Mettre en file une opération forward-backward

POST

https://api.fastino.ai

training

jobs

{job_id}

forward-backward

curl -X POST "https://api.fastino.ai/v1/training/jobs/$TRAINING_JOB_ID/forward-backward" \
  -H "X-API-Key: $FASTINO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "loss_fn": "cross_entropy",
    "datums": [
      {
        "model_input": {"chunks": [{"type": "encoded_text", "tokens": [151644, 872, 198, 3838]}]},
        "loss_fn_inputs": {
          "target_tokens": {"data": [872, 198, 3838, 30], "dtype": "int64"},
          "weights": {"data": [0, 0, 1, 1], "dtype": "float32"}
        }
      }
    ]
  }'

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward"
}

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward",
  "status": "completed",
  "result": {
    "loss_fn_output_type": "ArrayRecord",
    "loss_fn_outputs": [
      {"logprobs": {"data": [-3.1, -0.4, -1.2, -0.8], "dtype": "float32", "shape": [4]}}
    ],
    "metrics": {"loss:sum": 2.0}
  },
  "error": null
}

POST /v1/training/jobs/{job_id}/forward-backward Computes the loss of a batch of token-level datums and accumulates its gradients. The next optimizer step applies them. The response is a future; read the loss and per-token log probabilities from GET /v1/training/jobs/{job_id}/futures/{future_id}. POST /v1/training/jobs/{job_id}/forward takes the same body and computes the loss without gradients.

Request

string

requis

Your Fastino API key.

string

requis

Interactive job ID.

object[]

requis

One or more training examples. Tokenize them yourself with the base model’s tokenizer.

Afficher properties

object

requis

{"chunks": [{"type": "encoded_text", "tokens": [...]}]}. The input token IDs, N tokens in total.

object

requis

Per-position tensors, each {"data": [...], "dtype": "int64" | "float32"} with an optional shape. Every tensor has N values.

target_tokens (int64, required): the token expected at each position, usually the input shifted left by one.
weights (float32): per-position loss weight. Use 0 for prompt positions and 1 for completion positions.
logprobs and advantages (float32): required by every loss except cross_entropy. logprobs are the sampler’s log probabilities of the targets.

string

défaut:"cross_entropy"

cross_entropy, importance_sampling, ppo, cispo, or dro.

object

Loss settings. ppo and cispo take clip_low_threshold and clip_high_threshold; dro takes beta. Unknown keys are rejected.

curl -X POST "https://api.fastino.ai/v1/training/jobs/$TRAINING_JOB_ID/forward-backward" \
  -H "X-API-Key: $FASTINO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "loss_fn": "cross_entropy",
    "datums": [
      {
        "model_input": {"chunks": [{"type": "encoded_text", "tokens": [151644, 872, 198, 3838]}]},
        "loss_fn_inputs": {
          "target_tokens": {"data": [872, 198, 3838, 30], "dtype": "int64"},
          "weights": {"data": [0, 0, 1, 1], "dtype": "float32"}
        }
      }
    ]
  }'

Response

string

The queued future.

string

The job.

string

forward_backward.

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward"
}

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward",
  "status": "completed",
  "result": {
    "loss_fn_output_type": "ArrayRecord",
    "loss_fn_outputs": [
      {"logprobs": {"data": [-3.1, -0.4, -1.2, -0.8], "dtype": "float32", "shape": [4]}}
    ],
    "metrics": {"loss:sum": 2.0}
  },
  "error": null
}

Errors

Status	Meaning
`400`	Unknown `loss_fn`, a non-integer `target_tokens` tensor, out-of-vocabulary token IDs, or the job was stopped.
`402`	The team is out of credits or has no valid payment setup.
`404`	The job doesn’t exist or belongs to another team.
`422`	`datums` is empty or the body has an unknown field.

Problems found while the batch runs, such as a missing loss input, mismatched tensor lengths, an unknown loss_fn_config key, or a datum longer than the model’s limit, fail the future. Its error.category is user.

⌘I

curl -X POST "https://api.fastino.ai/v1/training/jobs/$TRAINING_JOB_ID/forward-backward" \
  -H "X-API-Key: $FASTINO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "loss_fn": "cross_entropy",
    "datums": [
      {
        "model_input": {"chunks": [{"type": "encoded_text", "tokens": [151644, 872, 198, 3838]}]},
        "loss_fn_inputs": {
          "target_tokens": {"data": [872, 198, 3838, 30], "dtype": "int64"},
          "weights": {"data": [0, 0, 1, 1], "dtype": "float32"}
        }
      }
    ]
  }'

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward"
}

{
  "future_id": "FUTURE_ID",
  "job_id": "SESSION_ID:train:0",
  "type": "forward_backward",
  "status": "completed",
  "result": {
    "loss_fn_output_type": "ArrayRecord",
    "loss_fn_outputs": [
      {"logprobs": {"data": [-3.1, -0.4, -1.2, -0.8], "dtype": "float32", "shape": [4]}}
    ],
    "metrics": {"loss:sum": 2.0}
  },
  "error": null
}

Inférence

Entraînement

Mettre en file une opération forward-backward

Request

Response

Errors

​Request

​Response

​Errors

Request

Response

Errors