.dropout,\n )\n self.vq = VectorQuantizer(\n self.latent_dim,\n self.num_latents,\n self.codebook_dropout,\n )\n self.out_dim = self.in_dim * self.patch_size**2\n self.decoder = STTransformer(\n self.model_dim,\n self.out_dim,\n self.num_blocks,\n self.num_heads,\n self.dropout,\n )\n\n def __call__(self, batch: Dict[str, Any], training: bool = True) -> Dict[str, Any]:\n H, W = batch["videos"].shape[2:4]\n outputs = self.vq_encode(batch["videos"], training)\n recon = self.decoder(outputs["z_q"]) # (B, T, H_down * W_down, C)\n recon = nn.sigmoid(recon)\n outputs["recon"] = unpatchify(recon, self.patch_size, H, W)\n return outputs\n\n def vq_encode(self, videos: Any, training: bool = True) -> Dict[str, Any]:\n # --- Preprocess + encode ---\n B, T = videos.shape[:2]\n x = patchify(videos, self.patch_size)\n N = x.shape[2]\n x = self.encoder(x) # (B, T, N, E)\n\n # --- Vector quantize ---\n x = x.reshape(B * T * N, self.latent_dim)\n z_q, z, emb, indices = self.vq(x, training)\n z_q = z_q.reshape(B, T, N, self.latent_dim)\n indices = indices.reshape(B, T, N)\n return dict(z_q=z_q, z=z, emb=emb, indices=indices)\n\n def decode(self, indices: Any, video_hw: Tuple[int, int]):\n z = self.vq.codebook[indices]\n recon = self.decoder(z)\n recon = nn.sigmoid(recon)\n return unpatchify(recon, self.patch_size, *video_hw)\n
python
tab
247
926,801
train_tokenizer.py
0
0
from dataclasses import dataclass\nimport os\nimport time\n\nimport einops\nfrom flax.training import orbax_utils\nfrom flax.training.train_state import TrainState\nfrom jax.sharding import Mesh, PartitionSpec, NamedSharding\nfrom jax.experimental.mesh_utils import create_device_mesh\nimport optax\nimport orbax\nfrom orbax.checkpoint import PyTreeCheckpointer\nimport numpy as np\nimport dm_pix as pix\nimport jax\nimport jax.numpy as jnp\nimport tyro\nimport wandb\n\nfrom models.tokenizer import TokenizerVQVAE\nfrom utils.dataloader import get_dataloader\n\nts = int(time.time())\n\n\n@dataclass\nclass Args:\n # Experiment\n num_steps: int = 300_000\n seed: int = 0\n seq_len: int = 16\n image_channels: int = 3\n image_height: int = 90\n image_width: int = 160\n data_dir: str = "data_tfrecords/coinrun"\n checkpoint: str = ""\n # Optimization\n vq_beta: float = 0.25\n batch_size: int = 48\n min_lr: float = 3e-4\n max_lr: float = 3e-4\n warmup_steps: int = 10000\n # Tokenizer\n model_dim: int = 512\n latent_dim: int = 32\n num_latents: int = 1024\n patch_size: int = 4\n num_blocks: int = 8\n num_heads: int = 8\n dropout: float = 0.0\n codebook_dropout: float = 0.01\n # Logging\n log: bool = False\n entity: str = ""\n project: str = ""\n log_interval: int = 5\n log_image_interval: int = 250\n ckpt_dir: str = ""\n log_checkpoint_interval: int = 10000\n log_gradients: bool = False\n\n\nargs = tyro.cli(Args)\n\n\ndef tokenizer_loss_fn(params, state, inputs):\n # --- Compute loss ---\n outputs = state.apply_fn(\n params,\n inputs,\n training=True,\n rngs={"params": inputs["rng"], "dropout": inputs["dropout_rng"]},\n )\n mse = jnp.square(inputs["videos"] - outputs["recon"]).mean()\n q_loss = jnp.square(jax.lax.stop_gradient(outputs["emb"]) - outputs["z"]).mean()\n commitment_loss = jnp.square(\n outputs["emb"] - jax.lax.stop_gradient(outputs["z"])\n ).mean()\n loss = mse + q_loss + args.vq_beta * commitment_loss\n\n # --- Compute validation metrics ---\n gt = inputs["videos"].clip(0, 1).reshape(-1, *inputs["videos"].shape[2:])\n recon = outputs["recon"].clip(0, 1).reshape(-1, *outputs["recon"].shape[2:])\n psnr = pix.psnr(gt, recon).mean()\n ssim = pix.ssim(gt, recon).mean()\n _, index_counts = jnp.unique_counts(\n jnp.ravel(outputs["indices"]), size=args.num_latents, fill_value=0\n )\n codebook_usage = (index_counts != 0).mean()\n metrics = dict(\n loss=loss,\n mse=mse,\n q_loss=q_loss,\n commitment_loss=commitment_loss,\n psnr=psnr,\n ssim=ssim,\n codebook_usage=codebook_usage,\n )\n return loss, (outputs["recon"], metrics)\n\n\n@jax.jit\ndef train_step(state, inputs):\n grad_fn = jax.value_and_grad(tokenizer_loss_fn, has_aux=True, allow_int=True)\n (loss, (recon, metrics)), grads = grad_fn(state.params, state, inputs)\n state = state.apply_gradients(grads=grads)\n if args.log_gradients:\n metrics["encoder_gradients_std/"] = jax.tree.map(\n lambda x: x.std(), grads["params"]["encoder"]\n )\n metrics["vq_gradients_std/"] = jax.tree.map(\n lambda x: x.std(), grads["params"]["vq"]\n )\n metrics["decoder_gradients_std/"] = jax.tree.map(\n lambda x: x.std(), grads["params"]["decoder"]\n )\n return state, loss, recon, metrics\n\n\nif __name__ == "__main__":\n jax.distributed.initialize()\n num_devices = jax.device_count()\n if num_devices == 0:\n raise ValueError("No JAX devices found.")\n print(f"Running on {num_devices} devices.")\n\n if args.batch_size % num_devices != 0:\n raise ValueError(\n f"Global batch size {args.batch_size} must be divisible by "\n f"number of devices {num_devices}."\n )\n\n per_device_batch_size_for_init = args.batch_size // num_devices\n\n rng = jax.random.PRNGKey(args.seed)\n if args.log and jax.process_index() == 0:\n wandb.init(entity=args.entity, project=args.project, group="debug", config=args)\n\n # --- Initialize model ---\n tokenizer = TokenizerVQVAE(\n in_dim=args.image_channels,\n model_dim=args.model_dim,\n latent_dim=args.latent_dim,\n num_latents=args.num_latents,\n patch_size=args.patch_size,\n num_blocks=args.num_blocks,\n num_heads=args.num_heads,\n dropout=args.dropout,\n codebook_dropout=args.codebook_dropout,\n )\n rng, _rng = jax.random.split(rng)\n image_shape = (args.image_height, args.image_width, args.image_channels)\n inputs = dict(\n videos=jnp.zeros(\n (per_device_batch_size_for_init, args.seq_len, *image_shape),\n dtype=jnp.float32,\n ),\n )\n init_params = tokenizer.init(_rng, inputs)\n\n # --- Initialize optimizer ---\n lr_schedule = optax.warmup_cosine_decay_schedule(\n args.min_lr, args.max_lr, args.warmup_steps, args.num_steps\n )\n tx = optax.adamw(learning_rate=lr_schedule, b1=0.9, b2=0.9, weight_decay=1e-4)\n train_state = TrainState.create(apply_fn=tokenizer.apply, params=init_params, tx=tx)\n\n # FIXME: switch to create_hybrid_device_mesh for runs spanning multiple nodes\n device_mesh_arr = create_device_mesh((num_devices,))\n mesh = Mesh(devices=device_mesh_arr, axis_names=("data",))\n\n replicated_sharding = NamedSharding(mesh, PartitionSpec())\n train_state = jax.device_put(train_state, replicated_sharding)\n\n # --- Load checkpoint ---\n step = 0\n if args.checkpoint:\n restore_target = {"model": train_state}\n restore_args = orbax_utils.restore_args_from_target(restore_target)\n train_state.params["params"].update(\n PyTreeCheckpointer()\n .restore(args.checkpoint, item=restore_target, restore_args=restore_args)[\n "model"\n ]\n .params["params"]\n )\n # Assume checkpoint is of the form tokenizer_<timestamp>_<step>\n step += int(args.checkpoint.split("_")[-1])\n\n # --- TRAIN LOOP ---\n tfrecord_files = [\n os.path.join(args.data_dir, x)\n for x in os.listdir(args.data_dir)\n if x.endswith(".tfrecord")\n ]\n dataloader = get_dataloader(\n # NOTE: We deliberately pass the global batch size\n # The dataloader shards the dataset across all processes\n tfrecord_files,\n args.seq_len,\n args.batch_size,\n *image_shape,\n )\n print(f"Starting training from step {step}...")\n while step < args.num_steps:\n for videos in dataloader:\n # npy_path = "overfit_dir/single_sample_corner.npy"\n # videos = np.load(npy_path)\n # print("batch shape: ", videos.shape)\n # while(True):\n # --- Train step ---\n rng, _rng, _rng_dropout = jax.random.split(rng, 3)\n\n videos_sharding = NamedSharding(\n mesh, PartitionSpec("data", None, None, None, None)\n )\n videos = jax.make_array_from_process_local_data(videos_sharding, videos)\n\n inputs = dict(videos=videos, rng=_rng, dropout_rng=_rng_dropout)\n start_time = time.time()\n train_state, loss, recon, metrics = train_step(train_state, inputs)\n elapsed_time = (time.time() - start_time) * 1000\n print(f"Step {step}, loss: {loss}, step time: {elapsed_time}ms")\n step += 1\n\n # --- Logging ---\n if args.log:\n if step % args.log_interval == 0 and jax.process_index() == 0:\n wandb.log(\n {\n "loss": loss,\n "step": step,\n "step_time_ms": elapsed_time,\n **metrics,\n }\n )\n if step % args.log_image_interval == 0:\n gt_seq = inputs["videos"][0]\n recon_seq = recon[0].clip(0, 1)\n comparison_seq = jnp.concatenate((gt_seq, recon_seq), axis=1)\n comparison_seq = einops.rearrange(\n comparison_seq * 255, "t h w c -> h (t w) c"\n )\n # NOTE: Process-dependent control flow deliberately happens\n # after indexing operation since it must not contain code\n # sections that lead to cross-accelerator communication.\n if jax.process_index() == 0:\n log_images = dict(\n image=wandb.Image(np.asarray(gt_seq[0])),\n recon=wandb.Image(np.asarray(recon_seq[0])),\n true_vs_recon=wandb.Image(\n np.asarray(comparison_seq.astype(np.uint8))\n ),\n )\n wandb.log(log_images)\n if step % args.log_checkpoint_interval == 0:\n ckpt = {"model": train_state}\n orbax_checkpointer = orbax.checkpoint.PyTreeCheckpointer()\n save_args = orbax_utils.save_args_from_target(ckpt)\n orbax_checkpointer.save(\n os.path.join(os.getcwd(), args.ckpt_dir, f"tokenizer_{ts}_{step}"),\n ckpt,\n save_args=save_args,\n )\n if step >= args.num_steps:\n break\n
python
tab
248
928,576
train_tokenizer.py
9,423
0
null
python
selection_command
249
928,836
train_tokenizer.py
9,401
0
null
python
selection_command
250
929,327
train_tokenizer.py
9,362
0
null
python
selection_command
251
929,418
train_tokenizer.py
9,344
0
null
python
selection_command
252
929,419
train_tokenizer.py
9,303
0
null
python
selection_command
253
929,441
train_tokenizer.py
9,277
0
null
python
selection_command
254
929,477
train_tokenizer.py
9,189
0
null
python
selection_command
255
929,513
train_tokenizer.py
9,148
0
null
python
selection_command
256
929,577
train_tokenizer.py
9,080
0
null
python
selection_command
257
929,609
train_tokenizer.py
9,005
0
null
python
selection_command
258
929,610
train_tokenizer.py
8,959
0
null
python
selection_command
259
929,639
train_tokenizer.py
8,902
0
null
python
selection_command
260
929,671
train_tokenizer.py
8,856
0
null
python
selection_command
261
929,703
train_tokenizer.py
8,830
0
null
python
selection_command
262
929,734
train_tokenizer.py
8,799
0
null
python
selection_command
263
929,766
train_tokenizer.py
8,723
0
null
python
selection_command
264
929,798
train_tokenizer.py
8,668
0
null
python
selection_command
265
929,834
train_tokenizer.py
8,595
0
null
python
selection_command
266
929,866
train_tokenizer.py
8,525
0
null
python
selection_command
267
929,902
train_tokenizer.py
8,482
0
null
python
selection_command
268
929,934
train_tokenizer.py
8,433
0
null
python
selection_command
269
929,935
train_tokenizer.py
8,356
0
null
python
selection_command
270
929,969
train_tokenizer.py
8,278
0
null
python
selection_command
271
930,003
train_tokenizer.py
8,198
0
null
python
selection_command
272
930,035
train_tokenizer.py
8,176
0
null
python
selection_command
273
930,067
train_tokenizer.py
8,107
0
null
python
selection_command
274
930,099
train_tokenizer.py
8,052
0
null
python
selection_command
275
930,130
train_tokenizer.py
7,970
0
null
python
selection_command
276
930,162
train_tokenizer.py
7,918
0
null
python
selection_command
277
930,194
train_tokenizer.py
7,869
0
null
python
selection_command
278
930,226
train_tokenizer.py
7,813
0
null
python
selection_command
279
930,257
train_tokenizer.py
7,791
0
null
python
selection_command
280
930,289
train_tokenizer.py
7,765
0
null
python
selection_command
281
930,290
train_tokenizer.py
7,726
0
null
python
selection_command
282
930,321
train_tokenizer.py
7,668
0
null
python
selection_command
283
930,353
train_tokenizer.py
7,626
0
null
python
selection_command
284
930,385
train_tokenizer.py
7,584
0
null
python
selection_command
285
930,427
train_tokenizer.py
7,558
0
null
python
selection_command
286
930,458
train_tokenizer.py
7,527
0
null
python
selection_command
287
930,460
train_tokenizer.py
7,448
0
null
python
selection_command
288
930,523
train_tokenizer.py
7,423
0
null
python
selection_command
289
930,559
train_tokenizer.py
7,393
0
null
python
selection_command
290
930,594
train_tokenizer.py
7,392
0
null
python
selection_command
291
930,626
train_tokenizer.py
7,370
0
null
python
selection_command
292
930,627
train_tokenizer.py
7,293
0
null
python
selection_command
293
930,661
train_tokenizer.py
7,232
0
null
python
selection_command
294
930,700
train_tokenizer.py
7,152
0
null
python
selection_command
295
930,733
train_tokenizer.py
7,115
0
null
python
selection_command
296
930,766
train_tokenizer.py
7,038
0
null
python
selection_command
297
930,798
train_tokenizer.py
7,037
0
null
python
selection_command
298
930,830
train_tokenizer.py
6,952
0
null
python
selection_command
299
930,831
train_tokenizer.py
6,938
0
null
python
selection_command
300
930,861
train_tokenizer.py
6,870
0
null
python
selection_command
301
930,893
train_tokenizer.py
6,825
0
null
python
selection_command
302
930,925
train_tokenizer.py
6,824
0
null
python
selection_command
303
930,956
train_tokenizer.py
6,761
0
null
python
selection_command
304
930,989
train_tokenizer.py
6,728
0
null
python
selection_command
305
931,020
train_tokenizer.py
6,705
0
null
python
selection_command
306
931,052
train_tokenizer.py
6,658
0
null
python
selection_command
307
931,117
train_tokenizer.py
6,621
0
null
python
selection_command
308
931,118
train_tokenizer.py
6,561
0
null
python
selection_command
309
931,155
train_tokenizer.py
6,527
0
null
python
selection_command
310
931,191
train_tokenizer.py
6,494
0
null
python
selection_command
311
931,224
train_tokenizer.py
6,442
0
null
python
selection_command
312
931,258
train_tokenizer.py
6,436
0
null
python
selection_command
313
931,318
train_tokenizer.py
6,414
0
null
python
selection_command
314
931,318
train_tokenizer.py
6,389
0
null
python
selection_command
315
931,333
train_tokenizer.py
6,367
0
null
python
selection_command
316
931,654
train_tokenizer.py
6,389
0
null
python
selection_command
317
932,154
train_tokenizer.py
6,414
0
null
python
selection_command
318
932,185
train_tokenizer.py
6,436
0
null
python
selection_command
319
932,208
train_tokenizer.py
6,442
0
null
python
selection_command
320
932,247
train_tokenizer.py
6,494
0
null
python
selection_command
321
932,280
train_tokenizer.py
6,527
0
null
python
selection_command
322
932,333
train_tokenizer.py
6,561
0
null
python
selection_command
323
932,378
train_tokenizer.py
6,621
0
null
python
selection_command
324
932,688
train_tokenizer.py
6,561
0
null
python
selection_command
325
932,845
train_tokenizer.py
6,527
0
null
python
selection_command
326
933,177
train_tokenizer.py
6,535
0
null
python
selection_command
327
933,832
train_tokenizer.py
6,535
0
#
python
content
328
933,833
train_tokenizer.py
6,536
0
null
python
selection_keyboard
329
933,866
train_tokenizer.py
6,536
0
python
content
330
933,868
train_tokenizer.py
6,537
0
null
python
selection_keyboard
331
934,155
train_tokenizer.py
6,536
0
null
python
selection_command
332
934,404
train_tokenizer.py
6,572
0
null
python
selection_command
333
934,937
train_tokenizer.py
6,563
59
# npy_path = "overfit_dir/single_sample_corner.npy"
python
selection_command
334
935,161
train_tokenizer.py
6,563
96
# npy_path = "overfit_dir/single_sample_corner.npy"\n # videos = np.load(npy_path)
python
selection_command
335
935,263
train_tokenizer.py
6,563
143
# npy_path = "overfit_dir/single_sample_corner.npy"\n # videos = np.load(npy_path)\n # print("batch shape: ", videos.shape)
python
selection_command
336
935,412
train_tokenizer.py
6,563
166
# npy_path = "overfit_dir/single_sample_corner.npy"\n # videos = np.load(npy_path)\n # print("batch shape: ", videos.shape)\n # while(True):
python
selection_command
337
935,857
train_tokenizer.py
6,571
0
null
python
selection_command
338
936,552
train_tokenizer.py
6,715
1
null
python
content
339
936,553
train_tokenizer.py
6,668
1
null
python
content
340
936,553
train_tokenizer.py
6,631
1
null
python
content
341
936,553
train_tokenizer.py
6,571
1
null
python
content